Home

An ontology-based approach to achieve inclusive design - E-LIB

1. Task Arcelik Washing Arcelik Washing Machine Panel 1 Machine Panel 2 Identify the on off button 8 8 8 8 Push on off button 6 8 6 8 Identify the set Program 8 8 8 8 Turn knob to set Program 2 8 2 8 Identify main control panel 8 8 8 8 Read and understand texts of 2 8 3 8 main panel Identify minor control panel 8 8 8 8 Read and understand texts of 2 8 4 8 minor panel Total 68 75 73 43 Table 5 26 Results of user study regarding washing machines Difficulty encountered Arcelik Existing Panel 1 Arcelik Mock up Panel 2 Cognitive features Meaning of labels is not intuitive Yes Yes Technical terminology used Yes Yes Sensory features Difficulty reading the labels Yes Yes Difficult to find programs Yes Yes Physical features On off button is hard to press Yes Yes Program selection knob difficult to Yes Yes turn Program selection knob is too No Yes small to hold and control Total number of issues 6 7 Table 5 27 Result comparison regarding washing machines by user study 135 5 EVALUATION Difficulty encountered Arcelik Existing Panel 1 Arcelik Mock up Panel 2 Cognitive features Order of use not intuitive Yes Yes Meaning of labels not intuitive Yes Yes Technical terminology used Yes Yes Meaning of formatting is unclear Yes
2. Grundig Ar elik Grundig large black IEC 5010 power symbol Volume Up Down on lower left marked with and on the left and right outer circle in the middle Channel Up Down on lower right as P and P on the top and bottom outer circle in the middle Width 45mm length 240mm height 17mm Arcelik Grundig small black IEC 5010 power symbol Volume Up Down on lower left marked with and on the left and right outer circle in the middle Channel Up Down on lower right as and on the top and bottom outer circle in the middle Width 50mm length 110mm height 17mm at lowest 27mm at highest point height increases gradually from top to bottom o Table 5 10 Existing and emerged remote controls used for evaluation of customer satisfaction 5 EVALUATION Source Model Physical Characteristics Image e On Off button on the right side of the panel with a IEC 5010 power symbol e Program selection as rotary knob in the middle with 12 settings e Buttons and rotary knob are Ame grey coloured Ar elik Washing Different labelling e g Machine under hand symbol Start Panel 1 Pause Cancel button e LED information panel between program knob and detergent dispenser e Width 590mm height 125mm at shortest point and 150mm at longest point e On Off butto
3. General characteristics Predicate Datatype Description Name String The name to identify an environment is the only primary predicate Mandatory to define it in an instance IDName String The ID name is unique for each object of the Ontology class Description String Textual description of the environment RoomType Integer 1 Living Room in which user trial took place room 2 Dining room 3 Kitchen 4 Living dining room 5 Kitchen dining room 6 Utility storage room 7 Kitchen din ing living room 8 Bathroom 9 Cellar 10 Other RoomWidth Integer 1 99 Estimate of room width in meters in which user trial took place RoomLength Integer 1 99 Estimate of room length in meters in which user trial took place Continued on next page 58 3 KNOWLEDGE MANAGEMENT Door Integer 1 999 Number of doors in room where field trial took place Window Integer 1 999 Number of windows in room where field trial took place Hearing Acoustics Integer 1 Good Acoustics in the room in which user 2 Bad trial took place BackgroundNoise Integer 0 No Level of background noise in room in Level background noise which user trial took place 1 Low 2 Loud BackgroundNoise Integer 1 Type of background noise in room in Type TV radio 2 which user trial took place People talking 3 Dog barking 4 Road works 5 Alarm 6 Traffic 7 Co
4. MDProfile2 User Profile Trevor V11 V12 MD1 MD2 Eileen Mark Source Doroth NEB y http www cardiac eu ora guidelines telecoms mobile htm Gandalf VProfile1 User Summary Keys should be as large as possible without reducing the distance between the keys to less than half the key width Environment y Text The spacing between the keys is as important as the size ofthe keys as spacing enables keys to be more easily distinguished both visually and by touch Keys should be as large as possible without reducing the distance between the keys to less than half the kew width OpenasRTF Open as PDF Found 57 Recommendations Figure 4 8 Designer role view with selection of VProfile2 User Model Product development workflow Using the Administration software component the designer is able to get recommen dations based upon selections and to change the complete Ontology This software is directly used by the designers in the evaluation In the Sketch Design phase the administrator role needs to provide a Socket Server on local or remote system Afterwards the designer role can start the sketch applica tion to connect to the server system Figure 4 8 presents the selection of moderate visual impaired users The pre sented recommendations only concentrate upon all factors and user needs related to VProfile2 User Model as defined in Vicon Consortium 2012a 85 4 APPLICATION IN DEVELOPMEN
5. Large buttons Good sized detergent drawer with enough space for any sized hand to fit in e It is difficult to know what to do first Do you press On Or do you select a program Why is an On button necessary at all What is the difference between Start Pause Cancel and On Off No colour contrast between buttons and their surround High gloss finish on buttons and their surround Avoid bold and italics in labels It is not clear why some program labels are in bold and italics while others are in regular grey text No audible and tactile click from any of the major or minor controls on activation Parallax issues the user has to kneel in front of the display in order to read the full program guide the program selection knob blocks the view of the bottom programs For a quick wash does the user select Express 39 from the program guide or Quick Wash from the minor controls What is the difference The names of some of the programs are not intuitive Rinse on the program selection knob versus Rinse Plus on the minor controls There is no home setting for the program selection knob This means that the starting point for the knob might be different every time the user puts on a wash For users who count the turns in order to find the desired program this is a particular issue Table 5 19 General Comments Ar elik Washing Machine Panel 2 Part A
6. Statement skip The Sketch Application provides O 2 5 2 2 0 a wide choice of scenarios The design recommendations of O 0 0 5 6 0 the toolset are necessary It takes a short time to go 2 1 3 3 2 0 through recommendation list I would need a user manual to 0 1 3 6 1 0 use the software The look and feel of the 0 2 3 4 2 0 application was suitable and pleasant The software is easy to use 0 2 0 8 0 1 Table 5 3 Results of questions regarding suitability for the task The suitability of the software framework to be added into existing product devel opment processes is covered in this section Table 5 3 presents these results The first statement is questioning if the choice of scenarios of user model typ ical environment in which the product would be used and typical task performed using the product is sufficient enough The result is slightly positive but ambigu ous Almost half of participants answered neutrally 5 11 answered infers that the current system describes a good base but can and probably should also be extended The necessity of design recommendations is throughout answered very pos itive All participants agreed resulting in the conclusion that the concept of a recommendation driven product development process is not just accepted but also needed Although the customer involvement method applied by each participant is very different
7. 4 APPLICATION IN DEVELOPMENT PROCESS is also possible representing the use of the product by different impairment groups in different typical environments performing different typical tasks and resulting in a merged set of recommendations for all scenarios see also 3 4 6 In the CAD phase software applications CAD CAE CAx are used for the cre ation of a virtual prototype With respect to the design process and the requirement of an as un obstructive system as possible an integrated module in Siemens NX was implemented The user input in this phase is the virtual product itself which is de signed simultaneously while using the module Virtual environments focus upon the representation of physical attributes and surfaces most often ignoring functional issues It is necessary to set up each compo nent and add context and type related attributes by the user Regarding this issue an annotation tool as a part of the CAD module was implemented by which the designer is able to annotate e g a cube as a press button Using this functional annotation for each component a set of recommendations is presented qualitative or quantita tive which should be considered Quantitative recommendations as presented in 1 4 contain limits of parameters e g minimum button label size of interface com ponents To support the designer these parameters can also be applied directly to the virtual component if the parameter mentioned is defined
8. ASEL ul sjaqe BY Jo Kan g Bay ayy S2npas Jou PINOYS VOLUN 1euJ 3ui aya q p 3eultunT i Neusa aq PINOYS sK aya MePI Men A 3u uuuojJinu3 UJ3 SAS3 SSUI PInD7DJO TiS 5eIpJe5 MMM TA U3q TIqorn7suJo5 1e31 s ul pinb7DJo Ti 5PIpJe5wwWW7 0331 T 5JnosS JASN z lo1dA JASN LIOA Jepuen K yioJog 10 Bul3u6n peq Joy sK 3 p 3eultunTi yeni BUEN ssia JOABIL JES ZAYON 135 LIOIN J SN z llJo13dH J SN LBIYOJAH 3u uu 3 U01133 35 ON MAJA uol3aepu ujujyop q IO xue3Joduul Y U Jnss Jd uol3eAI32e Key 90 g MM y PPoWwJasn oeusss s uoyang p Jls p JI U3 3J3LM ultuJ 3 p 07 seu J sriiy Z Z A uondu s q a gt ue3Jodw Y JY uoud ayy JO poq ay anoqe pasieJ aq PINOYS s 3 K3nI32e3 J 33 q JOY s0 Y E 82ue1J0dLu ybiy skay ayy uo SOU Je LW Jens a q 5a 12819 po Y Bm u033nq uado 4294 BWEN a gt ue3Jodw Wnipaw suol3puo gt Bulay n peg Joy shay p 3eulunTii EO 8 m uoewsojupseL 82ue7J0duu Y N uolgel3u J Jjip 49998q Joy Skay 5Je7 20 8 Es 10 10 9530 ON uondiuys q AsjojenJasuo WEN uoewsojuj JUSWUOJ AUZ 2ueVOdW WhIpaw sdo3 fay Jo 10 9333 3111283 BY BAOJdLUN Lo 8 m 5ue3JoduJ i Mo punoJJnos Jay pue suo33nq syeqe 1 pue suo33nq U A3 q 4seJQUOD JENSI poo 00 8 epuswworsy andusean b JUI ET N MBIASaJ seg 6p WOU3 suolqepu uuuuo2 u 199334007 Yodxg yodw
9. 3 4 2 Reasoning Engine Figure 3 5 presents the complete reasoning process for the final Ontology Based upon the initial Ontology as shown in the previous section the process contains five inferences up to the final model The first inference classifies User Model in stances using different rules according to WHO ICF user profiles see Organization et al 2012a and Organization et al 2012b With respect to the Ontology model this step adds new memberships for each User Model to different already created profile classes These classes are separated into no impairments e g HProfileO for no hearing impairment group mild e g VProfilel for mild visual impairment group and moderate e g MDProfile2 for moderate manual dexterity impairment group lev els for visual manual dexterity and hearing impairments The second inference deals with component recommendations resulting in analogue new classes with member instances for each recommendation related to an annotated component These rec IComplete syntax description can be found at http jena apache org documentation inference 65 3 KNOWLEDGE MANAGEMENT ommendations will be presented in the CAD phase The last three steps deal with the immediate textual recommendations presented in the sketch phase regarding the selection of the designer of a User Model typical environment and typical task Inference Action Initial Ontology lt lt lt User Model Pro
10. Part B 130 5 EVALUATION The side by side comparison regarding washing machines was conducted by a user study but also by an accessibility expert of NCBI In the first comparison see table 5 27 no accessibility issues were solved from existing to emerged user interface but one new issue appeared with respect to a smaller selection knob The second comparison by the expert figured out that one accessibility issue regarding the distance between the and button was solved but with the modification 10 new issues appeared see tables 5 28 and 5 29 On the contrary in the user trials the washing machine panel 2 performance 73 43 was slightly better than panel 1 68 75 see table 5 26 The reason of this lies in the amount of problems with labelling While issues regarding the labelling only count as one single accessibility issue they do have a much higher impact in practical use With respect to the software this recommends a high importance for accessibility to have good and easy readable labels In the recommendation list provided label accessibility issues can be found in several recommendations R 02 R 04 R 14 etc This part of the evaluation also covers a comparison of output of the framework with the real prototype design Recommendation lists provided by the framework were compared with the Doro Mock Up phone see table 5 13 if the recommenda tions were applied correctly or not One issue not included in t
11. To the VERITAS Simulation input is a Virtual User Model a Simulation Model one or more Task Models and a virtual 3D environment as part of the Simulation Plat form The Simulation Module creates a complete scenario in which the User Model performs tasks The Simulation Platform has three elements 1 Task Manager Module All task related issues are included in the Task Manager Module It divides the selected task into primitive tasks and manages the humanoid to perform each task separately 38 2 STATE OF THE ART 2 Humanoid Module The skeletal model of any selected user consists of 46 elements and 45 joints including different geometrical but also kinematic data as degrees of freedom 3 Scene Module This module creates the complete scene including objects and their attributes Open bottom drawer Simulation result Failure Shoulder amp Failure Wrist joint Elbow amp Wrist joint limit Simulation result limit a4 b4 Simulation result Failure Shoulder amp Elbow joint limit c4 Rheumatoid 5 E Adhesive shoulder Task arthritis Spinal cord injury capsulitis Hemiparesis Elderly Open top drawer i Simulation result E Simulation result N 5 n lat lt lat It Failure Shoulder ee n Sa Failure Shoulder joint a ye En od joint limit limit v bl yi gt al op cl zl
12. Tri VEB Cameras Gy Model History Akg Datum Coordinate S M8 Body 1 8 Body 2 M8 Body 3 M6 Body 4 M6 Body 5 WE Body 6 M8 Body 7 ZG Body 8 E Body 9 M8 Body 10 BE Body 11 B Body 12 BB Body 13 M8 Body 14 M8 Body 15 M8 Body 16 M8 Body 17 1 BE Body 19 M8 Body 20 E Body 21 M8 Body 22 M8 Body 23 BB Body 24 BE Body 25 E Body 26 M8 Body 27 M8 Body 28 M8 Body 29 M6 Body 30 IStart 8 Index of MViconinstall t WF NX 7 5 Gateway IP Figure 4 11 Example of a loaded CAD file DORO mobile phone Eile Edit View Format Tools Assemi sun M NA Py Command Finder IL Part Navigator E mrm No cue Name a Comment SP Gs HSH F E 3 Model History Bili Datum Coordinate S 1G Body 1 E Body 2 MB Body G BE Body 4 BB Body 5 M8 Body 6 Y Body 7 M8 Body 8 M8 Body 9 FAG Body 10 DB Body 11 BY Body 12 M8 Body 13 M8 Body 14 BE Body 15 BE Body 16 MB Body 17 M8 Body 18 M8 Body 19 M6 Body 20 FAG Body 21 AG Body 22 FG Body 23 ZG Body 24 M8 Body 25 M8 Body 26 BB Body 27 B Body 28 M8 Body 29 M8 Body 30 al E Recommendations Settings YODODODOO Dependencies Details Preview iblies Information Analysis Preferences Window Hel
13. i3 i13 For 8 19 beneficiaries the on off button was not obvious il i2 i5 i6 i7 i9 i10 i15 16 19 did not recognise A B and C buttons i2 i3 i4 i5 i6 i7 i8 i9 i10 113 114 i15 il6 i17 i18 i19 8 19 did not recognise the volume buttons i3 i4 i6 i7 114 i15 i116 i18 2 19 did not recognise scroll up down but tons 115 i18 Grundig Large Silver Remote Control Issue Participants 39 47 beneficiaries successfully identified the on off button il i2 17 i8 i9 i11 112 113 114 115 116 117 i18 gl g2 g4 g5 g6 g7 g8 g9 g10 g11 812 813 814 g15 g16 g17 g19 g20 tl t2 t3 t4 t5 t6 t7 t8 Continued on next page 174 APPENDIX Issue Participants A further three beneficiaries successfully identified the on off button on their second guess i5 il0 g18 Five beneficiaries had to be told where the on off button was i3 i4 i6 i119 i20 43 47 beneficiaries expected the on off but ton to be at the top of the remote control either on right or left il i2 i4 id i7 i8 ilO ill il2 113 il4 il5 i16 i17 i18 i20 gl g2 g4 g5 g6 87 g8 g9 g10 g11 g12 g13 g14 g15 g16 g17 g18 g19 g20 tl t2 t3 t4 t5 t6 t7 t8 Two beneficiaries failed to identify that on and off would be on the same button i4 i6 All 47 beneficiaries were able to physically
14. schnittstellen am beispiel von myui Technik f r ein selbstbestimmtes Leben 2012 G Stucki International classification of functioning disability and health icf a promising framework and classification for rehabilitation medicine American jour nal of physical medicine amp rehabilitation 84 10 733 2005 A Sundin M Christmansson and M Larsson A different perspective in participa tory ergonomics in product development improves assembly work in the automotive industry International journal of industrial ergonomics 33 1 1 14 2004 The Apache Software Foundation Reasoners and rule engines Jena inference sup port http jena apache org documentation inference 2013 VDLI Gesellschaft Entwicklung Konstruktion Vertrieb VDI 2221 Methodik zum En twickeln und Konstruieren Technischer Systeme und Produkte 1993 VDI Gesellschaft Konstruktion und Entwicklung Produktionstechnik ADB Gemeinschaftsausschu8 Produktplanung VDI 2220 Produktplanung Ablauf Begriffe und Organisation 1980 VDI Gesellschaft Konstruktion und Entwicklung Produktionstechnik ADB Gemeinschaftsausschuf8 Produktplanung VDI 2223 Methodisches Entwerfen technischer Produkte Verein Deutscher Ingenieure VDI Handbuch Konstruktion Berlin 2004 Vicon Consortium Project Deliverable 1 1 End user and environment field study 2010 Vicon Consortium Project Deliverable 1 2 Survey of Design Frameworks and Tools 2010a 152 REFERE
15. using the number keys on a phone Strongly Discomfort Integer 0 No Discomfort in hands when gripping 1 Slightly 2 small objects or operating controls Moderately 3 Strongly TouchSensitivity Integer 0 Sensitivity by touching different Normal 1 surfaces Mild 2 Moderate Table 3 8 Ontology class data properties used for User Model Manual dexterity Gandalf see figure 3 4 represents an active elderly gentleman who is used as a representative for a specific target user group Based upon textual issues presented in his description bold marked different abstract nominal and categorical values are extracted to form an analogue Ontology instance including different data proper ties General characteristics Predicate Value Name Gandalf IDName P5 Description Gandalf is an active older gentleman who refuses to let his age stop him from doing things Nickname N A VirtualModel N A Age 80 Gender M Continued on next page 56 3 KNOWLEDGE MANAGEMENT Hearing Predicate Value Hearing500Hz 30 Hearing 1kHz 45 Hearing2kHz 65 Hearing4kHz 75 SpeechWithBackgroundNoise O HearingAid 1 HearingAidWithProduct 1 Vision Predicate Value VisualAcuity 2 FieldOfVision 3 Colour 1 NearFocus 2 DepthPerception 2 ContrastSensitivity 2 Glare 1 Glasses 1 Glass
16. 1 Virtual User Model Similar as in this thesis Virtual User Models were used to describe user needs and requirements However the model includes general preferences disabili ties affected tasks motor visual hearing speech and cognitive and behavioural parameters 2 Task Model The interaction between the virtual user and the environment is described in the 37 2 STATE OF THE ART Physical Normal Rheuma Spinal Adhesive Hemi Elderly character values toid cord shoulder paresis Man istics arthritis injury capsulitis 75 79 Wrist 0 60 0 62 flexion Wrist 0 60 0 67 48 0 53 extension Shoulder 0 180 0 10 0 118 0 20 0 53 39 flexion Shoulder 0 90 0 15 0 74 0 10 abduction Shoulder 0 90 internal rotation Shoulder 0 50 0 15 0 31 0 10 external rotation Forearm 0 85 supination Elbow 0 150 0 91 09 flexion Table 2 3 Part of Virtual User Models as used in the VERITAS project Source Kaklanis et al 2010 Task Model Complex tasks are divided into primitive tasks and must be pre defined by designers developers according to the functionality of the designed prototype 3 Simulation Model The aim of the Simulation Model is to define all specific functionalities of the simulation result including information about possible tasks that can be per formed during simulation by the virtual user Simulation Platform
17. 129 5 EVALUATION Positive Negative e The Start Pause Cancel label is too close to the program labels The program selection knob is too difficult to turn The program selection knob does not give any tactile feedback when turned The program selection knob can sit between two programs The on off button is difficult to press The start button is a critical button which the user will use every single time the washing machine is in use but it is hidden amongst the other controls The location of the buttons relative to the order in which you use them is not intuitive The user presses On on the right then selects a program to the left then selects one or more of the minor controls to the left then presses Start to the right The layout should more closely mirror the user journey Difficult to press and hold the Start Pause Cancel button for three seconds e The Cancel label is mid way between two different controls Spacing should be used so that there is no confusion between buttons and their corresponding labels e The button must be pressed repeatedly to increase the time delay in increments of 5 minutes The labels are already wearing off Inconsistent font size on the program selection knob labels The program selection knob is very cluttered with 16 program options Table 5 20 General Comments Ar elik Washing Machine Panel 2
18. 4 1 3 Server Tool All connections between the back end and all front end applications are imple mented by the Socket Server Application middleware which provides reasoning us ing JENA McBride 2002 and commands to get all instances and relations between instances from the Ontology ConVic includes a JENA based interface to the Ontol ogy Ontology Editor and Administration Utilities and also the Sketch Application which can be started separately for preview purposes Vicon Consortium 2011b described these interfaces section 3 as back end and front end Each of the front end applications Sketch Design Application Administrator Software CAD Modules and Interfaces and the Virtual Reality Simulation Platform uses data which is pro vided by the Socket Server For each product mobile phone washing machine and TV remote the server creates a different port 65000 for mobile phone 65001 for washing machine and 65002 for TV remote on the server and provides all relevant information through commands The Socket Server is included in the setup file as an applicable Java JAR This JAR file contains all relevant information of the back end and the middleware itself To start it on the current machine it is just necessary to run the JAR file This software 77 OMO NO OI GOO N 4 APPLICATION IN DEVELOPMENT PROCESS component deals with direct communications between the client and the server side and requires ports 65000 to 6500
19. IPI d aen na noduu UDI IPR SANO asmu oj 21Au0D 81 4 APPLICATION IN DEVELOPMENT PROCESS uoneorydde U31S9d YPAS u JO JINN Weiserp seo sn 9 p 93H 4dd se aJOUI9Y AL suonepu uluio35 t uiypenw Bulysem qyuodx uoud l qoMW 0 221494 YINMS suoepuouuooy N ASEL M IA av9 104 1 lt sapn 2ui gt ASA Hodx3 o i spuajxa gt lt spu l x gt Fs i lt spu 1x gt suol epuswwossy JUSUWUOJIAUJ 1Xx91 U09 3U3WUOJ AUJ MAJA 129j95 129J9S suolnepu9ulul009Y saopnpul gt lUJoid 4 21901d 195N Suopepuawwo y IPPON 195N MAJA 720195 lle A IA ZsopnpulS sysel SIU9WUOJTAUJ S911J0Jd 19P0W 1 SN 39195494295 404 uo3I A HE1S suoT1epu uuo35 u ubrs q YII9NS NOJIA i uB6is q 13sn Y 40 e1 SIUIWUpY Y 82 4 APPLICATION IN DEVELOPMENT PROCESS Product development workflow The work flow of figure 4 6 demonstrates the support of the system in relation to user input for the role models administrative user and designer The Socket Server needs to be available as middleware see Vicon Consortium 201 1b for the designer to be able to use or to have access to the Ontology data It provides an interface for the selections of User Model profiles typical environment settings and typical tasks After the designer started the Sketch Design Application she or he can set up the target device type for the product develop
20. People who rely on touch to operate keypads benefit from keys that are as distinctive as possible to the touch ee oe are more easily distinguished than those that are flush against their surround Keys should therefore be raised above the body of the phone preferably by 5 Apply Recommendation As ls Ts 6 Mie HS UE Dependencies Details Preview EY Start index of Viconinstall NX 7 5 Gateway IP MAS J 1217PM NB Figure 4 13 Recommendation view in CAD Module defined by designers Tools gt Expression or Ctrl E in Siemens NX The interface presents all relevant parameters during annotation of objects For example if one recommendation defines that the button height should be at least 5 mm the affected parameter name is presented during annotation and must be defined by designers in Siemens NX 4 3 Impact on the Product Development Process Chapter 2 4 described various customer involvement methods used in different sce narios of product development Referring to different levels of customer involvement in terms of this thesis customers are beneficiaries the reviewed methods presented three levels design for design with and design bu Design for describes the perspec tive of product development to create a product for a specified target group without participation of real persons representing the target group Here designers create the product and decide about changes all by
21. Wu skip Figure 12 Question if the description of information in the sketch appli cation for environments is comprehensible 10 participants skip Figure 13 Question if the description of information in the sketch appli cation for tasks is comprehensible 161 APPENDIX participants 10 _ EZ I s skip Figure 14 Question if the description about recommendations is com prehensible Conformity with user expectations participants 10 I skip Figure 15 Question if from participant point of view the software has a consistent structure 162 APPENDIX 10 p 8 2 S g A 8 g A di N E Te skip Figure 16 Question about layout expectations of the software 10 8 2 E a oF E 8 A 4 2 0 I I l ii skip Figure 17 Question if some features of applications do not have an un predictable processing time 163 APPENDIX Suitability for learning participants I k j E FR skip Figure 18 Question how much time is required for learning to use the software 10 p participants PE skip Figure 19 Question if the software is easy to learn without prior knowl edge help or manual 164 APPENDIX participants T b skip Figure 20 Question if the software is easy use to use even without hav ing prior knowledge Complete Re
22. further links etc This data must be used for a dynamic extraction to present only relevant data which the designer needs based upon all input given as a selection of a specific impairment group of beneficiaries The next issue is how to present the data to the user as seamless as possible in the product development process Since designers use various software or not software tools e g draft sketches in a phase based sequence all support must be included as much as possible in the typical design process Thus the following research questions are topic of this thesis e Representation of information How to extract data from issues One main requirement is to include both quantitative and qualitative recom mendations into a database or context aware system see section 1 4 Adaptation of information How to maintain adaptation and sustainability All recommendations must be manipulable including an addition of designer s personal experience Exploitation of information How to use available data Each recommendation item must be represented in a designer friendly way the inclusion in different tools of the design process is preferred Impact without hindrance in the product development process How to maintain designer acceptance All issues must be included in the product development process smoothly to maintain the acceptance by designers This is a main point which is also men tioned by various authors as problematic see C
23. ment aims the purpose to raise the question of user needs by designers to their product 5 1 4 Results Statement The VICON virtual user concept 1 0 0 1 3 0 is capable in supporting the designers in creating inclusive products The VICON virtual user concept O 1 1 1 2 0 can help to involve the user s perspective into the development process earlier The VICON virtual user concept 1 1 0 0 2 1 is capable in product development acceleration The VICON virtual user concept O 0 1 0 2 2 provides knowledge concerning disabilities and derived requirements Table 5 1 Results of statements about general concept Table 5 1 presents the results of this user study The first statement shows the response of participants if the concept is capable of supporting designers in creating inclusive products All except one participant agreed The one participant strongly disagreed concerning a complete replacement of user trials with prototypes with the framework The concept itself may be not capable to replace user tests completely but is able to help and support especially for designers with minor experience in in clusive design guidelines 97 5 EVALUATION Comments to statement The VICON virtual user concept is capable in sup porting the designers in creating inclusive products It will give a very good reminder to work on inclusive design After al
24. ommendation instances On task side each instance can optionally contain a direct connection to the profile predicate of the recommendation class by the Impairment predicate in which recommendations are presented if the same im pairment profile groups are included see task parameters table 3 11 Environment The EnvRule predicate represents the connection between environment and rec ommendation instances Component For the component relation available annotation component options are speci fied in the Component predicate 3 4 5 Description Logic Expressivity Description Logic DL expressivity denotes the complexity of operators used through out the Ontology Baader 2003 Table 3 16 presents the expressivity used by the Ontology Naming convention Description AL Attributive language This is the base language which allows e Atomic negation negation of concept names that do not appear on the left hand side of axioms e Concept intersection e Universal restrictions e Limited existential quantification Cc Complex concept negation H Role hierarchy sub properties rdfs subPropertyOf D Use of data type properties data values or data types Table 3 16 Used DL Expressivity of Ontology The initial model applies the DL expressivity with role hierarchy expressions es pecially for a hierarchical structure of recommendations and data type properties for attribute values of i
25. that the system is capable in product development acceleration but one par ticipant mentioned that it may even result in a deceleration of product development as designers adapt prototypes to user needs All in all this leads to better products which is the main issue of the framework Comments to statement The VICON virtual user concept is capable in product development acceleration No I don t think so S Thinking on evaluation However also after I told about Sketch Application the opinion still remained that the acceleration cannot be reached It may result even in deceleration But this is not very negative It will make better products and that s great It will make better products but it will not make it faster Its a matter of redoing things S It can lead sometimes to redoing things I could imagine that it would accelerate It depends on the realisation You can prevent big faults and big mistakes You can save money and time by virtual prototyping I think it s really good If it does it would probably accelerate in the wrong direction So we ll get not so good products very fast If you find a way to make it more accurate some products could be helped if they are easier to map The mobile phones are more complex than the tool currently can handle Currently the tool is oversimplifying the reality There might be products where it could help but for mobile phones it is too simpl
26. tl t2 t3 t4 t5 t6 t7 t8 7 8 beneficiaries reported that knob is hard to hold and turn around to set Program B tl t2 t3 t4 t5 t6 t7 5 8 beneficiaries reported that knob is too small to hold and control it tl t3 t4 t5 t7 All beneficiaries successfully identified the set Program C tl t2 t3 t4 t5 t6 t7 t8 7 8 beneficiaries reported that knob is hard to hold and turn around to set Program C tl t2 t3 t4 t5 t6 t7 5 8 beneficiaries reported that knob is too small to hold and control it tl t3 t4 t5 t7 7 8 beneficiaries have successfully under stand the Program Guide tl t3 t4 t5 t6 t7 t8 All beneficiaries have some problems to un derstand some parts of guide tl t2 t3 t4 t5 t6 t7 t8 5 8 beneficiaries reported that it is hard to find and understand programmes t2 t3 t4 t6 t7 4 8 beneficiaries reported that text was hard to read without glasses t5 t6 t7 t8 All beneficiaries successfully identified the Main Control panel tl t2 t3 t4 t5 t6 t7 t8 5 8 beneficiaries mentioned that some of the text is not easy to understand tl t2 t3 t4 t5 All beneficiaries successfully identified the Minor Controls panel tl t2 t3 t4 t5 t6 t7 t8 4 8 beneficiaries reported that some of la bels are not easy to understand tl t3 t4 t5 183
27. 0 without having prior knowledge Table 5 6 Results of questions regarding conformity with user expecta tions 110 5 EVALUATION The first statement asks how time consuming it is to learn how to use the soft ware see table 5 6 In average the result of this statement was slightly positive among participants concluding that it takes some time to learn The next both statements consolidate this statement regarding the software is easy to learn with out prior knowledge help or manual with a strong agree by participants This goes in line with the conclusion of results that a user manual would be advantageous 5 EVALUATION 5 2 5 Discussion The evaluation of the software framework by designers showed that user involvement methods applied by design studios of participants vary the assumption can be made that the software can be included independently of the current used method In conclusion two main issues were identified e The software is suitable to be included in existing product development pro cesses independently of user involvement method applied This issue is the main result of this part of the evaluation All 7 customer involvement methods were covered by participants but also each participant stated a positive feedback Especially results about the necessity of design rec ommendations see second question of table 5 3 or figure 6 of the annex empha sized the need by designers for a recommendation drive
28. 4 279 287 1985 James Rumbaugh Ivar Jacobson and Grady Booch Unified Modeling Language Reference Manual The 2nd Edition Pearson Higher Education 2004 ISBN 0321245628 S J Russell and P Norvig Artificial intelligence a modern approach Prentice hall 2010 R C Schank and C K Riesbeck Inside computer understanding Five programs plus miniatures Lawrence Erlbaum 1981 A Schmidt and K Van Laerhoven How to build smart appliances Personal Com munications IEEE 8 4 66 71 2001 Wolfgang Schneider Ergonomische Gestaltung von Benutzungsschnittstellen Kom mentar zur Grundsatznorm DIN EN ISO 9241 110 Beuth Verlag GmbH 2008 Sascha Segan Enter the phablet A history of phone tablet hybrids http www pcmag com slideshow story 294004 enter the phablet a history of phone tablet hybrids 2012 151 REFERENCES REFERENCES E H Shortliffe Computer based medical consultations MYCIN volume 388 Elsevier New York 1976 Howard E Shrobe Supporting and optimizing full unification in a forward chaining rule system In AAAI pages 710 715 1993 H Snellen Dr H Snellen s Probebuchstaben zur Bestimmung der Sehsch rfe H Peters 1863 S Staab and R Studer Handbook on Ontologies Springer Verlag Berlin Heidelberg 2009 T Strang and C Linnhoff Popien A context modeling survey In Workshop Proceed ings 2004 O Strnad A Felic and A Schmidt Context management f r selbst adaptive nutzer
29. 4 9 presents the use case diagram for this phase Analogue to the sketch design phase the module needs a Socket Server available to access all relevant data for this phase The administrator starts the server as described in Vicon Consortium 2011b for interface purpose The designer starts Siemens NX and creates a virtual prototype in the virtual environment Figure 4 10 presents the selection of an internal special toolset role in Siemens NX provided by the installation program having the possibility to start the CAD module from the internal Siemens NX toolbar In four steps recommendations are achieved for a design 1 Create CAD prototypes As seen in the Use Case Diagram see figure 4 9 the user needs to have an existing object e g press button as cube in the virtual environment for an notation by the module Figure 4 11 presents the title screen of Siemens NX including an imported CAD model 2 Start the CAD module To start the CAD module a shortcut button was included into Siemens NX see the small V Icon in the upper corner of figure 4 11 After start the annotation view is presented 3 Annotate CAD objects Designers are able to add semantic information about CAD objects using the Annotation View of the CAD module Figure 4 12 presents the annotation of a CAD object as a press button 88 4 APPLICATION IN DEVELOPMENT PROCESS san 2 K o BB AO Cx y Befehis D mm pa ZIRE Kein Auswahlfilter Gesamte Baug
30. 95 dB or more People who suffer from profound hearing loss are very hard of hearing and rely mostly on lip reading and or sign language Table 3 2 Audiometric descriptors and hearing loss according to the Eu ropean Group on genetics of hearing impairments EGGHI by the variability of different illnesses and test procedures an abstraction of visual preferences of a person into three different profile groups concerning no mild and moderate visual impairments was used see table 3 3 Regarding manual dexterity impairments there are two ICF definitions available B710 represents the functions of the range and ease of movement of a joint focusing upon all different functions regarding the mobility of single joints B730 concentrates upon the force generated by contraction of different muscles and muscle groups 48 3 KNOWLEDGE MANAGEMENT Visual descriptors Definitions of visual ability No visual impairment Mild visual impairment Moderate visual impairment The subject does not use glasses and does not have any restrictions of visual ability Mild visual impairments result in the use of glasses Subject is slightly sensitive to light and glare without glasses things appear to be indistinct or blurry and does have some minor problems to adjust to changes in light levels The user does have moderate impairments regarding vision Glasses are necessary to see distant objects due to a mode
31. Computing Human Computer Interaction Series pages 23 48 Springer London 2013 ISBN 978 1 4471 5081 7 doi 10 1007 978 1 4471 5082 4 2 URL http dx doi org 10 1007 978 1 4471 5082 4_2 148 REFERENCES REFERENCES Patrick Langdon and Harold Thimbleby Inclusion and interaction Designing inter action for inclusive populations Interacting with Computers 22 6 439 448 2010 P M Langdon and P Biswas Clustering user data for user modelling in the guide multi modal set top box In Patrick Langdon John Clarkson Peter Robinson Jonathan Lazar and Ann Heylighen editors Designing Inclusive Systems pages 195 204 Springer London 2012 ISBN 978 1 4471 2866 3 doi 10 1007 978 1 4471 2867 0 20 URL http dx doi org 10 1007 978 1 4471 2867 0_20 Ora Lassila Ralph R Swick World Wide and Web Consortium Resource description framework rdf model and syntax specification 1998 M Lawo P Kirisci M Modzelewski J O Connor A Fennell T Fiddian H G kmen M Klann M Geissler S Matiouk and Y Mohamad Virtual user models ap proach and first results of the vicon project eChallenges e 2011 Conference Pro ceedings 2011 Rensis Likert A technique for the measurement of attitudes Archives of psychology 1932 Quentin Limbourg Jean Vanderdonckt Benjamin Michotte Laurent Bouillon and Victor L pez Jaquero Usixml A language supporting multi path development of user interfaces In Engineering human com
32. Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig 3 1 3 2 3 3 3 4 3 5 3 6 4 1 4 2 4 3 4 4 4 5 4 6 4 7 4 8 4 9 4 10 4 11 4 12 4 13 4 14 5 1 5 2 5 3 Age groups of participants 6000000 46 Simulation of a vision impairment with cataracts and macular degeneration sss v w us ee w a 4 X AUS UN aS 47 Hands affected by rheumatoid arthritis in early intermediate and late PASES laa a WN ia S a e rik ae 50 The Gandalf User Model 54 The reasoning approach o ooo 66 Recommendation Rules to create User Model Recommendation for impaired groups 4 6000 68 The complete software framework containing back and front end applications usso s ma dh a m a a pe a de bs 8 73 User input of the designer supporting the sketch design and CAD phases of Product Development Process 76 VSF Manifest xml example file providing the selections of the de SINEP ai mase A AE aaa AS E A ee 78 VSF meta xml example file providing the annotations of compo MENU Parameters ws va bad A pl el 79 Sketch Design application On the left the user is able to select User Models typical Environments and typical Tasks A multiple selectionisalsopossible 81 Use case diagram UML of the Sketch Design application 82 Knowledge base inter
33. Moderate Moderate No i7 67 Male Mild Mild No i8 65 Female No Moderate No i9 68 Male Mild Moderate No ilO 75 Female No Moderate Mild ill 78 Male Mild Moderate No il2 70 Female No Moderate Mild il3 81 Female Mild Mild No il4 75 Female Moderate Moderate Moderate il5 65 Male Mild Mild No il6 65 Female No Mild Mild il7 68 Female No Mild Mild il8 79 Female Moderate Moderate No il9 77 Female Mild No Moderate i20 85 Male Mild Mild No tl 73 Female No Mild No t2 67 Male Mild Mild Mild t3 68 Male Mild Mild No t4 83 Male Moderate Moderate No t5 73 Male Mild Moderate No t6 71 Female Mild Moderate No t7 Continued on next page 166 APPENDIX Age Gender Hearing Vision Manual Beneficiary Code Dexterity g Germany i Ireland t Turkey 70 Female Mild Mild No t8 167 APPENDIX Device Images Doro PhoneFasy 332 Doro Mock Up created using the framework 168 APPENDIX Arcelik Grundig large silver Arcelik Grundig large black Arcelik Grundig small black 169 APPENDIX Ar elik Ar elik Washing Machine Panel 1 wa celik Arcelik Arcelik Washing Machine Panel 2 Prewash Quick Rinse Anti Wash Plus creating g 170 APPENDIX Doro PhoneEasy 332 Issue Participants Correctly identified on button i3 i4 i8 ill 114 117 i20 gl g2 g4 g6 g7 g8 g9
34. URL http www sciencedirect com science article pii 0148296382900340 Panagiotis Moschonas Athanasios Tsakiris Nikolaos Kaklanis Georgios Stavropou los and Dimitrios Tzovaras Holistic accessibility evaluation using vr simulation of users with special needs In UMAP Workshops 2012 A Naumann and M Roetting Digital human modeling for design and evaluation of human machine systems MMI Interaktiv 12 27 2007 Alan F Newell and Peter Gregor User sensitive inclusive design in search of a new paradigm In Proceedings on the 2000 conference on Universal Usability CUU 00 pages 39 44 New York NY USA 2000 ACM ISBN 1 58113 314 6 doi 10 1145 355460 355470 URL http doi acm org 10 1145 355460 355470 World Health Organization et al Deafness and hearing impairment See http www who int mediacentre factsheets fs300 en index html last checked 12 Nov 2012 2012a World Health Organization et al Visual impairment and blindness See http www who int mediacentre factsheets fs282 en index html last checked 12 Nov 2012 2012b M Peissner A Schuller and D Spath A design patterns approach to adaptive user interfaces for users with special needs Human Computer Interaction Design and Development Approaches pages 268 277 2011 Matthias Peissner Dagmar H be Doris Janssen and Thomas Sellner Myui gener ating accessible user interfaces from multimodal design patterns In Proceedings of the 4th ACM SIGCHI sy
35. Yes why some labels are in bold and italics others are not Sensory features Bold and italics used in labels Yes Yes Glossy finish on button surround Yes Yes Glossy finish on button No Yes No obvious audible click from No Yes buttons No obvious tactile click from No Yes buttons Parallax issues user needs to Yes Yes bend down to read lower control labels as the knob blocks view Start button is hidden among Yes Yes the other controls Start label is too close to program No Yes guide labels Cancel label is midway between Yes Yes two buttons Poor visual contrast between label Yes Yes and surround No colour contrast between No Yes buttons and surround Program selection knob cluttered No Yes 16 programs versus 12 on Panel 1 Labels wearing off Note this may No Yes be due to the fact that it is a prototype Physical features On off button is hard to press Yes Yes Start button is hard to press Yes Yes Difficult to hold and press Start Yes Yes button for three seconds Program selection knob difficult to Yes Yes turn Table 5 28 Result comparison regarding washing machines by expert Part A 136 5 EVALUATION Difficulty encountered Arcelik Existing Arcelik Mock up Panel 1 Panel 2 Physical features continued from Part A No home setting for the program Yes Yes selection knob so the starting point will change Program selection knob does not No
36. Yes give any tactile feedback on turning Program selection knob can sit No Yes between two settings i e does not click into place Difficult to press button Yes Yes repeatedly and buttons too close Yes No together Inconsistent font size on program No Yes selection knob Total number of issues 17 26 Table 5 29 Result comparison regarding washing machines by expert Part B was solved resulting in new accessibility issues As a result with respect to the hypothesis 3 3 the following issues were resolved e The use of product interfaces relies on readable and understandable labelling information so recommendations regarding labels and text information are very important Results of the study show a product can only be as much accessible as the user understands the features Country specific terminology or pictograms must be evaluated and applied to product interfaces for accessibility and a better comprehension by customers It is not advantageous to concern only one single recommendation all issues must be solved for a product to be more inclusive otherwise the modification can also imply new accessibility issues The coverage of each different recommendation is important If only one single recommendation is covered even new accessibility issues can appear as seen regarding washing machine panels Therefore recommendations can be func tions of each other e If al
37. al Kaklanis et al 2012b see chapter 2 6 3 for a more detailed review presented a different view including Virtual User Models for specification of impairment issues From the perspective of including user needs DHM systems highly focus upon substantial design studies during product development process and are not able to give the designer recommendations of how which parts of the product should be changed Designers are able to perform tasks in a virtual environment and to iden tify e g reachability issues The presented tools focus on the evaluation of products in a virtual environment As an input a virtual prototype of the product must already be available DHM tools are able to simulate tasks performed by virtual avatars providing indicators for ergonomic issues In terms of this thesis support of inclusive design must occur in earlier stages during first product drafts and CAD design 26 2 STATE OF THE ART Hold headlamp assuming horizontal force of 20 N Hold headlamp assuming horizontal force of 100 N Hold headlamp assuming horizontal force of 180 N Figure 2 7 Digital Human Model JACK presenting the task Hold Head lamp with three different force magnitudes Source Zhou et al 2009 27 2 STATE OF THE ART 2 6 Related Projects This thesis was created within the context of the VICON project to support designers of physical products like mobile phones washing machines or TV remotes b
38. base is defined as a flat hierarchy the inference Ontology contains a tree based taxonomy for the recommendation model 1 Classify User Model instances to user profiles mass customization see also Pine and Davis 1999 This first reasoning step will be needed to define different profiles based upon the possibilities and user needs of the beneficiaries 2 Add recommendations to each User Model profile class as members After this step recommendations can be connected to User Models 3 Add recommendations to each environment class as members Thus each environment instance is an instance and cannot contain members Comparison is made in lower case 67 GG W Ne OO 3 KNOWLEDGE MANAGEMENT as a pre step each instance needs to have an analogue class where members can be applied 4 Add recommendations to each task class as members This step is analogue to the previous environment step involving all recommen dations having an impact on specific task selections of the user 5 Add recommendations to each component class as members This step deals with the presentation of recommendations for the second phase CAD where based upon annotations of the components of the virtual prototype different recommendations should be presented 3 4 4 Ontology Inference The Forward Chain Reasoning steps of the Ontology create new classes Exemplary rules were used to create new User Model classes for each specified hearing im
39. be used Labelling Symbols should be accompanied by text Pass Room for improvement Symbols should be easily recognisable Pass Room for improvement The text and background colour combination should have Pass high contrast A clear open typeface font should be used for text Pass Text should not be placed over a background image or over Pass a patterned background White or yellow type on black or a dark colour is more Fail legible The typeface weight and size are suitable Pass Room for improvement Upper and lower case is used Pass Washing Machines Minimum strength is needed to open and close the door N A Controls are easy to grip and turn Fail easy to grip but stiff to turn The door opens flat or as wide as possible for maximum access N A The dome in the door does not provide an obstruction to N A access Wheels are added for ease of moving top loading machines N A Table 5 24 Recommendation Checklist of Arcelik Washing Machine Panel 2 Part B 134 5 EVALUATION Recommendation Result Washing Machines continued from Part B The door handle or button is easily activated N A The drawer for the soap powder is fairly large Pass Noise emission is at a minimum level N A Result 14 25 Table 5 25 Recommendation Checklist of Arcelik Washing Machine Panel 2 Part C
40. burden for the users regarding the necessity to learn how to use such a product Accordingly one main challenge and opportunity of human computer interaction is the involvement of each functionality in a respective and self descriptive way to the user On the other hand especially due to demographic changes user require ments must also be considered in the design process Existing guidelines and standards define approaches and recommendations regarding design issues related to different devices and user impairments but are not con sequently included in product development Designers have the challenge to respect both topics and create either individual products or products for an as wide spread customer group of people as possible This thesis describes a possible approach supporting designers with im pact in product development phases from the first stage While designers create product drafts and virtual prototypes they are able to get concept information about end user needs and requirements before physical pro totyping Outline This thesis consists of six chapters The first chapter will present the back ground motivation and all challenges within the topic of product develop ment with focus upon elderly user groups Also design recommendations are presented and clustered into semantic groups The second chapter has three main parts relevant approaches of data rep resentation existing methods to infer data and sim
41. button presents the letters SMS On the Mock Up the button is represented as an envelope also on the right side Open and read an incoming text message On both mobile phones participants had to press the message button to open and read an incoming text message 115 5 EVALUATION Source Model Physical Characteristics Image Doro Doro PhoneEasy 332 Small IEC 5010 power symbol on red button for on and off over a telephone handset Receive call button marked green with telephone handset SMS button for messaging Width 102mm length 50mm height 16mm doro FO Doro Doro High button spacing All buttons labels are white On and off button similar to PhoneEasy 332 but Mock Up without IEC 5010 power created symbol using the framework e Messaging button marked with an envelope e Width 123mm length 53mm height 16mm Table 5 9 Existing and emerged mobile phones used for evaluation of customer satisfaction 116 5 EVALUATION large silver Width 50mm length 224mm height 12mm at lowest 22mm at highest point height increases gradually from top to bottom Source ae Model Physical Characteristics Image e IEC 5010 power symbol e Volume Up Down on lower left as a right triangle marked with and e Channel Up Down on lower z elik right as P with and
42. by the re searcher il4 Continued on next page 175 APPENDIX Issue Participants All 39 beneficiaries were able to physically press the 5 1 7 buttons il i2 i3 i4 i5 16 17 i8 i9 ilO ill 112 113 i14 115 116 117 118 i19 i20 gl g2 g4 g5 g6 87 g8 g9 gl0 g11 g12 813 g14 g15 g16 817 818 819 g20 Ten beneficiaries reported difficulty reading the labels i2 i3 i4 i6 i8 i9 110 113 114 i19 25 28 beneficiaries successfully identified the location of compartment il i2 i3 i4 i5 i6 i7 i8 i9 i10 ill 112 113 il4 i15 116 i17 i19 i20 tl t2 t3 t4 t5 t8 24 28 successfully identified how to open the compartment il i2 i3 i4 i5 i6 i7 i8 i9 i10 ill i12 113 il4 i15 116 i17 i19 i20 tl t2 t3 t4 t5 Ten had difficulty when opening the com partment i2 i4 i5 i9 i11 i12 i14 i16 i18 t7 27 28 successfully identified how to close the compartment il i2 i3 i4 i5 i6 i7 i8 i9 i10 ill i12 113 114 i15 116 i17 i19 i20 tl t2 t3 t4 t5 t6 t7 t8 Eleven had difficulty when closing the com partment i4 i9 i11 il4 i15 i17 i18 i20 t6 t7 t8 Three beneficiary failed to successfully open or close the battery compartment il8 t7 t8 39 47 reported a general difficulty when reading the labels i2 i3 i4 i6 i7 i8 i9 i110 111 112 il3 i14
43. can also apply recommendations directly to existing virtual objects A server provides recommendations based on ontology based models and reasoning as seen in chapter 3 With respect to section 2 4 the framework can be seen as an extension to Quality Function Deployment by an iterative factor allowing designers to modify product de signs based upon predefined scenario specifications Chapter 5 will evaluate the presented framework by the impact on the product development process with involvement of designers and beneficiaries 94 Chapter 5 Evaluation This chapter concerns the third hypothesis of chapter 1 6 For a reasonable and comprehensible evaluation this hypothesis was split into three sub hypotheses The first sub hypothesis concentrates on the account of the general concept of the frame work and includes interviews with designers The second focuses upon the improve ment for the complete development process and was implemented including an on line questionnaire with end users of the framework and the third targets end users of the products customer satisfaction so real products were tested with beneficiaries While section 5 1 concerns the theoretical concept of the system section 5 2 focuses on the use of the software by designers Section 5 3 concentrates on the view by real people of the target group directly to see if the products can be used by a wider group of people 5 1 General Concept 5 1 1 Thesis and Predict
44. capabil ities of end users This thesis focuses upon the support and application of inclusive design theories principles and methods into the product development process to successfully inte grate end user requirements so the product can be used and accessed by the largest possible group of users see Kirisci et al 201 1b and Kirisci et al 201 1al Different projects deal with this topic to change and maintain product development process by creating guidelines for designers to add background knowledge about the end users further referred to as beneficiaries of the products One example is the exclusion calculator of the inclusive design toolkit Clarkson 2003 which de fines what percentage of users of a target population can not perform a specific task e g kneeling down Other projects like TIRESIAS Abbott 2007 or CARDIAC Car diac Consortium 2012 collect expertise based upon end user studies guidelines or other projects to present recommendations and information about design principles 1 INTRODUCTION and methods for different physical and non physical products These guidelines can not immediately be integrated into the product development process but rather should be seen as a type oriented textual collection of factors which should be advised while designing a product for beneficiaries By the defini tion of beneficiaries of products especially impaired and elderly people are addressed to maintain the abilit
45. capable to be included into real product development processes and does not affect existing product devel opment processes as an obstacle in typical product design Also a strong learning curve was observed raising context awareness of end customers on designer side However a broad expanse of recommendations regarding product interaction is re quired helping designer in the creation of suitable inclusive design for an as wide group of end customers as possible 140 Chapter 6 Discussion and Future Work 6 1 Discussion The aim of this thesis was a contribution to support product designers during the product development process solving the problem of inclusion of beneficiary needs More precisely designers should be able to access contextual information about customers of their product to include related issues as early as possible A framework was implemented and applied in industrial field The data used in the system has to be widely extended so also cognitive impairments can be included by the addition of new classes and rules analogously as existing impairment profiles described in section 3 2 1 and 3 4 4 Regarding the definition of cognitive parameters and the classification of User Models into no mild and moderate cognitive impairment groups user trials are necessary Also different new target products can be implemented One main factor during this thesis was the software not restraining the designer A possible sof
46. contain instances of User Models Environments Tasks Components and Recommendations Data properties representing attributes for instances were specified The properties will be more granularly described in 3 3 2 for the User Model 3 3 3 for the Environment 3 3 4 for the Task 3 3 5 for the Component and 3 3 6 for the Recommendation related attributes To represent all data an Ontology implementation was chosen due to aspects presented in 2 2 In summary this decision was endorsed by the following factors Object oriented data structure An Ontology formally represents knowledge data including instances and rela tions Each instance e g User Model can be related to different other classes and inherit various attributes like the age of a target user or if she or he needs glasses Highly adaptable vocabulary In addition or as a consequence of the object oriented data structure ontolo gies have the advantage to be highly adaptable to a problem by extending the Ontology vocabulary Availability of reasoning Aside of the main purpose of the application of ontologies reasoning is used to infer new states based upon initial models These engines can be used to automate classification processes and decisions There are multiple Ontology frameworks on the market with different pros and cons For the realization of the knowledge base Jena was used see McBride 2002 and McBride 2001 by concerning the following reaso
47. desired program this is a particular issue Table 5 17 General Comments Arcelik Washing Machine Panel 1 Part A 127 5 EVALUATION Positive Negative e The program selection knob is too difficult to turn The program selection knob does not give any tactile feedback when turned The start button is hard to press The on off button is hard to press The start button is a critical button which the user will use every single time the washing machine is in use but it is hidden amongst the other controls The location of the buttons relative to the order in which you use them is not intuitive The user presses On on the right then selects a program to the left then selects one or more of the minor controls to the left then presses Start to the right The layout should more closely mirror the user journey Difficult to press and hold the Start Pause Cancel button for three seconds e The Cancel label is mid way between two different controls Spacing should be used so that there is no confusion between buttons and their corresponding labels e The button must be pressed repeatedly to increase the time delay in increments of 5 minutes Table 5 18 General Comments Arcelik Washing Machine Panel 1 Part B 128 5 EVALUATION Positive Negative Clear typeface Visually clear and tactile marking on program selection knob
48. e1 a Open bottom drawer E Simulation result niea re Simulation result Failure Shoulder amp Simulation result Simulation result joint limit Success Wrist joint limit Success Success a2 c2 42 e2 Open top drawer Simulation result Simulation result Simulation result Simulation result Simulation result Failure Shoulder Failure Shoulder Success a e ir Success Success ES joint limit joint limit e3 b3 d3 e a3 c3 o de 5 A Simulation result Success d4 Simulation result Success e4 Table 2 4 Simulation results of the VERITAS project Source Kaklanis et al 2010 39 2 STATE OF THE ART An exemplary output of the VERITAS Project framework is presented in table 2 4 Each task is performed by a virtual humanoid with different impairments like rheumatoid arthritis resulting in a value for success or failure and the problem issue 2 6 4 VICON As previously mentioned this thesis evolved during the VICON project providing there a supporting framework for designers during product development process VICON aims to provide support to designers during the complete product development life cy cle allowing designers a recommendation driven product development as presented by this thesis but also to evaluate virtual products in a predefined virtual environ ment The virtual simulation platform VIRTEX is used to create a comprehensive scenario and t
49. expert of NCBI was conducted as part of the VICON project Results are divided into general com ments and a checklist evaluation with respect to recommendations produced by the system The expert evaluation was carried out with the devices Doro Mock Up Phone see table 5 9 Washing Machine Panel 1 and 2 see table 5 11 5 3 2 Experimental Setup The study took place in Ireland and Germany see Focus Group Report of the VICON project Vicon Consortium 2013b A total of 48 subjects participated all over the age of 65 All participants had at least one mild to moderate hearing vision or manual dexterity impairment appendix 6 2 presents the full list of participants Concerning the individual impairments table 5 8 presents the levels of each im pairment type by categories of none mild and moderate impaired subjects As washing machine panels were not available during the beneficiary tests an additional accessibility expert evaluation as an alternative to the beneficiary testing has been conducted to assess the washing machine panels 1One participant has had severe vision impairment so was excluded from the study 113 5 EVALUATION Age Male Female Total 65 69 years 5 6 11 70 79 years 1 9 10 80 89 years 5 8 13 90 years 0 5 5 Total 10 29 39 Table 5 7 Participants of the user study by age and gender Impairment type None Mild Moderate Hearing 19 19 9 Visi
50. find a virtual model e g in form of a wavefront obj file format Age Integer 65 116 Age in years Gender String M or F Gender of person Table 3 5 Ontology class data properties used for User Model General Characteristics The tables 3 5 3 6 3 7 and 3 8 present the different data properties of the User Model class Each parameter can be used to define a specific User Model instance and will be used to classify the instance as a member of impairment groups In order to the output of recommendations each User Model impairment group is resolved to emit different recommendations The reasoning classifies each single User Model instance into separate impairment profiles see section 3 4 4 53 3 KNOWLEDGE MANAGEMENT Hearing Predicate Datatype Description Hearing500Hz Integer 10 Threshold hearing level in dB at 500Hz 120 without aid Hearing 1kHz Integer 10 Threshold hearing level in dB at 1kHz 120 without aid Hearing2kHz Integer 10 Threshold hearing level in dB at 2kHz 120 without aid Hearing4kHz Integer 10 Threshold hearing level in dB at 4kHz 120 without aid SpeechWithBack Integer 0 Threshold of speech intelligibility with groundNoise 200 background noise as percentage of background noise volume compared to speech volume HearingAid Integer 0 No Indicator for worn hearing aid 1 Yes HearingAidWith Integer 0 No Will th
51. following sources e Information about available input and output devices from the Device Pro file e Information about user and environment from the User Profile e Customization settings that must be predefined by Ul developers of appli cations from the Customization Profile 2 UI Preparation Additionally the most suitable selection of Ul components is made in this stage including the following input e All possible application interactions are predefined in the Abstract Applica tion Interaction Model which defines different situations for each state of the application e To maintain the accessibility of the user interface requirements for end users of the interface are specified in the User Interface Profile and are related to the current user environment and device setup After this preparation a complete set of information about the current user device and interactions is available 3 UI Generation and Adaptation Based on previous data the interface is generated to user needs and can dy namic and system initiated be Ul adapted at runtime If the user changes the three stages of adaptation must be repeated e User Interface Generation This activity creates and renders the Ul based on all provided data The result is shown in the last image of figure 2 8 see Peissner et al 2012 User interface adaptations during use The possibility to adapt the Ul to the user is included If for instance new components and elem
52. http www cardiac eu org guidelines keys htm http www cardiac eu org guidelines telecoms mobile htm ComponentRule button height gt 5 Component turning knob press button Text People who rely on touch to operate keypads benefit from keys that are as distinctive as possible to the touch Raised keys are more easily distinguished than those that are flush against their surrounding Keys should therefore be raised above the body of the phone preferably by 5 mm 3 Level Table 3 15 One instance of the recommendation class defined parameters Analogue steps are performed for the classification of visual and manual dexterity impairments For the classification of recommendations to each selection of the user instances contain values to connect with various classes Table 3 15 presents one instance of the recommendation class including all defined attributes With respect to the purpose of defining recommendations based on different selections of the user each instance contains information about target User Models Environment Models Task Models and Component Models e User Model The impact between one recommendation instance and their importance to dif ferent impairments is described in the Profile predicate e Task The TaskRule predicate is used to describe the relation between tasks and rec 187 recommendation instances available in total 69 3 KNOWLEDGE MANAGEMENT
53. il5 il6 il7 il8 il9 i20 gl g2 g4 85 g6 87 g8 89 g10 g11 g12 g13 g14 g15 g16 g17 818 g19 g20 t7 t8 13 47 specifically reported a difficulty un These difficulties however relate to the buttons derstanding the labels or symbols that were omitted from the tasks above i2 i5 i8 i9 111 112 113 i20 gl g2 g14 t7 t8 Doesn t like the rocker switch 115 Grundig Large Black Remote Control Issue Participants 27 47 beneficiaries successfully identified the on off button at first guess il i2 17 i8 i9 i11 112 113 114 115 116 117 i18 gl g2 s4 g6 g8 g12 g15 816 g19 13 14 15 t6 t8 Two successfully identified the correct but ton for on only i4 i5 while one successfully identified it for off only i6 Continued on next page 176 APPENDIX Issue Participants Accordingly these three beneficiaries failed to recognise that on and off would be on the same button i4 i5 i6 Nine failed to identify either on or off cor rectly i3 i10 i19 i20 tl t2 t7 36 of the 39 beneficiaries expected the on off button to be at the top of the remote control il i2 i4 id i7 i8 ilO ill il2 113 il4 il5 i16 i17 i18 i20 gl g2 g4 g5 g6 87 g8 g9 g10 g11 g12 g13 g14 g15 g16 g17 g18 g19 g20 42 47 beneficiaries were able to physically press the on off button il i
54. ontologies as a knowledge base concen trate upon more domain specific approaches like KONWERK see G nter and Hotz 1999 and Funke and Sebastian 1996 KONWERK represents a modular configu ration tool which is able to perform domain specific reasoning including the specifi cation of a task configuration aim objects relations and previous knowledge about the configurational process Objects function as instances which can inherit prop erties representing e g domain specific preferences Using constraint propagation 20 2 STATE OF THE ART value ranges of the problem domain are successive narrowed by interpolation of con straints KONWERK consists of four basic modules focusing upon the following general tasks Representation of domain objects Domain objects define various models or virtual representations of all objects which are or can be involved in the problem domain area Representation and processing of relations constraints and heuristics Relations between all objects are used for the problem definition Formulation of the configuration task The task represents the problem that should be modelled Objectives or criteria for the goal system which should be maximized or minimized by selecting or constructing an appropriate solution must be defined to process a suitable configuration Control of the configuration task In addition the configuration task must be manipulable by the user to change the goa
55. possible without reducing the Pass distance between the keys to less than half the key width Ideally the keys should be internally illuminated but the Fail internal illumination should not reduce the legibility The visual markings on the keys should be high contrast Pass Room for clear and as large as is possible on the key top improvement The pressure to activate a key should be between 0 5 and N A 1 Newton There should be auditory and tactual feedback of control Pass activation For controls that do not have any physical travel audio or Pass tactile feedback should be provided so the user knows when the control has been activated e g a toggle switch or a push in pop out switch There is a clearly labelled reset control N A Buttons or keys have tactile markings Pass Room for improvement Buttons or keys including touch screen buttons are large Pass and easily identifiable from each other Buttons or keys are operable with one hand Pass Glare is minimised on the surface of the product Fail Instructions Program Guide Use simple clear concise language Fail Be task orientated Pass Use a typeface with good legibility Pass but bold and italics should not be used Labelling Symbols should be accompanied by text Pass Room for improvement Symbols should be easily recognisable Pass Room for improvement The text and backgr
56. suitability and usability of the software framework in product development environ ments An online questionnaire was done based on ISO 9241 110 Schneider 2008 with focus on usability and end user suitability of the software framework 5 2 1 Thesis and Prediction In this section the following hypothesis will be concerned Hypothesis 3 2 Improvement for Development Process The software framework is suitable to be adapted into existing product develop ment processes and can be used by designers without hindrance to their typical tasks The main question of hypothesis 3 1 deals with the manipulations of current product development processes by the system To analyse the change a questionnaire was created in which designers actively use the system It is based upon the ISO 9241 110 Schneider 2008 The result of this evaluation refers to designer acceptance directly but is also connected to user involvement methods applied by design studios Itis a crucial issue 101 5 EVALUATION for designers to use the presented tools without hindrance As seen in section 2 4 there is a variety of methods available to involve customers in product development The independent integration of applied user involvement methods during product development is crucial for a successful acceptance by designers so an optimal result would be positive regardless of the method applied by participants of the study 5 2 2 Experimental Setup Th
57. the second phase module recommendations can be applied which were already presented in the first phase software The VSF for the transfer between the second and the third phase 78 15 16 17 18 19 20 21 22 23 24 25 26 27 4 APPLICATION IN DEVELOPMENT PROCESS lt xml version 1 0 encoding utf 8 gt lt Product gt lt ComponentList gt lt Component type PressButton id button_on gt 0 009 gt lt LocalRotation x 0 y 0 z 0 gt lt Dimension x 0 014 y 0 007375001 lt Color r 0 g 0 b 0 gt lt MinimumForce gt 28 lt MinimumForce gt lt PressDepth gt 0 2 lt PressDepth gt lt LocalPosition x 0 01455851 y 0 04557789 z z 0 002 gt lt Component type Text id button_on_text gt lt LocalPosition x 0 y 0 z 0 001 gt lt LocalRotation x 0 y 0 z 0 gt lt Dimension x 0 014 y 0 007375001 z 0 0001 gt lt Color r 1 g 1 b 1 gt lt FontSize gt 12 lt FontSize gt lt Component gt lt Component gt Dresd lt ComponentList gt lt Information gt lt Name gt Doro 332gsm lt Name gt lt Type gt Cell Phone lt Type gt lt Vendor gt Doro lt Vendor gt lt OntologyServer 1p XXX XXX XXX XXx port 65000 gt lt Information gt lt Product gt Figure 4 4 VSF meta xml example file providing the annotations of com ponent parameters 79 4 APPLICATION IN DEVELOPMENT PROCESS a
58. to imagine designing a product for a fictional end user called Eileen where the Persona Eileen is described as followed Eileen retired a few years ago at the same time as her husband She has mild moderate hearing and manual dexterity impairments but they don t affect her everyday life She does not wear a hearing aid but is aware that she is listening to the TV much louder these days and would probably benefit from a hearing aid She has no visual loss so does not wear glasses or contact lenses Eileen is generally healthy and active In the week she helps look after her young grandchildren and at weekends she enjoys travelling and gardening with her husband This scenario also raises the issue to the participants How do you know what is important for the design ifyou do not have any experience with products for impaired people The sketch design tool can be used to get design support for this exact situation based upon different guidelines as already mentioned see 3 3 6 105 5 EVALUATION Questionnaire After the presentation of inclusive design issue and each software front end ques tions are raised The questions relate to ISO Norm 9241 110 see Schneider 2008 with four cate gories according to the following themes e Suitability for the task Does the software help designers to complete a task of product design for elderly people without burden e Self descriptiveness Each software tool explained suffic
59. values this issue refers to an abstract view on the product design interface During the product development process it can be very problematic to adapt to these issues automatically due to all different as sertions of natural speech This type of information is referred to as qualitative recommendations Due to the differences between qualitative and quantitative recommendations both types should be adapted and used separately Qualitative recommendations relate to non measurable challenges of a design and can be very abstract e g good contrast Also relations and functional dependen cies between values can be stated For this qualitative recommendations need to be included in an early stage of the design process so designers are able to incorporate them 10 1 INTRODUCTION Quantitative recommendations focus on nominal parameters of a design resulting in the need of an available virtual design of the product These recommendations should be included in later design phases as parameters can change during the de sign of a product e g total width of a device can be modified ifa new button is added 1 5 Research Questions The focus of this thesis contains research questions extracted from all different fields of the inclusion of a supporting system into the product development process First all different data must be usable included into a representation which can handle all different kind of data textual information images
60. 0 Bordered 2009 Figure 1 4 Projected population pyramid for EU 27 2060 excluding french overseas departments Source European Commission 2011 Figure 1 4 EUROPOP2008 convergence scenario see European Commission 2011 shows the projected population pyramid for the 27 European countries for the year 2060 compared to the pyramid of 2009 The amount of elderly people compared to 2009 will be much higher resulting in a more extreme scenario In the VICON project an ethnographic user study was carried out with elderly people see 3 2 Out of this various Personas were created defining average and abstract attributes for different mild to moderate impairments e Hearing Impairments Hearing impaired people have problems with acoustic feedback or acoustic in volvement e Visual Impairments Visual impaired users have problems with too small visual output e Manual Dexterity Impairments Manual dexterity impaired users have grasp problem e g if buttons are too small or too close together 1 INTRODUCTION 1 3 Inclusive Design Design represents the process of creation The concept of inclusive design deals with the capability to create and provide an interface which can be theoretically used by everybody This concept has gained many names Design for All Universal Design etc Newell and Gregor Newell and Gregor 2000 described inclusive design to be user sensitive with respect to the concept of universal usability Langd
61. 0 excluding french overseas departments s sos 5 The TIRESIAS project website containing guidelines for designers 6 The CARDIAC project website containing guidelines for designers 7 The exclusion calculator of the inclusive design toolkit 9 Exemplary output of the exclusion calculator of the inclusive de sign toolkit sx ww Se e wama ii rd e b 9 Product Development Process 15 ContextualExtendedORM 666 19 KONWERK example hierarchy of concepts 21 KONWERK example definition of a constraint 22 Methods of involvement reviewed 23 RAMSIS 3D CAD ergonomics tool presenting geometric kinematic digital human model 25 Digital Human Model JACK presenting the task Hold Headlamp with three different force magnitudes 27 Automatic adaptation with implicit confirmation 28 GUIDE an open architecture for various multi modal user in terface technologies 4 06 31 User Initialisation Application of the GUIDE Project 35 GUIDE Simulation of visual impairments without and with mild visual impairment 36 Simulation input of VIRTEX 40 Simulation interface of VIRTEX 41 Simulation output of VIRTEX 41 LIST OF FIGURES LIST OF FIGURES Fig Fig Fig Fig Fig Fig
62. 0 A li RS Q AA SS 133 Park A iii aaa qa k Uu a A A A le 133 PartB s sosoca s susta ag masala saa k Ua pa ae 134 Recommendation Checklist of Ar elik Washing Machine Panel 2 Patt aia Sa eee A S us k B New E a e e BS 135 Results of user study regarding washing machines 135 Result comparison regarding washing machines by user study 135 Result comparison regarding washing machines by expert Part A136 Result comparison regarding washing machines by expert Part B137 viii List of Theorems Page 1 Hypothesis Ontology based model application 12 2 Hypothesis Suitable Reasoning 12 3 Hypothesis Designer acceptance 60000600 12 3 1 Hypothesis General Concept e 95 3 2 Hypothesis Improvement for Development Process 101 3 3 Hypothesis Customer Satisfaction 113 Abstract In recent years more and more sophisticated devices are created includ ing an in the worst case exponential growth of functionality In current versions mobile phones are not just telecommunication devices but also a Camera music player browser email interface etc resulting in new terms like smart phones A television device can also be used as a browser using a wireless internet connection and washing machines contain more programmable functions than customers will ever need and use This com plexity can most often be reflected as a
63. 010 46 3 KNOWLEDGE MANAGEMENT 3 2 1 Impairments Before execution of a user study the first question relates to impairments and their definition Hearing impairment represent a total or partial loss of hearing ability in one or both ears ICF B230 see Organization et al 2012a With respect to this study a classification based upon the European Group on genetics of hearing im pairments EGGHI was used see Martini 1996 and table 3 21 Similar definitions can be found from the British Society of Audiologistics BSA and the Royal National Institute for Deaf and Hard of Hearing People RNID consequently indicating that there is a consensus for four hearing impaired levels see Vicon Consortium 2010 and table 3 2 Vision can be described as sensory function relating to sensing the presence of light and sensing the form size shape and colour of the visual stimuli B210 see Organization et al 2012b Figure 3 2 Simulation of a vision impairment with cataracts left and macular degeneration middle and right Source Vicon Con sortium 2010 A wide range of tests exists to measure different types of vision or vision impair ments Vision impairments can be very different due to specific issues dealing with sensorial functionality see examples in Vicon Consortium 2010 The most famil iar method of tests is the assessment of visual acuity using the Snellen chart see Snellen 1863 where a series of indivi
64. 2 Tools 260 V usuy u aer S Sup s AAA A W kuq aw zat 80 4 2 1 SketchbDesignTool 60600600 80 4 2 2 Administration Tool 00 0 83 4 2 3 CAD Mod le 244 42248468 80583544 OX Ge Eee Ee OS 86 4 3 Impact on the Product Development Process 91 4 4 Conclusion 4 452280 us 09 AAA RR ma a eS 94 iii CONTENTS CONTENTS 5 Evaluation 5 1 General Concept 5 1 1 Thesis and Prediction 5 1 2 Experimental Setup 5 1 3 Execution 5 1 4 Results 5 1 5 Discussion 5 2 Improvement for Development Process 5 2 1 Thesis and Prediction 5 2 2 Experimental Setup 5 2 3 Execution 5 2 4 Results 5 2 5 Discussion 5 3 Customer Satisfaction 5 3 1 Thesis and Prediction 5 3 2 Experimental Setup 5 3 3 Execution 5 3 4 Results 5 3 5 Discussion 5 4 Conclusion 6 Discussion and Future Work 6 1 Discussion 6 2 Future Work References Appendix List of Figures Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig Fig 1 1 1 2 1 3 1 4 1 5 1 6 1 7 1 8 2 1 2 2 2 3 2 4 2 5 2 6 2 7 2 8 2 9 2 10 2 11 2 12 2 13 2 14 Page BMW iDrive controller and display 2 Beneficiary user with mobile phone 3 Population pyramid for EU 27 2009 excluding french overseas departments 4 484045 44 nera e a Sp Q Yes 4 Projected population pyramid for EU 27 206
65. 2 i3 i4 i5 i6 i7 i8 i9 i10 ill 112 113 i14 115 116 117 118 119 i20 gl g2 g4 g5 g6 g7 g8 g9 g10 gil g12 g13 g14 g15 g16 817 g18 g19 g20 t7 t8 39 47 of the beneficiaries identified the vol ume up down buttons at first glance il i2 i4 i5 17 i8 i9 110 111 112 113 115 116 i17 i18 i20 gl g4 g5 g6 g7 g8 g9 gl0 g11 g12 813 814 g15 g16 817 818 tl t2 t3 t4 t5 t6 t8 Four mistook the P buttons for the vol ume buttons i3 i6 114 i19 All 47 beneficiaries were able to physically press the volume up down buttons il i2 i3 i4 i5 i6 i7 i8 i9 i10 ill i12 113 il4 115 116 117 118 119 i20 gl g2 g4 g5 g6 g7 g8 g9 g10 gil g12 g13 g14 g15 g16 g17 g18 g19 g20 t1 t2 t3 t4 t5 t6 t7 t8 42 47 beneficiaries successfully identified the 5 1 7 buttons il i2 i3 i4 i5 i6 i7 i8 i9 i10 ill i12 113 i15 il6 il7 i18 i19 i20 gl g2 g4 g5 g7 gll g12 g13 g14 g15 g16 g17 g18 tl t2 t3 t4 t5 t6 t8 All 39 beneficiaries were able to physically press the 5 1 7 buttons il i2 i3 i4 i5 i6 i7 i8 i9 i10 ill i12 113 il4 115 116 117 i18 119 i20 gl g2 g4 g5 g6 87 g8 g9 gl0 gll g12 g13 g14 g15 g16 817 818 819 g20 Five beneficiaries were unable to identify how to find channel 517 g6 g8 g9 g10 t7 One beneficiary declined to carry out the task s
66. 2 to be open Manipulation is possible using the administration software lt xml version 1 0 encoding UTF 8 gt lt Phasel gt lt UserModel gt Gandalf lt UserModel gt lt Environment gt Cellar lt Environment gt lt Task gt Check wash dial lt Task gt lt Phasel gt lt Phase2 gt lt CADFilename gt Arcelik WashingMachine stp lt CADFilename gt lt Phase2 gt Figure 4 3 VSF Manifest xml example file providing the selections of the designer 4 1 4 Vicon Status Files The Vicon Status File VSF is used regarding intercommunication between the phase specific applications In the first phase the designer creates a product as a draft getting textual qualitative recommendations to consider ensuring inclusive design These recommendations are also relevant for the second phase software in which the user creates a virtual prototype of the same draft product Vicon Status File can be used to represent the setting from the first phase to the second phase Vicon Status Files are containers including documents or other files Each VSF contains one main file Manifest xml which describes the selection which is already made in a previous stage of product development Figure 4 3 presents an exemplary Manifest xml file defining information about the first phase concerning a selection of the User Model Gandalf the environment Cellar and the task Check wash dial Using these selections in
67. 3 Personal knowledge of participants about Inclusive Design 156 APPENDIX 10 participants 0 a T a T not at all slightly moderately familiar very familiar familiar familiar familiar Figure 4 Personal knowledge of participants about Virtual User Mod elling VUM Suitability for the task 10 8k g2 E a 61 8 5 q a 4k 2 0 I skip Figure 5 Question if the software provides a wide choice of scenarios 157 APPENDIX participants 0 5 I 5 yi skip Figure 6 Question if the design recommendations of the toolset are nec essary 10 8k 8 E a 61 8 gt q a Ak ar skip Figure 7 Question if it takes a short time to go through recommendation list 158 APPENDIX 10 participants an skip Figure 8 Question about the need of an user manual 10 participants a 7 a iu skip Figure 9 Question if the look and feel of the application is suitable and pleasant 159 APPENDIX 10 participants o EZ a skip Figure 10 Question if the software is easy to use in general Self Descriptiveness participants 0 l l l I E skip Figure 11 Question if the description of information in the sketch appli cation for user profiles is comprehensible 160 APPENDIX 10 participants l
68. 32 scale with 7 values was presented Statements The following statements were presented 1 The VICON virtual user concept is capable in supporting the designers in cre ating inclusive products This statement focuses on the complete concept of the VICON project including the framework part as described by this thesis but also an evaluation software part VIRTEX for simulation of virtual products is sufficient to create products for a wider group of customers 2 The VICON virtual user concept can help to involve the user s perspective into the development process earlier This aims especially the product development phases presented by this thesis to enhance and improve contextual information into the process of design in which designers are creating the product by sketch and the CAD software 3 The VICON virtual user concept is capable in product development accelera tion Regarding all parts of the software framework design and evaluation parts should support designers in terms of time which also results in cost decreases of product development 1See Vicon Consortium 2012b of the VICON project for more information about VIRTEX 96 5 EVALUATION 4 The VICON virtual user concept provides knowledge concerning disabilities and derived requirements As a final statement but also for future purposes of the software in terms of learning of contextual information by designers about end customers this state
69. 4 144 means a complete a partial and no fulfilment of the requirement 45 3 KNOWLEDGE MANAGEMENT 3 2 User Study With respect to the topic of the VICON project a detailed ethnographic research was carried out with involvement of a group of elderly people and designer groups This user study was executed by Royal National Institute for Deaf People RNID 2010 see Vicon Consortium 2010 and involved a test scenario including washing machines and mobile phones The target group contained 58 elderly people who had a range of three different types of mild to moderate WHO classified impairments see Stucki 2005 Hearing loss B230 sight loss B210 and manual dexterity B710 730 Each participant had either one minor developed physical impairment or a combination of all target impairments Figure 3 1 presents the age groups of all 58 participants With respect to their impairments the age is relevant in order to ensure the classification of mild to moderate impairments IThe aim of this thesis refers to a part of the VICON project Virtual User Concept for Supporting Inclusive Design of Consumer Products and User Interfaces The project deals with the support through the complete product development phases including an evaluation of the target product in a virtual envi ronment 2 m65 69 Mm70 74 m75 79 m30 84 Mg85 89 90 94 m 95 99 Figure 3 1 Age groups of participants Source Vicon Consortium 2
70. 9y JO AWN urei8erp seo sn 6 y IMSY S19 9We J1ed uu q Da esapnpul gt uonepu uuuo35 til Ajddy lt spuayxe gt th jnpow NOIA HE JS in faynasuo gt s adAqoj01d JENA 9 lqelleAe sanas 794205 AS suonmepu uiuo35 t MOLA s uauodwo sysel S USWUOJTAUZ 43A49S 19 90S S911J0Jd 1 POW 4asn 404 uoo A HEJS SUOT EPUsWWOIZY ad 10301d ENYA ponaysuoD lt sapnpul gt A Ss XN suauwaIs HEJS uot eat iddy ubTs q AVI NOJIA a3u61s3a 13sn Y JojeajsiulWpYy Y 87 4 APPLICATION IN DEVELOPMENT PROCESS eters For instance an external parameter such as recommended button size can be used in context of defining a keypad with round buttons and in parallel a keypad with square buttons The final prototype will not provide a complete set of templates but the templates for the most typical components such as keypads displays etc used in mobile phones washing machines and TV remotes Additionally the user has the possibility of defining own customized templates supported by the core parameters that are provided by the recommendation system Product development workflow The integration of inclusive design into the CAD process was developed as a module within Siemens NX The user designer is able to connect to the socket server and to get support by visual recommendations Recommendations can be directly applied to objects Figure
71. An ontology based approach to achieve inclusive design support in the early phases of the product development process UJ Universitat Bremen Markus Modzelewski Department of Computer Science University of Bremen FB3 A thesis submitted for the degree of Dr Ing February 11 2014 First Reviewer Second Reviewer Prof Dr Michael Lawo University of Bremen Department of Computer Science Dr Patrick Langdon University of Cambridge Department of Engineering If you can t explain it simply you don t understand it well enough Albert Einstein Acknowledgements First of all I would like to thank my advisor Prof Dr Michael Lawo for motivating and supporting me during my work This thesis would not be possible without him I would also like to thank Dr Patrick Langdon as my second advisor for support and feedback whenever needed Both always had an open ear for me and it was a pleasure to work with them Pierre Kirisci and Patrick Klein from BIBA supported my work with many discussions and concepts I always enjoyed our work on projects I would like to thank the VICON team Dr Antoinette Fennell and Joshue O Connor from CFIT Dr Yehya Mohamad and Svetlana Matiouk from Fraunhofer FIT Haluk G kmen from Arcelik Thomas Bergdahl and Christina Johansson from DORO without whom the VICON project and so my thesis would not be successful Philipp Klaffert and Dr Hannes Baumann supported me by comments and rev
72. Hands affected by rheumatoid arthritis in early intermedi ate and late phases left to right Source Vicon Consortium 2010 3 2 2 Methodology In order to define problems of each target group related to impairment levels as presented in 3 2 1 the following methodology was carried out 1 Introduction of the researcher and briefly to aims of this study An introduction should give the subject a proper view of issues and topics 2 Application and realization of each task The subject performs different tasks with product The tasks were predefined and describe a typical use 3 After completion a questionnaire was used to figure out problems related to impairments and functionalities 50 3 KNOWLEDGE MANAGEMENT In this step some problematic areas could be identified while performing differ ent tasks 4 In addition the researcher records observations of each task to define how many participants have had problems with their task This is necessary for further observations and extraction of results that were not covered by the questionnaire 3 2 3 Outcomes Regarding this thesis especially two outcomes were relevant The identification of problematic issues while performing different tasks by the subjects resulted in tex tual recommendations for the designers For further information see D1 1 of the VICON project Vicon Consortium 2010 The other outcome is the separation of subjects into different profiles concerni
73. Mac OS Self Descriptiveness Statement skip The description of information in O 1 1 6 3 0 the sketch application for user profiles is comprehensible The description of information in O 1 1 4 3 2 the sketch application for environments is comprehensible The description of information in O 1 1 7 2 0 the sketch application for tasks is comprehensible The description about 0 1 0 7 3 0 recommendations is comprehensible Table 5 4 Results of questions regarding self descriptiveness Information on user profiles environments tasks and recommendations is col lected next see table 5 4 The results show a trend towards approval most partic ipants answered concluding a general approval but also some ambiguities with respect to comprehension of each model More information about the scenario would be advantageous As presented in figure 5 3 part A meta information about the cur rent scenario selection can be seen in the software For instance impairment levels 109 5 EVALUATION for each user model instance are presented but designers do not exactly understand the meaning An improvement of the scenario explanation would be beneficial Conformity with user expectations Statement skip The software has a consistent 0 0 3 7 1 0 structure The layout was as expected 0 1 2 6 1 1 Some features
74. NCES REFERENCES Vicon Consortium Project Deliverable 1 3 Virtual Humans in a human centred design process a critical review 2010b Vicon Consortium Project Deliverable 1 4 Functional and system requirements dossier 201la Vicon Consortium Project Deliverable 3 1 System Architecture and Interface Spec ification 2011b Vicon Consortium Project Deliverable 2 2 Virtual User Model Final release 2012a Vicon Consortium Project Deliverable 3 4 Final prototype of the virtual user mod elling software framework 2012b Vicon Consortium Project Deliverable 4 3 Evaluation report on how convenient it is to use Virtual User Model and adapted prototype 2013a Vicon Consortium Project Deliverable 4 4 Focus Group Report 2013b H Wache T Voegele U Visser H Stuckenschmidt G Schuster H Neumann and S H bner Ontology based integration of information a survey of existing ap proaches In IJCAI 01 workshop ontologies and information sharing volume 2001 pages 108 117 Citeseer 2001 X H Wang D Zhang T Gu and H K Pung Ontology based context modeling and reasoning using owl In Pervasive Computing and Communications Workshops 2004 Proceedings of the Second IEEE Annual Conference on pages 18 22 IEEE 2004 D Waterman A guide to expert systems Addison Wesley Pub Co Reading MA 1986 David Wechsler Manual for the wechsler adult intelligence scale 1955 D D Woods Decomposing auto
75. S text message voice mail or email but also different new feature applica tions such as facebook google messenger skype chat etc One solution implemented by the framework includes internationally recognised standards in icons pictograms and symbols as recommendations This part of the evaluation also covers a comparison of output of the framework with the real prototype design Recommendation lists provided by the framework were compared with the Doro Mock Up phone see 5 13 if the recommendations were applied correctly or not One issue not included in the Mock Up relates to holding issue of beneficiaries with a weak grip Since the material of the Doro Mock Up phone and so weight and surface material of the functioning phone was not available results 10 of 12 issues for Doro Mock Up shows that recommendations were almost properly applied 125 5 EVALUATION TV Remotes In this section results of the TV remote user study with beneficiaries are presented Task Grundig large Grundig large Grundig small silver black black Identify on button 31 39 22 39 37 39 Press on button 39 39 39 39 39 39 Identify the volume 29 39 32 39 37 39 button Press Volume 39 39 39 39 39 39 up down key Identify the channel 38 39 35 39 24 39 up button Press the channel 39 39 39 39 39 39 up button Identify location of 19 20 39 39 36 38 compartment to change batteries Ide
76. T PROCESS 4 2 3 CAD Module Overview During the CAD design phase designers use commercial CAD software In the VI CON project Siemens NX e g was used see Vicon Consortium 2010a and Vicon Consortium 2010b In order to push the sketch phase recommendations to the CAD system a software module using NX s API toolkit has been developed Siemens modelling environment has a collection of API toolkits called NX Open NX Open al lows access and manipulation of models designed with NX as well as customization of the NX user interface to suit individual needs The Common API toolkit compatible with the requirements as defined in Vicon Consortium 201 1a exposes the same ob ject model for a number of programming languages Java C etc A comprehensive understanding of the core concepts such as how the API exposes objects within NX is necessary Itis an advantage that the common API gives access to the same object model used by NX developers Extensive interfaces can be established with the modelling environment The ele ments of the object model are semantically incomplete Modelling environment con centrates upon primary visual and surface parameters functional parameters are not included Siemens NX offers the possibility to add custom parameters which does not need to be related to current components These attributes are stored within the component and can be used to represent functional parameter values like the force needed to p
77. TE OF THE ART Modelling Framework For purpose of simulation and adaptation the GUIDE Project conducted different user trials Three impairment levels for each modality of visual hearing manual dexterity and cognitive impairments were implemented based on a qualitative user study The steps for this approach were see Guide Consortium 201 1 1 2 8 Obtain and collate survey data and user trial data Reduce the dimensionality of the data set by eliminating highly correlated vari ables Cluster the survey data for each modality Vision Hearing Cognition Physical Reduce the dimensionality of the data set by eliminating non significant vari ables in the k means clustering Take the resulting clusterings and characterize the cluster centres in terms of the combined contributions to the clusters Repeat for User trial data Examine distance of users from cluster centres as indication of sensitivity to adaptation Improve and refine with additional data and overlapping clustering techniques The full data set contains 46 users with different impairments at the age range of 49 90 years It includes a variety of parameters to specific capabilities of each user Based on these parameters a k means Clustering was applied with 3 clusters for low medium and high levels for each impairment type k 3 see Kanungo et al 2002 Non significant variables were eliminated due to their contribution to t
78. TE OF THE ART Social TV Upload Schedule Feedback Home gt Entertainment Select the Video El CricketWorldCupWinningMoment Duration 00 02 11 Ma Failure is the pillar of success Select the Language English e o pi Duman OPH tanve a Pa pmi Om ba t peer Figure 2 11 GUIDE Simulation of visual impairments without top and with mild visual impairment bottom Source GUIDE Con sortium b YesPrimeMinisterBBCcomedy Duration 00 02 52 NAMMA_OORU_final i Duration 00 03 09 Notification 0 Logout Search stream Search EaglesHotelCaliforniaLive Duration 00 06 34 Motivational_Video Duration 00 03 25 India UK Advanced Technology Centre of Excellence in Next Generation Networks Systems and Services IU ATC nam esas turan Seb ab 36 HS w O Coma Duramam ei RR Cation of Cantons e Mari ewe sin ers her ard ea IU ATC 2 STATE OF THE ART 2 6 3 VERITAS Project The VERITAS Project Virtual and Augmented Environments and Realistic User In teractions To achieve Embedded Accessibility Designs focuses on a virtual simu lation framework including end user impairments to infer problematic usability is sues Kaklanis et al 2012b Using this designers are able to simulate end user behaviour when performing predefined tasks with virtual product prototypes Con textual models including various general values but also impairment specific values
79. able to physically press the up button il i2 i3 i4 i5 i6 i7 i8 i9 i10 ill i12 113 i14 115 116 i17 118 i19 i20 gl g2 g4 g5 g6 87 g8 g9 g10 g11 g12 g13 g14 g15 g16 g17 g18 g19 g20 tl t2 t3 t4 t5 t6 t7 t8 One beneficiary declined to carry out this task g2 43 46 beneficiaries successfully identified the location of compartment i2 i3 i4 i5 i6 i7 i8 i9 i10 ill i12 113 114 115 il6 117 i19 i20 gl g2 g4 g5 g6 87 g8 g9 g10 gil g12 g15 g16 g17 g18 g19 tl t2 t3 t4 t5 t6 t7 44 46 beneficiaries successfully identified how to open compartment il i2 i3 i4 i5 i6 i7 i8 i9 i10 ill i12 i13 il4 115 116 117 118 119 i20 gl g2 g4 g5 g6 g7 g8 g9 gl0 gll 812 g15 g16 817 818 819 tl t2 t3 t4 t5 t6 t7 t8 Seven had difficulty when opening the com partment i9 i10 i11 i114 i19 i20 g20 t3 All 46 beneficiaries successfully identified how to close compartment il i2 i3 i4 i5 i6 i7 i8 i9 i10 ill i12 113 i14 115 116 i17 118 i19 i20 gl g4 g5 g6 87 g8 g9 g10 g11 g12 g13 g14 g15 g16 g17 g18 g19 g20 tl t2 t3 t4 t5 t6 t7 t8 Four had difficulty when closing the com i3 i9 i10 t4 partment Issues reported by the beneficiaries il i3 i4 i5 i6 i8 i9 i11 i12 i13 i15 i17 n 22 47 in relation to the look and feel of the keys and controls r
80. aised keys are more easily distinguished than those that are Rush against their surround Keys should therefore be raised of the phone preferably by mm above the body Figure 5 3 Mainframe explanation of the sketch design tool during the online questionnaire There is section A Meta Information about the current scenario selection is seen Also impairment levels for each user model instance are presented in this field HPro filel mild hearing impaired HProfile2 moderate hearing impaired VProfile for visual and MDProfile for manual dexterity impaired User Model 104 5 EVALUATION In section B designers can select user model representatives Personas typical environments in which the target product can be used and typical tasks which can be performed using the product A multiselection is also possible A search field is presented in section C in which designers can search for rec ommendations e g the search term button will result in a recommendation list with only relevant button recommendations Section D presents a list including all resulting recommendations based on the current selection of User Model Environment and Task or search The full text of one single recommendation is presented in section E It includes a name required impairment profiles source summary and a text containing relevant information For the test a use case scenario is described The participant has
81. an obstruction to the designer In the first phase draft the user is able to use a stand alone system to get recommendations for the desired tar get group and product type These textual suggestions for the design process refer to different aspects of the product which are already specified during drafting For instance if the product should be for visual impaired people the designer should be aware of large fonts as soon as possible in the product development process Some recommendations which should be considered do have a very strong geometrical 92 4 APPLICATION IN DEVELOPMENT PROCESS form and surface impact so already in this phase the designer must be aware of dif ferent aspects of the end users The second tool can be used within daily work software of designers as an inte grated module of the CAD software Siemens NX This results in acceptance by the users getting qualitative but also quantitative recommendations directly applied to product parameters All tools create a user awareness with the designers for a proper understanding of impaired user s needs Product development is no longer an encapsulated pro cess see 2 1 in which not only the surface of a product and its functional design aspects are considered but also context of the focus group typical environments tasks and component specifications The method provides user involvement by using existing user studies with specified target groups Also the context inform
82. and P Robinson Towards a practical inclu sive design approach In Proceedings on the 2000 conference on Universal Usability pages 45 52 ACM 2000 M Kifer Rule interchange format The framework Web reasoning and rule systems pages 1 11 2008 Michael Kifer rule interchange format logic programming s second wind In Pro ceedings of the 20th international conference on Inductive logic programming ILP 10 pages 1 1 Berlin Heidelberg 2011 Springer Verlag ISBN 978 3 642 21294 9 URL http dl acm org citation cfm id 2022735 2022737 P Kirisci P Klein M Modzelewski M Lawo Y Mohamad T Fiddian C Bowden A Fennell and J Connor Supporting inclusive design of user interfaces with a vir tual user model Universal Access in Human Computer Interaction Users Diversity pages 69 78 2011a P T Kirisci K D Thoben P Klein and M Modzelewski Supporting inclusive product design with virtual user models at the early stages of product development In Proceedings of the 18th International Conference on Engineering Design ICED 11 Vol 9 pages 80 90 201 1b C Kiss Composite capability preference profiles cc pp Structure and vocabularies 2 0 W3C Working Draft 8 2006 Pat Langdon Developing an interactive tv for the elderly and impaired An inclu sive design strategy In Pradipta Biswas Carlos Duarte Patrick Langdon Luis Almeida and Christoph Jung editors A Multimodal End 2 End Approach to Ac cessible
83. and down before green button i15 to answer call Pressed red button at first attempt to an g10 swer call Unclear speech due to crackling or feedback il g5 Difficult to hear speech i2 i7 87 gl5 Does not use text messaging i9 i10 i19 g2 g5 g6 g7 g8 g9 g13 g14 Failed to successfully type the complete text message i6 i7 i14 g10 g18 g19 7 24 assumed the phone was set to predic tive text il i2 i3 i13 g10 g18 g20 17 24 assumed the phone was set to non predictive text i4 i5 i8 111 i15 116 117 118 i20 gl g4 g11 g12 g15 g16 g17 g19 18 27 had difficulty with typing double let ters in a word In all cases the beneficiaries pressed the button for the second time too quickly i2 i3 14 i8 111 113 115 117 i20 gl g4 gl0 g12 815 817 818 819 g20 Successfully identified text message button il 12 i3 14 18 ill 117 gl g4 gl0 g11 g12 g15 g16 817 818 819 g20 Successfully opened text message il i3 14 i5 17 i8 ill 113 114 115 116 117 118 i20 g1 g4 g10 g11 g12 g15 816 817 g18 g19 g20 Continued on next page 172 APPENDIX Issue Participants Successfully read text message il i3 i4 i5 17 i8 ill 113 114 115 116 117 118 i20 g1 g4 g10 g11 g12 g15 g16 g17 g18 gl9 g20 6 38 reported that the labelling was too small i9 i10 i14 g
84. as de signers in the whole product development process Haines et al 1998 By being a part of design and physical construction of the product this approach focus upon experience of all participants of product development Sundin et al 2004 Figure 2 5 presents the outcome of the review of the above mentioned methods In relation to different product development phases each method has its pros and cons Three main impact fields for customer involvement were identified These in clude specification concept development and prototyping Related to this the sis a separation of product development into phases is suitable There is no single best method for all products The most suitable customer involvement method is defined by cost time and suitability of end product Customer involvement methods are used to get feedback and reactions stepwise or during product development It is advantageous to create a possibility to include as much of this information into early stages of product development as possible 24 2 STATE OF THE ART As mentioned in the last point a knowledge base including as much context informa tion about end users is advantageous In the next chapter 3 the possibility to create a suitable knowledge base will be discussed 2 5 Digital Human Models In current design approaches DHMs digital human modelling is used for a virtual representation of humans in a virtual environment DHMs like RAMSIS o
85. ation is used through the complete product development process to support designers in the sketch design and CAD phase Referring to section 2 4 the software framework can be seen as an extension to Quality Function Deployment by an iterative factor In addition to QFD in each phase the scenario is specified and used to generate quality function similar recom mendations based upon predefined specifications 93 4 APPLICATION IN DEVELOPMENT PROCESS 4 4 Conclusion Based on the implementation of the knowledge base in the previous chapter 3 this chapter presents the implementation of the framework providing a support for de signers by providing recommendations based on a pre specified set of target User Models typical environments and typical tasks which can be performed using the product A system architecture including designer front ends was presented and tools im plemented for the purpose of a phase based supporting framework providing design ers with qualitative and quantitative recommendations as seen in chapter 3 3 6 Hypothesis 2 Suitable Reasoning Ontology based models can be used to give statements from knowledge base for specified scenarios described by the questions of who is using a product where to perform what task The result presents the answer to hypothesis 2 as a standalone tool presenting qualitative recommendations for sketch design phase and an integrated module in a CAD environment Siemens NX which
86. beyond the device it came with For example a DVD player remote may be programmed to operate Contact us 5 ta various brands of televisions The future of remote controls Touchscreen remotes These remote controls feature an LCD screen that can be either monochrome or full color The buttons are actuallyimages on the screen which when touched will send infra red signals out to control devices Some models have multiple screens that are accessed through buttons on the touch screen and other models have a combination of the touchscreen and hard traditional buttons Some models of the touchscreen remotes are programmed using a graphical interface program on a PC which allows the user to customize the screens backgrounds buttons and even the actions the buttons perform This project that is created is then downloaded into the remote through a USB cable or in most recent models bywireless Universal Remote Consoles ie Figure 1 6 The CARDIAC project website containing guidelines for de signers Source Cardiac Consortium 2012 1 INTRODUCTION Regarding existing tools for designers applications like the exclusion calculator of the inclusive design toolkit Clarkson 2003 focus upon a selective result of rec ommendations for designers based upon input of specific user impairments Figure 1 7 presents the calculator and a selective input of requirements for visual hearing cognitive and manual dexterity
87. biquitous computing systems Regarding requirements of the creation of a knowledge base including human en vironment task and other factors in terms of this thesis a different main focus is aimed 1 Partial validation pu Due to requirements of this thesis various models and relationships must be described e g User Model profiles or recommendations based upon different values Additionally a correct syntactical inference is needed for the purpose to provide accurate data and correct scenarios 2 Level of formality for The level of formality describes how precise contextual facts and interrelation ships between instances and models can be represented Regarding require ments as presented in this thesis formality is a very important issue to indicate different values abstract nominal etc in one and the same model 3 Applicability to existing environments app Applicability represents the possibility to use the knowledge base in different other applications This feature is relevant especially regarding future possibil ities like import of and export into other knowledge bases 4 Distributed composition dc This requirement is irrelevant with respect to existing server client architecture 44 3 KNOWLEDGE MANAGEMENT for maintenance purposes see requirement dossier of the VICON project Vicon Consortium 201 1a 5 Richness and quality of information qua With respect to sensorial data this requirement descr
88. ble Install System is a professional open source system to create Windows installers It is designed to be as small and flexible as possible and is therefore very suitable for the presented framework See http nsis sourceforge net 2The questionnaire can be executed using a browser and the address http 134 102 95 211 eval 3See http www w3 org html 4See http php net 102 5 EVALUATION Online Questionnaire for the evaluation of an ontology based approach to achieve inclusive design support in the early phases of the product development process This questionnaire is part of my doctoral thesis which is also a part of the european funded research project VICON The project is investigating the potential of user modelling for designing inclusive products My thesis An ontology based approach to achieve inclusive design support in the early phases of the product development process focuses upon the impact and extension of support for designers for the creation of products especially for elderly people will use the results of this Hypothesis 3 2 Improvement for Development Process questionnaire to answer hypothesis 3 2 of my thesis The software framework is suitable to be included into real product development processes and can be used by designers without hindrance to their typical tasks It will take about 25 minutes and includes 24 questions and is opened until 02 08 2013 All data will be held confidentially an
89. button The channel up button is located on the lower half of the Grundig silver large remote control On the Grundig black large it is realized as the P in the middle and on the Grundig small black as Identify location of compartment to change batteries On all remote controls the location of compartment is on the lower back side of the device Identify how to open battery compartment This task aims to describe problems with the handling force and precision needed to open the compartment Washing Machines figure 5 11 Identify the on off button On both washing machine panels the on off button is located on the right side of the panel Push on off button This task aims to describe problems with the force and precision needed to push the button 119 5 EVALUATION Identify the set Program On both washing machine panels the program setting was realized as a rotary knob in the middle right of the panel Turn knob to set Program This task aims to describe problems with the force and precision needed to rotate the knob Identify main control panel On both panels the main control panel is located in the lower middle Read and understand texts of main panel This task aims to describe problems with labels of the rotary knob Identify minor control panel On both panels the minor control panel is located in the middle Read and understand texts of minor panel This task aims to describ
90. ccessfully identified the volume up down buttons il i2 i3 i4 i5 i8 i9 111 il2 113 114 115 il6 i17 i18 i19 i20 gl g2 g4 g5 g6 87 g8 g9 g10 g11 g12 g13 g14 g15 g16 g17 g18 g19 g20 tl t2 t3 t4 t5 t6 t7 t8 Five mistook the up down buttons for the volume buttons at first i3 i4 i5 i6 i14 Four mistook the mute button for the vol ume button at first i7 ilO ill i14 Of the nine beneficiaries who failed to iden tify the buttons at first guess three failed to correctly identify the volume up down but tons at all i6 i7 i10 All 47 beneficiaries were able to physically press the volume up down buttons il i2 i3 i4 i5 i6 i7 i8 i9 i10 ill i12 113 114 i15 il6 117 i18 119 i20 gl g2 g4 g5 g6 87 g8 g9 g10 g11 g12 g13 g14 g15 g16 g17 g18 g19 g20 tl t2 t3 t4 t5 t6 t7 t8 Continued on next page 179 APPENDIX Issue Participants 32 47 beneficiaries successfully identified the channel up button il i2 i4 i5 i8 i9 ilO 114 116 i17 118 i20 g4 g8 g9 g10 g11 g12 g13 g14 g15 g16 817 g19 tl t2 t3 t4 t5 t6 t7 t8 Nine beneficiaries suggested the Menu but ton as an option i8 i15 i17 i18 i19 i20 g2 g8 g18 Six beneficiaries were unable to suggest a button that might bring them to channel 517 i3 i6 i7 i111 112 i113 All 47 beneficiaries were
91. cher as to where the on button was i6 i8 i9 110 i13 i15 i18 Continued on next page 173 APPENDIX Issue Participants Once told how to turn the phone on 9 19 successfully turned the phone off without prompting il i2 i4 i5 i7 i8 i9 i10 i13 All of the beneficiaries successfully dialled the number il i2 i3 i4 i5 i6 i7 i8 i9 i10 ill 113 il4 i15 i16 i17 i18 i19 i20 18 19 of the pressed the green button to connect the call beneficiaries successfully il i2 i3 i5 i6 i7 i8 i9 i10 i11 i13 i14 il5 i16 i17 i18 i19 i20 handed as desired 1 19 pressed the green button before di i4 alling the number 1 19 was unable to use the phone one il 8 16 successfully guessed that the envelope button was linked to text messaging i2 i3 i4 i8 113 116 117 i18 All 16 successfully read and identified the letters on the keys il i2 i3 i4 i5 i6 i7 i8 i11 113 114 i15 il6 i17 i18 i20 All 16 successfully pressed the correct keys il i2 i3 i4 i5 i6 i7 i8 i11 113 114 i15 il6 i17 i18 i20 All users were happy with the size of the nu meric buttons and with the labels on those il i2 i3 i5 i6 i7 i8 i9 i10 i11 i13 i14 115 116 i17 i18 i19 i20 tween buttons buttons 1 19 user would prefer one handed use il 2 19 users would prefer less spacing be
92. cified EnvRule parameter of cach recommendation which defines when a single recommendation should be presented see a more detailed review in section 3 3 6 and 3 4 4 3 3 4 Task The task class represents one specific task which the beneficiaries can perform using the product Each task refers to a different set of recommendations General characteristics Predicate Datatype Description Name String Name of the task is presented to the user in the Ul IDName String The ID name is unique for each object of the Ontology class Nr Integer 1 999 The task number identification code unique for every task Description String Textual description of the task Impairment String Each impairment profile can be defined here as in recommendation class as comma separated values for impairment groups for direct connection see table 3 13 Component String Specific component name involved in a task see Component Model Continued on next page 60 3 KNOWLEDGE MANAGEMENT Complexity Integer 0 Not The complexity estimate of a task complex 1 Medium complex 2 Severe InputRequired Integer 0 No Identifies if an input to the task object 1 Yes is required Input String Input character chain if required can be extended to regular expression describing the input InputDescription String Optional textual description of the input NumberOf Integer 1 99 Number of s
93. d anonymously The questionnaire is completely voluntary You may decline to answer any question or stop filling in the questionnaire at any time and for any reason When the data is shared described or interpreted there will be nothing to identify you or your company If you have any questions or additional feedback don t hesitate to contact me by E Mail Markus Modzelewski Figure 5 1 Introduction of questionnaire Questions related to your company and pre experience The following questions will be used to determine the method of customer involvment Please choose the following options if applicable If there is no customer involvement please continue We start with a check list for the design about what key features are needed and use this list through the m complete design process QFD a We create non functional prototypes and evaluate them with our customers user oriented product development a We evaluate our first sketches with our customers directly concept testing a We evaluate first functional prototypes with customers beta testing e Groups of target customers create first designs with support by our designers consumer idealized design a Users who face needs directiy and benefit most by our products design our products lead user method O We do not have any comitted designers at all and create the product design by other workers and customers participatory ergonomics Figure 5 2 Ques
94. de of the TreeView provides a legend and an orientation control frame Additionally by holding the right mouse button and moving the mouse for ward backward the perspective zooms in out of the TreeView Administrators are able to add new instances to each class by using the Add new Instances dialogue in the knowledge base view A description for the predicates is presented too The Predicate Tool of the Administration software provides the feature to change the predicates of each Ontology class VUMS Cluster XML files can be imported and exported directly to the Ontology 84 4 APPLICATION IN DEVELOPMENT PROCESS ConVic x File Edit Import Export Look amp Feel Recommendations Knowledge Base Treeview Meta Inf l User Model Information Recommendation List J Name VProfile2 User Profile VProfile2 is R 00 Good visual contrast between buttons and their labels buttons and their sourround Lowimportance Groups R 01 1 i i ium i Dasctintian Moderate visual impaired user hi 01 Improve the tactile detection of key tops medium importance R 02 Large keys for better differentiation high importance Scenario l ii R 03 Illuminated keys For bad lighting conditions medium importance User Model ES No Selection Element Recommendation View HProfile1 User HProfile2 User MDProfile1 User Name Large keys for better differentiation
95. dual letters decreasing in size are presented on a wall chart and the person is asked to read the chart from a specified distance The resulting measure of visual acuity VA indicates an individual s ability to read the chart in comparison with an individual with perfect visual acuity Determined lAudiometric Descriptors are based on the average of the pure tone hearing threshold levels at 250 500 1000 2000 and 4000Hz 47 3 KNOWLEDGE MANAGEMENT Audiometric descriptors Definitions of hearing loss dB Mild hearing loss Moderate hearing loss Severe hearing loss Profound hearing loss On average the most quiet sounds that people can hear with their better ear are between 25 and 40 dB People who suffer from mild hearing loss have some difficulties keeping up with conversations especially in noisy surroundings On average the most quiet sounds heard by people with their better ear are between 40 and 70 dB People who suffer from moderate hearing loss have difficulty keeping up with conversations when not using a hearing aid On average the most quiet sounds heard by people with their better ear are between 70 and 95 dB People who suffer from severe hearing loss will benefit from powerful hearing aids but often they rely heavily on lip reading even when they are using hearing aids Some also use sign language On average the most quiet sounds heard by people with their better ear are from
96. ducted to rate the accessibility compared to recommendations given by the system The comparison and the expert study identified the following issues e The use of product interfaces relies on readable and understandable labelling information so recommendations regarding labels and text information are very important Participants have had several problems regarding the identification of single 139 5 EVALUATION functions with respect to their icons or characteristics For instance the SMS button on mobile phones could not be identified correctly in Ireland as it is not commonly used for messages as e g in Germany Icons instead lead to overcome barriers related to language or literacy but the functional meaning of logos can also be ambiguous It is not advantageous to concern only one single recommendation all issues must be solved for a product to be more inclusive otherwise the modification can also imply new accessibility issues This issue was raised regarding the washing machine panels in which one single recommendation was solved but 10 new accessibility issues appeared If all recommendations are included product interfaces can be used by a wider group of people Most participants were satisfied with the new accessibility and stated the same recommendations as the system The evaluation concludes a positive feedback from both perspectives designers and end customers but with additional comments The system is
97. e 99 5 EVALUATION The last statement issues the aim of the framework to provide knowledge con cerning disabilities and derived requirements which except one participant agreed One participant neither agreed or disagreed and stated that the system provides use ful knowledge but it needs sophisticated data about the end users such as a high amount of recommendations Comments to statement The VICON virtual user concept provides knowledge concerning disabilities and derived requirements Yes its what it is about It provides me with this knowledge You learn a lot about humans with disabilities The text is always about hu mans and devices Well there was some really good knowledge The list from the sketch tool is useful but if you base test on the too limited data it could mislead the designers in their process Further suggestions for improvement Usage of VICON in any CAD software 1 VICON could give us physical data dimensions colours if needs light and sound forces 2 This data must be given to the engineers during design interactive on time I missed an active part of designing E g if I would be designing a mobile phone I would like to combine the designing part in the CAD program and directly get a visual feedback notifying me about some problems Sometimes there is no target user group specified so it would be helpful to have a possibility to adjust t
98. e and solve problems In terms of this work section 2 2 describes possible approaches of a knowledge base Furthermore section 2 3 deals with the question of how to use connect and infer the data for a manipulable system as required These systems are referred to as expert systems Requirements as presented in 1 5 as well as daily life situations are governed by deterministic rules Rule based expert systems represent an efficient and compre hensive way to handle knowledge base information by functions and inference The concept of expert systems emerged in the late 1960s see Davis et al 1977 im cluding systems like DENDRAL Lindsay et al 1993 or MYCIN Shortliffe 1976 which focus upon the medical field using rule based engines Analogously Schank and Riesbeck 1981 wrote AT has gotten into the knowledge business in a big way in the late few years partially because of the success of MYCIN DENDRAL and other programs see Schank and Riesbeck 1981 Since then the field of expert systems grew continuously extended and was used in new domains see Castillo and Alvarez 1991 Castillo et al 1997 Hayes Roth et al 1984 Waterman 1986 and Giarratano and Riley 1998 Otherwise the field of ontologies is growing similarly as expert systems see Wache et al 2001 Staab and Studer 2009 and Russell and Norvig 2010 filling the gap between knowledge management and reasoning Current expert systems implementing
99. e when a rec ommendation is relevant In this thesis only the Sketch and CAD phases are focused 3 4 Reasoning A reasoning step is needed to infer from available data information As presented in the previous chapter all relevant recommendations to the user are based upon a predefined setting This section deals with the syntax used for the realisation and presents the complete reasoning approach The Jena framework used here includes a general purpose rule based reasoner henceforth referred to as generic rule reasoner which is able to apply rules to the current Ontology state with the output of a new state 3 4 1 Rules Rule bare rule or bare rule or ruleName bare rule bare rule term term gt hterm hterm forward rule or bhterm lt term term backward rule hterm term or bare rule term node node node triple pattern or node node functor extended triple pattern or builtin node node invoke procedural primitive bhterm node node node triple pattern functor functorName node node structured literal node uri ref e g http foo com eg or prefix localname e g rdf type Table 3 14 Informal description of the simplified text rule syntax of rea soner Source The Apache Software Foundation 2013 Rules are used to infer from one state and setting to new states by application of rule sets For instance all Use
100. e conversion of consumer demands into quality characteristics and the iterative development of a design quality function describing a relation between consumer and product In QFD the Thttp www w3 org 2005 rules wiki RIF_Working_Group 22 2 STATE OF THE ART Concept Detailed Specification Development Design _ Prototyping Final Product Phase of i i i the design Design OFD process for i Z User oriented Product Development Design i with Concept i Beta f Testing i testing Lead User Method Design by lt Customer Idealized Design Participatory Ergonomics Type of customer involvement Figure 2 5 Methods of involvement reviewed Source Kaulio 1998 only contact point of designers and consumers is before the creation of the product to specify all relevant parameters e User oriented product development In relation to QFD user oriented product development focuses upon the in volvement of the user after the first prototype creation It includes an use analysis phase into product development in which prototypes of the target product are used by beneficiaries Rosenblad Wallin 1985 Due to cost in tensive prototype generation this method is mostly suitable for products in which functional issues are primarily important Concept testing This method connects first sketch designs of the target product with an evalua tion by customers Concept testing should be supplemented with late
101. e framework consists of two different end user applications as explained in chap ter 4 Both applications can be used after the installation using a software installer implemented using a scriptable install system During the online questionnaire the installer can be downloaded and used to set up the Sketch Design Tool 4 2 1 and the Siemens NX module which is installed automatically if a local installation of Siemens NX is available In the current version only Windows OS is supported The evaluation itself was implemented using HTML and PHP to create a questionnaire capable of providing an installer during the process but also raising the questions 5 2 3 Execution Methodology The aim of this evaluation is to obtain a value for designer acceptance and the im pact on existing product development processes As target participants 11 physical product designers were interviewed The first questions referred to the familiarity of participants with inclusive design and Virtual User Models Figure 3 presents a bar graph of participant knowledge about inclusive design Most participants are partners of the VICON project so they were already familiar Virtual User Model see 3 3 contain contextual information about target end users of products VUMs were more often discussed and reviewed during the project process than inclusive design in general so participants were more familiar with this term see figure 4 INSIS Nullsoft Scripta
102. e problems with labels of the minor panel Product Comparison A product comparison was made and separated into three feature categories Cognitive features Potential issues identified in both panels included those relating to the order of use and possible difficulties with interpretation of labels use of unfamiliar terms and visual formatting to communicate information Sensory features Potential issues included by increasing the labels the spacing between labels decreased and they became harder to distinguish as well as a lack of audible feedback from the buttons Physical features Potential issues related to the buttons and controls being difficult to press and turn 120 5 EVALUATION 5 3 4 Results The following results show the summary of the user trials complete tables are included in appendix 6 2 An expert evaluation was conducted by an NCBI accessi bility expert The results are divided into general comments positive and negative and the results of the checklist evaluation Mobile Phones In this section results regarding mobile phones are presented It consists of four parts General comments table 5 12 present the expert evaluation by an NCBI accessibility expert Table 5 13 shows the checklist of recommendation provided by the software framework also by NCBI accessibility expert Table 5 14 summarizes the results of the beneficiary study with real participants as mentioned in section 5 3 2 The la
103. e user wear hearing aid when Product 1 Yes using this kind of product Table 3 6 Ontology class data properties used for User Model Hearing Gandalf 80 Gandalf is an active older gentleman who refuses to let his age stop him from do ing things He has a moderate hearing loss and wears digital hearing aids all day long He can follow conversa tions in quiet places without them but the aids make his life much easier Due to his moderate visual impairment he wears his new varifocal glasses all of the time Moderate arthritis in both hands does not stop him doing things but can cause him discomfort especially in cold weather So he often wears gloves in all seasons except the height of summer Gandalf still drives a car and enjoys walk ing his Labrador dog He lives alone and tries to go to as many daytime social events as he can for company and enter tainment Figure 3 4 The Gandalf User Model Source 2012a Vicon Consortium 54 3 KNOWLEDGE MANAGEMENT Vision Predicate Datatype Description VisualAcuity Integer 0 Visual acuity describes the sharpness Normal 1 of vision value of normal Mild 2 20 12 5 20 25 mild 20 32 20 63 Moderate moderate 20 80 20 160 FieldOfVision Integer 0 No Reduced field of vision finds it hard to 1 Slightly 2 see things to the side top bottom of Moderately 3 what they are looking at Str
104. elated to confusion about the functions of certain buttons Most commonly the buttons that were not tested in the tasks above il8 i19 i20 g5 g6 g13 g16 g17 t3 t7 Continued on next page 180 APPENDIX Issue Participants One beneficiary mentioned the glossy finish g14 on the remote control would get dirty as there were fingerprints on it Two beneficiaries reported that buttons were t3 t7 difficult to push 181 APPENDIX Arcelik Washing Machine Panel 1 Issue Participants All beneficiaries successfully identified the on off button tl t2 t3 t4 t5 t6 t7 t8 Two beneficiaries reported that button was hard to hold and push t3 t7 All beneficiaries successfully identified the set Program A tl t2 t3 t4 t5 t6 t7 t8 6 8 beneficiaries reported that knob is hard to hold and turn around to set Program A tl t3 t4 t5 t6 t7 One beneficiary mentioned that knob would be hard to use when hands are wet or soapy t7 All beneficiaries successfully identified the set Program B tl t2 t3 t4 t5 t6 t7 t8 6 8 beneficiaries reported that knob is hard to hold and turn around to set Program B tl t3 t4 t5 t6 t7 All beneficiaries successfully identified the set Program C tl t2 t3 t4 t5 t6 t7 t8 6 8 beneficiaries reported that knob is hard to hold and turn around to set Prog
105. en as a logical sequence of consecu tive steps The complete sequence of product development is most often separated into specific phases The Association of German Engineers VDI described this pro cess in various guidelines see VDI Gesellschaft Konstruktion und Entwicklung Produktionstechnik ADB Gemeinschaftsausschuf Produktplanung 1980 VDI Gesellschaft Entwicklung Konstruktion Vertrieb 1993 and VDI Gesellschaft Kon struktion und Entwicklung Produktionstechnik ADB Gemeinschaftsausschuf Produktplanung 2004 which are established on an european level and often in cluded into companies structures Vicon Consortium 2010a Especially VDI 2221 describes an accurate hierarchy based upon the main phases draft phase con cept phase and elaboration phase Entwurfsphase Konzeptphase and Ausar beitungsphase including a definition of requirements functional parameters and drafts in the first geometrical modelling and form design in the second and prototype construction in the third phase 14 2 STATE OF THE ART Since it is not possible to define a process which is valid for all products and all issues this thesis focuses upon a product development process including the first two phases henceforth referred to as sketch design and CAD computer aided design phases In the draft sketch design phase designers create prototype drafts As already mentioned the priority in this phase lies within t
106. ented in this thesis as different user tests would be necessary to obtain the information which devices are suitable for which functionality When restricted to single devices results using the presented framework can be optimal Regarding a more idealistic view an optimal scenario would be to get rec ommendations by target functionality or functionalities The target functionality would be the main hierarchical root defining suitable de vices The current evolutions such as smart phones are in line with this theory Existing devices could be selected by their suitability based on a set of target func tionalities Even new devices could be generated by need if a set of functionalities can not be provided by existing ones 6 2 Future Work As mentioned in chapter 3 the data used in the presented framework can be widely extended The software is already used in industry and is available as open source A further development is advantageous especially due to the connection of function ality and device With respect to the software framework but also inclusive design the following areas would be interesting e Standardisation of User Model The VUMS cluster prepared a position paper providing input to the standardi sation of User Models Based on User Models of the projects VERITAS VICON MyUI and GUIDE a standard definition of a representative virtual user includ ing parameters for hearing visual manual dexterity and cognitive impai
107. entioned a CAD module in Siemens NX was implemented Using the module the designer is able to view recommendations for the current prototype but also applies rules like e g The minimum size of a button for visual impaired users is 1cm These quantitative recommendations as defined in 1 4 always refer to a specific component of the prototype see 4 2 3 for a detailed review 3 3 6 Recommendation The recommendation class defines the presented output of the system for the user Both qualitative and quantitative recommendations see 1 4 can be represented General characteristics Predicate Datatype Description Name String The recommendation name will be presented in the Select Recommendation Form Priority Integer 1 Low The importance level of one 2 Middle 3 recommendation High priority means High that the recommendation is a MUST HAVE Summary String An optional summary of a recommendation Text String The complete text of a guideline recommendation Continued on next page 62 3 KNOWLEDGE MANAGEMENT Source String The source of a recommendation e g ISO Guideline or experience Attachment String An URI Uniform Resource Identifier see Masinter et al 2005 where an attachment can be found Profile String The profile or profiles of a recommendation used for the rules 6 profiles are available VI1 and VI2 for mild a
108. ents have been selected this activity triggers adapta tions to the current available Ul so the new components can be included at runtime Profile Updates Regarding a user change the stages must be re initiated again triggered by this activity 30 2 STATE OF THE ART 2 6 2 GUIDE Project The GUIDE Project Gentle user interfaces for elderly people see Langdon and Biswas 2012 Biswas et al 2012 and Langdon 2013 is targeting Web applica tions and related platforms The aim is to create a software framework and design tools for developers to integrate accessibility issues and personalization features into applications E ICT Application Logic Multimodal Input Fusion Multimodal Output Fission y at Multimodal Adaptation se gt User model i mulatior The GUIDE principles Figure 2 9 GUIDE an open architecture for various multi modal user interface technologies Source Jung and Hahn 2011 By using a variety of human interaction modalities as e g speech commands or visual gestures a logic controller can react and infer the most suitable configuration of an input device related to the customer see figure 2 9 User Models including parameters describe end user capabilities regarding impairments preferences are used by reasoning for scenario definition These User Models are based on various tests and user trials with elderly and impaired customers Jung and Hahn 2011 31 2 STA
109. esWithProduct 1 Manual dexterity Predicate Value Arthritis 1 Grip 3 Buttons 2 Discomfort 2 TouchSensitivity 2 Table 3 9 User Model definition for Gandalf Using the inference model the designer can select one single User Model Gan dalf including different categorical and nominal values see table 3 9 resulting in the output of all recommendations referring to impairment groups of the selected User Model instance As a pre inference the User Model Gandalf is classified into specific impairment profile groups so the system is able to connect the Persona to recommendation in stances Section 3 4 4 will present the reasoning in more detail 57 3 KNOWLEDGE MANAGEMENT All attributes data properties are also included in a cluster submission as part of the VUMS cluster Interoperable and Inclusive User Modelling concept for Simulation and Adaptation Kaklanis et al 2012a which deals as a definition which can be used by all VUMS projects VERITAS Chalkia et al 2010 VICON Lawo et al 2011 GUIDE Jung and Hahn 2011 and MyUI Strnad et al 2012 3 3 3 Environment The environment model is used to classify most used environments to represent dif ferent aspects of environments e g lighting levels as nominal and abstract values Each environment instance contains of different categorical or numerical values rep resenting different aspects of an environment
110. esponse to inclusive design Designing accessible technology pages 31 39 2006a J Goodman PM Langdon and PJ Clarkson Providing strategic user information for designers methods and initial findings Designing accessible technology pages 41 51 2006b K Goodwin Getting from research to personas Harnessing the power of data Cooper Newsletter 2002 GUIDE Consortium User Initialization application Prototype http www guide project eu index php mainItem Publications subItem Project Deliverables amp pageNumber 1 item 38 a Accessed 2013 08 21 GUIDE Consortium User Simulator Prototype http www guide project eu index php mainItem Publications amp subItem Project Deliverables amp pageNumber 1 amp item 22 b Accessed 2013 08 21 Guide Consortium Project Deliverable 7 1 Initial User Tests and Model 2011 A G nter and L Hotz Konwerk a domain independent configuration tool In Con figuration Papers from the AAAI Workshop pages 10 19 1999 HM Haines JR Wilson Health and Nottingham United Kingdom Safety Executive London United Kingdom Institute for Occupational Ergonomics Development of a framework for participatory ergonomics Sudbury HSE Books 1998 T A Halpin A J Morgan and T Morgan Information modeling and relational databases Morgan Kaufmann 2008 146 REFERENCES REFERENCES P Hamisu G Heinrich C Jung V Hahn C Duarte P Langdon and P Biswas Ac ce
111. ework in real product development processes and focuses on the suitability to support designers during product development without hindrance Product design ers were able to install and test the software framework in their typical environment to validate if the support is productive and can be included in existing processes without hindrance to typical design tasks Results show a positive acceptance by designers even throughout different user involvement methods as presented in chapter 2 4 In conclusion two main issues were identified e The software is suitable to be included in existing product development pro cesses independently of the user involvement method applied All seven different customer involvement methods were covered by participants of the study as presented in section 5 2 Although the acceptance by designers depends on the output of recommendations itis necessary for better acceptance to extend the amount of recommendations e An improvement regarding more comprehensive scenario information is advan tageous Regarding a more sophisticated scenario comprehension more background in formation is needed This improvement can be made by an extended user man ual including background information or a further presentation in the software framework The last sub hypothesis adds the perspective of product customers by comparing emerged products created using the system with existing ones In addition an expert evaluation was con
112. face of the Administration software 83 Designer role view with selection of VProfile2 User Model 85 Use case diagram UML of the CAD module 87 Special Vicon Role selection in Siemens NX 89 Example of a loaded CAD file DORO mobile phone 90 Annotation of a CAD object as a press button 90 Recommendation view in CAD Module 91 Impact on different methods of involvement 92 Introduction of questionnaire 00 103 Questions related to customer involvement method 103 Mainframe explanation of the sketch design tool during the on line questionnaire 104 List of Tables Tab 2 1 Tab 2 2 Tab 2 3 Tab 2 4 Tab 2 5 Tab 3 1 Tab 3 2 Tab 3 3 Tab 3 4 Tab 3 5 Tab 3 6 Tab 3 7 Tab 3 8 Tab 3 9 Tab 3 10 Tab 3 11 Tab 3 12 Tab 3 13 Tab 3 14 Tab 3 15 Tab 3 16 Tab 5 1 GUIDE Vision Hearing and Manual Dexterity related k means Cluster Centres as a result of user Survey GUIDE Cognition related k means Cluster Centres as a result of USE SUIVEY nome A kemi Part of Virtual User Models as used in the VERITAS project Simulation results of the VERITAS project Focus of related projects r r e s Results according to Strang and Linnhoff Popien Audiometric descriptors and hearing loss according to
113. fil e d ing by User Model Ontology with pro filed User Models E Inference of Rec ommendations to Ontology with Rec meet User Model Profiles ommendations for User Model profiles tee Inference of Rec ommendations to Environments Ontology with Recommen dations for Environments eae Inference of Recom ns mendations to Tasks Ontology with Recom mendations for Tasks 22 Inference of Recommen dations to Components Ontology with Recommen dations for Components Resulting Virtual User Model Ontology Figure 3 5 The reasoning approach 3 4 3 Reasoning Approach As previously mentioned the VICON reasoning consists of several stages in the cre ation of the final Virtual User Model In the first stage based upon a user study Vicon Consortium 2010 User Model instances are added as members to impair ment groups To simplify the reasoning the user can add or change rules directly The syntax was reduced so that predicate values of instances are already assigned The user can 66 3 KNOWLEDGE MANAGEMENT use them without a definition For each step single files are applied so the vari able generation could be automatically produced if the name of the variable is the same as the attribute name defined in class properties For instance if the variable visualacuity is used the definition
114. form The parameters were used to create a User Model for simulation of impairments During the User Initialisation Application customers generate their specific User Model which is classified by the definition of the k means cluster as seen in the previous section Figure 2 10 shows the web interface of the GUIDE Project for user initialisation Different aspects of customer needs and preferences are defined based on selection and behaviour including visual hearing manual dexterity and cognitive capabilities of the end user The resulting User Model can be used as a specification for the definition of an optimized accessible user interface For instance a suitable color 1Trail Making Test see Reitan 1986 2Auditory verbal learning test see Ivnik et al 1990 3Wechsler Adult Intelligence Scale see Wechsler 1955 34 2 STATE OF THE ART configuration of buttons of a TV application are defined by the selection of the user as seen in figure 2 10 The User Model can be used by customers to personalize their own device but also as a simulation for application developers Figure 2 11 presents such a simula tion including a Social TV application without and with mild visual impairments Visual Preferences Button Colour Which one of the following buttons do you prefer row E Cognitive TMT Game Figure 2 10 User Initialisation Application of the GUIDE Project Source GUIDE Consortium al 35 2 STA
115. formalisms for knowledge representation in general This is reflected in the dialect of the rules interchange format RIF is still in development and only available as a draft version see also Kifer 2011 and Kifer 2008 of the W3C 2 4 Customer involvement in product development This chapter presents approaches for incorporation of human factors into the prod uct development process M A Kaulio Kaulio 1998 presented a review on selected methods of user involvement and compared seven different methods by the cate gories of customer involvement design for design with and design by Design for denotes approaches in which products are designed without a direct confrontation with customers Products are created by designers using data general theories and models of customer behaviour instead Design with focuses on a similar product design process as design for approaches but including a presentation of concepts and prototypes to customers Feedback is used in product design for adaptations of products to end user needs In the last category of design by customers are actively involved in product development and create products Using these categories the following customer involvement methods were compared e Quality function deployment QFD Quality function deployment was introduced by Yoji Akao in 1983 see Akao 2004 and describes an analytical approach for the first design phases with the involvement of end users It represents th
116. g11 g12 g15 816 g17 g20 Failed to correctly identify on button at first attempt il i2 i5 i6 17 i9 ilO 1183 il5 116 118 i19 g5 g10 g13 g14 g18 g19 Correctly interacted with the on button with out assistance i3 i4 113 115 116 i17 i19 gl g2 g4 g6 g7 g8 g9 g10 g11 g12 g13 g14 g15 g16 g17 gl9 g20 Mistook the green button for the on button i5 i9 i11 i13 116 g5 g10 g13 814 g19 Mistook the lock button for the on button il i2 i5 i8 i9 113 i15 i16 i18 Mistook the scroll up down buttons for the on button i6 i10 i11 i15 ton Mistook the flashlight button for the on but i2 i18 ton Mistook the flashlight itself for the on button i13 Mistook the volume buttons for the on but i17 Failed to correctly interact with on button without assistance il 12 i3 16 17 i8 i9 110 i11 114 118 120 g9 Correctly identified off button without assis tance il i2 i4 i5 i8 i10 113 114 i15 i16 il7 il9 i20 gl g2 g4 g5 g6 g7 g8 g9 g10 g11 g12 g13 g14 g15 g16 g17 g18 g19 g20 Correctly interacted with off button without assistance il i2 i4 i5 i8 il0 i13 il5 i16 i17 i19 i20 gl g2 g4 g5 g6 g7 g8 g9 g10 g11 g12 g13 g14 g15 g16 g17 g18 g19 g20 Failed to correctly identify off button without assistance i3 i6 i7 i9 111 i18 Difficult
117. he final clustering The following tables 2 1 and 2 2 presents the resulting classification into each level without non significant variables 32 2 STATE OF THE ART Vision Close vision level able to read 20 20 20 60 20 80 perfectly Distant vision level able to read 5 5 20 perfectly metres General eyesight good excellent normal Seeing at distance good poor poor Seeing at night normal poor poor Colour perception good bad bad Hearing Able to hear a sound of 500Hz Yes Yes No Able to hear a sound of 1Khz Yes Yes Yes Able to hear a sound of 2Khz Yes Yes Yes Able to hear a sound of 3Khz Yes Yes Yes Able to hear a sound of 4Khz Yes Yes No Able to hear a sound of 8Khz Yes No No How do you define your hearing excellent good poor Conversation from a noisy excellent normal normal background Movie dialogue only excellent good poor Ringing noises only excellent good normal Phone rings with a movie in excellent good poor background Manual Dexterity Mobility diagnosis none hernia none slipped disc Muscular weakness never A few Frequently occasions Write No difficulty No difficulty Mild difficulty Push a heavy door No difficulty No difficulty Mild difficulty Change a bulb No difficulty No difficulty Mild difficulty Use of transport bus etc No difficulty No difficulty Moderately difficult Tingling of limb No difficulty Mild difficulty Mild difficulty Weakness No difficulty Mild difficult
118. he Mock Up relates to holding issue of beneficiaries with a weak grip Since the material of the Doro Mock Up phone and so weight and surface material of the functioning phone was not available results 10 of 12 issues for Doro Mock Up show that recommendations were almost properly applied 5 3 5 Discussion This part of the evaluation also covered a comparison of output of the framework with the real prototype design Recommendation lists were compared to the Doro Mock Up phone Arcelik washing machine panel 1 and 2 by an expert see tables 5 13 5 21 5 22 5 23 5 24 and 5 25 if the recommendations were applied correctly or not The results 10 12 for Doro Mock Up 20 25 Arcelik washing machine panel l and 14 25 Arcelik washing machine panel 2 show that some recommendations were not properly applied This issue can also be identified in the side by side comparison results as seen in table 5 27 As already mentioned the Arcelik washing machine panel 2 is a prototype in which one single recommendation of the framework 1See focus group report of the VICON project Vicon Consortium 2013bJ 131 5 EVALUATION Recommendation Result Controls Good visual contrast between the keys and the appliance Pass Major controls Good visual contrast between the keys and the appliance Pass Room for Minor controls improvement Key tops should be convex or flat with a raised edge Pass Keys should be as large as
119. he parameters of the users environments etc i e to create your own profiles The CAD application had a lot of problems installing it I wish the application would be more available to different platforms OS X version would be also great But thank you to your work it was an eye opener Whenever there is risk that the information can be misguided it is best to highlight it well The tools are not able to replace the real user tests But if you say this is something that should point out the issues of a product that need to be tested with real users then the tool can be really useful 100 5 EVALUATION 5 1 5 Discussion In summary this evaluation issued that the concept of the system including an eval uation software for virtual prototypes does have a good basis but the amount of information needs to improve In the next section designers using the software in their product development review the system more specifically Regarding the an swers of the participants three main issues were identified e The system is as useful as data and recommendations provided e With the system it is possible to prevent big faults and big mistakes before prototyping e It does not necessarily lead to an acceleration but can also result in a deceler ation due to product customization to user needs 5 2 Improvement for Development Process This section is an expert evaluation with designers to obtain a value of acceptance
120. he result of the study of the two mobile phones Doro PhoneEasy 332 and Doro Mock Up was limited due to non functionality of the Mock Up Cognitive sensory and physical difficulties of use could be identified 5 15 The side by side comparison suggests that the Mock Up Phone has less accessi bility issues than the existing phone Especially some physical issues were no longer present in the Mock Up The tests revealed that for some customers there was not enough vertical spacing between front side buttons of the Doro PhoneEasy 332 On the Mock Up phone greater spacing between the buttons resulted in an increased overall size of the phone so some beneficiaries were unable to use the phone one handed as desired This suggests that an optimal spacing between buttons lies somewhere between the 332 and the Mock up There is a direct link between button size button shape and button spacing By changing one it is possible to eliminate problematic issues of others If buttons are too big and too spaced the product can be less comfortable to use To get an appro priate optimal setting additional beneficiary trials to compare modified interfaces or iterative testing with prototypes is necessary In the presented user tests problems with the 332 such as difficulties with the size of the buttons were solved with the Mock Up But new problematic issues were created as difficulties with the spacing of the buttons Similar links can also been found on o
121. he surface design functional is sues are not as relevant Additionally these drafts are highly subjective due to the influence by the designer s knowledge creativity and preferences which results in concentrated views on surface and form design that all functional aspects have to be adapted to Product Development Process Phases Outputs Z N Creation of Prototypes Final Prototype A Z x CAD Phase D l Virtual Prototype Draft Phase gt Product Drafts Product Planning Principle Solutions 4 lt Structure Analysis Functional Structure lt lt t gt gt Specification of Requirements l Requirement List Figure 2 1 Product Development Process according to VDI Gesellschaft Entwicklung Konstruktion Vertrieb 1993 Concept design or computer aided design phase describes the virtual construc tion of the previously designed drafts using computer aided technologies CAx like Siemens NX The product is specified including all parameters values and surfaces resulting in a specification which is used for the construction of first real proto types With respect to the first phase instead of a surface and form design this Berthold described these and other methodologies and definitions for product development process Berthold 2002 and compared VDI guidelines with other definitions resulting in the hypothesis It is not possible
122. his issue resolves the fact that there is a strong connection between output data and usability in existing product development processes It is very important to have an as brought expanse of recommendations as possible to cover all relevant aspects of inclusive design With the system it is possible to prevent conceptional and usability faults before prototyping Mostly in sketch design phase this issue infers additional informa tion about end user requirements which were or could not be covered in this early state The benefit to have an impact already in the first design phase pre vents design mistakes as early as possible before virtual or physical prototyping It does not necessarily lead to an acceleration but can also result in a decelera tion of the design process due to product customization to user needs Regard ing modifications of the target product with respect to end user requirements the fact to have customer information in an early stage must not necessarily 138 5 EVALUATION result in a acceleration but can also result in a deceleration of the product de velopment process The creative process of design to cover issues regarding accessibility can decrease the speed of the design process but can also lead to more inclusive products This issue is relevant especially when there are no suitable solutions to accessibility issues yet The second sub hypothesis 3 2 expands the evaluation to the practical use of the fram
123. ibes support for quality and richness of incoming data This issue is not relevant in cases of this thesis 6 Incompleteness and ambiguity inc This issue represents the importance of the feature to manipulate and use data even if it is incomplete Regarding the VICON project this issue is not impor tant due to the non existence of sensorial data Approach pv for app de qua inc Key Value Models a E Markup Scheme Models i z Graphical Models z Object Oriented Models Logic Based Models z Ontology Based Models Table 3 1 Results according to Strang and Linnhoff Popien Source Strang and Linnhoff Popien 2004 With respect to thesis related requirements pu for and app requirements are primarily important Table 3 1 see Strang and Linnhoff Popien 2004 presents a comparison between all different approaches including an appropriateness value for each of them In consequence Ontology based models are most suitable for the implementation of a knowledge base Regardung the theses presented in 1 6 Ontology based models would be suitable for the implementation of all requirements for elderly people In the next chapter and using Ontology based models a separation between initial 3 3 and inferred Ontology 3 4 4 will be presented including a reasoning step including an application of specific rule sets to the initial model 3
124. iently and comprehensive e Conformity with the expectations Is the structure of the software ok and does it conform with habits of designers e Suitability for learning Do the tools require pre knowledge 5 2 4 Results Personal not at all slightly moderate familiar very knowledge familiar familiar ly familiar familiar of participants about Design of 0 3 0 1 7 physical products Computer O 3 1 3 4 Aided Design Inclusive 0 1 3 6 1 Design Virtual 0 5 2 4 0 User Modelling VUM Table 5 2 Results of questions regarding personal knowledge of partici pants 106 5 EVALUATION General Questions Table 5 2 presents the results of the general questions about pre knowledge of each participant 8 participants describe themselves as familiar or very familiar about de sign of physical products 8 11 72 As mentioned in 5 2 3 mostly designers were questioned but also three researchers in the field of design participated The reason was to get a more resourceful perspective on the results from practical but also theoretical point of view on design and ergonomic factors The answers of CAD knowledge are due to the fact not all designers work with a virtual environment but rather with sketch drafts or other tools All participants are at least slightly familiar with CAD Both front ends were evaluated by each experience level of designers Although not all partici
125. iew of the final versions Finally I would like to thank all other colleagues from the Al research group for numerous discussions on this thesis Contents Page Contents ii List of Figures v List of Tables vii List of Theorems ix 1 Introduction 1 1 1 Background and Motivation ee eee 1 1 2 Targeted Impairments l s y s 0000 4 1 3 Inclusive Design ee a 6 1 4 Design Recommendations 00060 10 1 5 Research Questions w w w w w w w w k UW e w W weta 11 16 Hypotheses s ki bare Sa E a e Sw AS D w W U 12 1 7 Conchusiony s r o d e a ee Ae AG a Was Ww b 13 2 State of the Art 14 2 1 Current Product Development Process 14 2 2 Context Modelling a 16 2 2 1 Key Value Models 0 17 2 22 MarkupSchemeModels amp 17 2 2 3 GraphicalModels 6 18 2 2 4 Object Oriented Models 0 0000 18 2 2 5 Logic Based Models sua amra w wa w s 18 2 2 6 Ontology Based Models o e 18 2 3 Expert SyStems was ek 20 2 4 Customer involvement in product development 22 2 5 DigitaaHumanModels 60600 20 2 6 Related Projects 60 28 2 0 1 My l Project ve V X era e a a ei S 28 CONTENTS CONTENTS 2 6 2 GUIDE Project ee 31 2 6 3 VERITAS Project lt e us sn s s s sus muy DED aaa ae da 37 DOA VICON esya duy
126. ilar solutions which deal with the topic of context awareness in product development The concept chapter 3 and 4 represents an approach to the issue of in clusion regarding different end user scenarios focussing upon elderly and impaired beneficiaries which was also used in the VICON project Using the concept an evaluation chapter 5 was created with designers as test subjects In Chapter 6 the result of the concept and additional future plans are discussed lSee http www vicon project eu Chapter 1 Introduction 1 1 Background and Motivation The emergence of new embedded mobile technologies leads to a substantial growth of functionality in technical products In turn this growth in functionalities stimulates accessibility and economic issues These issues contain accessibility and ergonomic issues regarding the use of product interfaces as the result of an overload of func tions and capabilities For instance mobile phones are no longer just telecommunication devices but also a camera music player browser email interface etc becoming smart phones Devices with voice calling capability cellular connectivity and a screen size of at least 5 but less than 7 inches are now called phablets see Segan 2012 Smaller and more effi cient electronic components can be included into products resulting in continuously expanding functionality The fascination about new possibilities often obscures the fact that technol
127. impairments The output of the calculator is an exclusion value which defines how much of the population is excluded by a specific design based upon a selection of different tasks within a product see figure 1 8 For example if the user input defines the task which includes bending down to reach various distances below the waist to a level of kneeling down demand level 3 the output of the tool presents an overall exclu sion of 7 17 of the target population gender both minimum age 16 maximum age 102 Regardless this output defines a task related exclusion of a target population there is no direct connection to any product capabilities except by the tasks Fur thermore no recommendations are presented which should be considered if a prod uct is designed but rather tasks which should be avoided to include into a product Zitkus Langdon and Clarkson Zitkus et al 2011 compared various already existing tools to support design teams to explore the accessibility value of a product see chapter 2 5 Virtual techniques like DHM Digital Human Modelling see Duffy 2008 support the development process during virtual product design phase in which a target product is available in a virtual environment A virtual human is able to perform different tasks including the product but these tools mostly do not include impairments see a more detailed review in chapter 2 5 1 INTRODUCTION Inclusive design toolkit Toolk
128. ion Hypothesis 3 1 General Concept The concept of supporting designers during product development as a software framework is able to support the design of inclusive products For this hypothesis designers were interviewed Some designers were without any pre knowledge others had strong knowledge about ergonomic issues and require ments of elderly people the software usage and impact on product development pro cess The aim of these interviews was to get a better understanding of possibilities but also limitations from the designer s perspective 95 5 EVALUATION 5 1 2 Experimental Setup The user study is documented in detail in deliverable 4 3 of the VICON project see Vicon Consortium 2013a Four different statements were presented to participants that were already familiar with the software 5 1 3 Execution Methodology As this user study was part of the VICON project it includes the software package VIRTEX for evaluation of already existing virtual products see section 2 6 4 The result from this simulation presents different issues and recommendations as in the sketch design and CAD design phases Statements as listed below were presented to participants asking for agreement Sug gestions of improvements were collected with open questions The original study had nine participants four of them only using the evaluation soft ware tool VIRTEX not part of this thesis For each statement a Likert see Likert 19
129. ion presents one type of user requirement the system must be able to integrate into knowledge base Visual markings on the keys should be characters at least 4 mm high and should have good contrast with the colour of the key e g white characters on matt black keys Gill 1997 This recommendation presents one example of what type of information needs to be included Marked words define important contextual information which needs to be transferred to a nominal or textual form into the knowledge base It includes both qualitative and quantitative issues The definition that the characters should be at least 4 mm high is a quantitative recommendation including a minimum value On the contrary the issue regarding the good contrast describes a qualitative recom mendation with an abstract definition 16 2 STATE OF THE ART Accordingly the following issues contain all main requirements for the system described in this thesis see Strang and Linnhoff Popien 2004 and Baldauf et al 12007 1 Comprehensible human readable To maintain a modification ability the whole data structure should be human readable e g XML This would also ease the manipulation of the knowledge base 2 Dynamic Modifications It must be possible to change and modify objects and structures of the data storage with respect to variability of knowledge 3 Models A model based architecture is recommended e g User Model Task Model Environment M
130. iotis Moschonas Konstantinos Moustakas and Dimitrios Tzovaras Enforcing accessible design of products and services through simulated accessibility evaluation Tangible Information Technology for a Better Ageing Soci ety page 59 2010 Nikolaos Kaklanis Yehya Mohamad Matthias Peissner Pradipta Biswas Patrick Langdon and Dimitrios Tzovaras An interoperable and inclusive user modelling concept for simulation and adaptation In UMAP Workshops 2012a Nikolaos Kaklanis Panagiotis Moschonas Konstantinos Moustakas and Dimitrios Tzovaras Virtual user models for the elderly and disabled for automatic simulated accessibility and ergonomy evaluation of designs Universal Access in the Infor mation Society pages 1 23 2012b ISSN 1615 5289 doi 10 1007 s10209 012 0281 0 URL http dx doi org 10 1007 s10209 012 0281 0 147 REFERENCES REFERENCES Tapas Kanungo David M Mount Nathan S Netanyahu Christine D Piatko Ruth Silverman and Angela Y Wu An efficient k means clustering algorithm Analysis and implementation Pattern Analysis and Machine Intelligence IEEE Transactions on 24 7 881 892 2002 M A Kaulio Customer consumer and user involvement in product development A framework and a review of selected methods Total Quality Management 9 1 141 149 1998 Mohsen Kazemi Adhesive capsulitis a case report The Journal of the Canadian Chiropractic Association 44 3 169 2000 S Keates P J Clarkson L A Harrison
131. it home Exclusion Calculator What is inclusive design Why do inclusive design How to get started Population of Interest Sex Minimum Age Maximum Age Design process checklist ees vin O Business case materials M a None Set AAA Hearing ED A hare 7 PEO Cambridge simulation maos Remember messa Ya ge Cambridge simulation Thinking pal express rem Change glasses Impairment simulator None Set a sofware Dexterity Fe SS O Example set of personas Reach amp F None Set ETE Strech Q A L User capabilities None Set a Change colours Locomotion amp WA Change Default Y Fontsize A 5 MN Figure 1 7 The exclusion calculator of the inclusive design toolkit Source Clarkson 2003 Reset All Demands Calculate Exclusion Population of int Total 45 59 million adu Effects of Reducing Demands Table of Results Overall Exclusion Fe Exclusion based on each capability separately z 2200 GB adults thousands 9 5 of target population 4 33 million GB adults Note that because some people have multiple capability losses the numbers for the individual capabilities may not add up to the overall exclusion Figure 1 8 Exemplary output of the exclusion calculator of the inclusive design toolkit Source Clarkson 2003 1 INTRODUCTION 1 4 Design Recommendations Existing user studies and pro
132. jects contain information about problems and issues re garding different kinds of user interfaces These guidelines refer to a textual output of theories and data due to experience dealing with user needs of impaired customers of products During the product development process designers need to have an as good awareness about user needs when dealing with problematic issues regarding the usage of each designed product as possible To maintain this awareness the connection between the designers and beneficia ries of their products needs to be revised and optimised Referring to the product development process a supporting system will be used to present and adapt issues of these guidelines and additional personal experience The following example recom mendation presents e g one issue derived from the TIRESIAS website Visual markings on the keys should be characters at least 4 mm high and should have good contrast with the colour of the key e g white characters on matt black keys see Abbott 2007 This sentence defines already two very different recommendation with respect to re quirements and user needs 1 Visual markings on the keys should be characters at least 4 mm high This item defines an already specified minimum value for a font size of charac ters upon keys Thus it refers to a nominal value this issue type is defined as a quantitative recommendation 2 Good contrast with the colour of the key With respect to nominal
133. l g10 g17 5 38 reported that the keys themselves were too small i9 116 g4 g10 g13 12 38 did not think the buttons were i14 i15 i16 g4 g5 g6 g7 g8 s9 g10 13 spaced apart enough gl5 3 38 did not like the surface shape of the i11 i15 g12 keys reporting that fingers could not easily press the keys without sliding off 1 38 thought the number keys should be g4 raised more from the surface of the phone 1 38 reported that the keys were too big 111 13 19 did not recognise the A B and C speed dial buttons i2 i3 i6 I7 i8 i9 ilO 113 114 i16 117 il8 i19 11 19 did not recognise the lock button il i2 i5 i6 i8 114 i15 i16 i17 i18 i19 3 19 did not recognise the volume buttons i3 i4 i6 i7 2 19 did not recognise the select menu op i6 i9 tion buttons 1 19 thought the select menu option but i14 tons were too small Doro Mock Up phone developed using the software phone one handed as desired Issue Participants 2 19 beneficiaries found the phone too wide il i20 to comfortably hold 1 19 beneficiaries was unable to use the il 12 19 beneficiaries did not immediately recognise the on off button il i2 i4 i5 i6 i7 i8 i9 110 i13 i15 i18 One person who successfully identified the on button failed to identify the off but ton il9 7 19 people had to be prompted by the re sear
134. l you have all the documents you have no knowledge of Also if the designer is well knowledgeable he she can forget S The system would remind him her on inclusive design challenges For designers with no experience it will be even bigger help Yes if it could give more physical data for mechanical engineers H the model has a sufficient amount of parameters then yes I e all parame ters you need to depict disabilities IUs a tool that can help But the designer should not trust the software in any case Otherwise you ll get for ten years always the same stuff I think inclusive products have to be innovative And for innovation you need freedom The database is limiting Using VICON only as a support of the design process can work but relying only on the VICON environment can be limiting You could support but only to a very limited amount We believe that you need to meet the real users and you cannot do that in a machine environment Regarding the second statement the answers were diverse Some participants agreed the concept can help to involve user s perspective into development process earlier but also considered the issue that a complete replacement of user trials is not advantageous Another participant noted that designers should not solely rely on the virtual concept but rather see the system as a supporting tool set than a replace ment of user trials Comments to statement The VICON vir
135. l if necessary The first step in developing a knowledge base of a specific domain consists of the definition of all different concepts involved in a problem Figure 2 3 presents an example hierarchy in which all objects are derived from the most general root object Hence the taxonomy level of an object description defines how specific a concept is E g Main River is a River and River is an Object Object A River Human A rl Main Stream Tributary peter Hans Klaus Figure 2 3 KONWERK example hierarchy of concepts All objects also can consist of different parameters like the length of a river or the name of a person The following example from fig 2 3 represents a constraint for the oxygen saturation of a river with minimum value 21 1 2 3 4 2 STATE OF THE ART def conceptual constraint name oxygensaturation_of_rivers patterns riv name river formula riv oxygenmin lt 10 Figure 2 4 KONWERK example definition of a constraint In the first line of figure 2 4 the name of the constraint is given Lines 2 4 define various attributes of this constraint as a pair of keys and values The formula is using the attribute of the river ocygenmin as a variable representing if the variable of a river is less then 10 mg O2 1 The river is included in the set of outputs In summary expert systems like MYCIN represent a quite prominent approach of rule
136. l recommendations are included product interfaces can be used by a wider group of people 137 5 EVALUATION This issue was one main result of this part of the evaluation as recommenda tions were in line with recommendations given by the expert but also recom mendations stated by customers 5 4 Conclusion The aim of this thesis is to contribute in research by creating a solution for support ing product designers during the product development process The solution was implemented as a supportive framework including different tools for designers as seen in chapter 4 To evaluate the complete framework and especially the benefit of the framework hypothesis 3 was separated into three sub hypotheses Due to the impact of the framework on the product development process the first sub hypothesis 3 1 stated to support designers to create more inclusive designed products The eval uation was conducted as interviews about opinions and knowledge of designers if the concept was suitable for adaptation Hypothesis 3 Designer acceptance The involvement of context awareness for designers about impairments of prod uct beneficiaries into different phases of product development provides adequate flexibility and designer acceptance by requirement traceability due to the focus of each phase upon different scenario issues Three main issues were identified from this study e The system is as useful as data and recommendations provided T
137. l volume control which can be easily reached or a software interface for adjusting the volume There is variable amplification Audio connectors There is a standard audio connector Figure 1 5 The TIRESIAS project website containing guidelines for de signers Source Abbott 2007 1 INTRODUCTION Currently the concept of inclusive design is referred to from many existing use studies and guidelines for designers The TIRESIAS Abbott 2007 see figure 1 5 and CARDIAC Cardiac Consortium 2012 figure 1 6 projects collect these use stud ies into one website presenting information which technical features surfaces and issues must be adapted for different devices including mobile phones or remote con trols These existing guidelines are not directly included in the design process of devices or in existing tools used by designers with the consequence that most designers do not use them or even know about their existence CARDIAC EU ORG Home Remote controls About us Aremote control is an electronic device used for the remote operation of a machine Due to the majority of modern devices being controlled by this kind of device and the amount of function controls found on most modern remote controls blind and Advisory Group area partially sighted people and those with other disabilities often encounter difficulties with remote controls that render them inaccessible Consortium area PER Abo
138. larkson et al 2003 Dong et al 2005 Goodman et al 2006a Goodman et al 2006b Dong et al 2004 and Cassim and Dong 2007 11 1 INTRODUCTION 1 6 Hypotheses Based upon the research questions the following hypotheses can be derived Hypothesis 1 Ontology based model application Ontology based models can be used to store and manipulate various data con cerning requirements especially of elderly people for the use of products This thesis also refers to the extraction and description of already available context information and requirements into one single knowledge management solution It must be possible to include all kind of data involved in the performance of a task by elderly This will be addressed in chapter 3 Hypothesis 2 Suitable Reasoning Ontology based models can be used to give statements from knowledge base for specified scenarios described by the questions of who is using a product where to perform what task From the context information as presented in the first hypothesis statements must be inferred so designers get only relevant information for specific scenarios This refers to a general verification of all software related terms see chapter 4 Hypothesis 3 Designer acceptance The involvement of context awareness for designers about impairments of prod uct beneficiaries into different phases of product development provides adequate flexibility and designer acceptance by require
139. ld be robust and not require Pass extending by the user Result 10 12 Table 5 13 Recommendation Checklist of Doro Mock Up Phone 122 5 EVALUATION Task Doro Doro Mock Up PhoneEasyp332 Phone Identify on button 20 38 7 19 Successfully dial a number 38 38 19 19 Press green button to 37 38 18 19 connect call Identify that a call is coming 38 38 N A in Press green button to 38 38 N A receive call Identify the message button 7 17 8 16 SMS Open and read an incoming 25 27 16 16 text message Total 83 4 N A Table 5 14 Results of user study regarding mobile phones Difficulty encountered Doro Doro Mock Up PhoneEasypR332 Cognitive features Difficulty recognizing SMS Yes Yes but envelope button icon was easier than SMS in Ireland Sensory features Difficulty identifying on Yes Yes button Difficulty identifying off Yes Yes button Difficulty reading letters or Yes No numbers Physical features Difficulty with the size of the Yes No buttons height was too narrow Difficulty with the spacing of No Yes the buttons too spaced Difficulty with the spacing of Yes No the buttons not enough space vertically Difficulty with the shape of Yes No the buttons Total number of issues 7 4 Table 5 15 Result comparison regarding mobile phones 123 5 EVALUATION T
140. leni Chalkia Evangelos Bekiaris Karel Van Isacker Serge Boverie Onorino Di Tanna Nikos Partarakis Kostas Moustakas Hans Joachim Wirsching Maria Fer nanda Cabrera Elena Tamburini and Mytas Nicolas Accessible and assistive ict veritas deliverable 1 7 1 final use cases and application scenarios 2010 S Ciccantelli and J Magidson From experience consumer idealized design involv ing consumers in the product development process Journal of Product Innovation Management 10 4 341 347 1993 J Clarkson Inclusive design Design for the whole population Springer Verlag 2003 J Clarkson R Coleman S Keates and C Lebbon A designer centred approach Inclusive design Design for the whole population 2003 R Coleman and C Lebbon Inclusive design Helen Hamlyn Research Centre Royal College of Art 2005 R Davis B Buchanan and E Shortliffe Production rules as a representation for a knowledge based consultation program Artificial intelligence 8 1 15 45 1977 H Dong S Keates and P Clarkson Inclusive design in industry barriers drivers and the business case User Centered Interaction Paradigms for Universal Access in the Information Society pages 305 319 2004 H Dong P J Clarkson J Cassim and S Keates Critical user forums an effective user research method for inclusive design The Design Journal 8 2 49 59 2005 Vincent G Duffy Handbook of Digital Human Modeling Research for Applied Er g
141. ls Erweiterung des Kon struktionssystems WISKON PhD thesis Kassel University 2002 Pradipta Biswas Pat Langdon Christoph Jung Pascal Hamisu Carlos Duarte and Luis Almeida Developing intelligent user interfaces for e accessibility and e inclusion In Proceedings of the 2012 ACM international conference on Intelligent User Interfaces IUI 12 pages 405 408 New York NY USA 2012 ACM ISBN 978 1 4503 1048 2 doi 10 1145 2166966 2167060 URL http doi acm org 10 1145 2166966 2167060 BMW AG BMW Website BMW Techniklexikon Controller http ww bmw com com de insights technology technology_guide articles controller html content_type com de insights technology technology_guide articles control_display html source com de insights technology technology guide articles idrive html amp article controller Accessed 2013 03 04 Cardiac Consortium Advancing research development in the area of accessible and assistive ict http www cardiac eu org 2012 J Cassim and H Dong Empowering designers and users Case studies from the dba inclusive design challenge Design for inclusivity a practical guide to accessible innovative and user centred design page 89 2007 144 REFERENCES REFERENCES E Castillo and E Alvarez Expert systems uncertainty and learning WIT Press 1991 E Castillo J M Guti rrez and A S Hadi Expert systems and probabilistic network models Springer Verlag 1997 E
142. lso contain the model file which was used in the CAD software see CADFilename tag in Figure 4 3 Additionally to the Manifest xml from the second to the third phase a meta xml file is included in the VSF The meta xml file provides all meta information regarding the model and prod uct components see figure 4 4 Each object in the virtual environment is added including the annotation type of a component the current id and the file name of the CAD model which is also included in the VSF container Physical data as local positioning of components are automatically included VSF containers including the three files Manifest xml meta xml and the corresponding model file are used for the transition from second to third phase 4 2 Tools This section describes all front end applications included in the development of a product Both tools support designers by providing recommendations for the target product 4 2 1 Sketch Design Tool Overview The Sketch Design tool see figure 4 5 will support the first design step phase 1 sketch design phase The system uses a choice of a User Model Persona an envi ronment and a task The output of the application is a number of textual recommendations and at tached files e g specific templates for graphic design software The Sketch Design tool is distributed as an applicable JAR file included in the installation setup To start the application it is necessar
143. mall It should be noted however that these buttons were not included in the tasks above i5 i9 110 113 114 i15 116 117 118 gl g4 g5 87 gl0 g13 g14 g15 g16 g19 t2 13 t4 t6 t7 t8 Nine beneficiaries reported that there were too many buttons on the remote control ill i14 i15 i16 gl g2 g13 g18 g19 178 APPENDIX Small Black Remote Control Issue Participants 45 47 beneficiaries successfully identified the on off button il i2 i4 i5 17 i8 i9 110 111 112 113 114 115 i16 i17 i18 i20 gl g2 g4 g5 g6 g7 g8 g9 gl0 g11 812 g13 g14 g15 816 17 g18 g19 g20 tl t2 t3 t4 t5 t6 t7 t8 One successfully identified the off button only i6 Two failed to correctly identify either on or off i3 il9 45 47 expected the button to be at the top of the remote control il i2 i4 i5 i6 i7 i8 i9 110 i11 i12 113 il4 il5 il6 il7 i18 i20 gl g2 g4 g5 g6 g7 g8 g9 g10 g11 g12 g13 g14 g15 g16 g17 g18 g19 g20 t1 t2 t3 t4 t5 t6 t7 t8 All 47 beneficiaries were able to physically press the on off button i2 i3 i4 i5 i6 i7 i8 i9 i10 ill i12 113 114 i15 i16 117 i18 119 i20 gl g2 g4 g5 g6 87 g8 g9 g10 g11 g12 g13 g14 g15 g16 g17 g18 g19 g20 tl t2 t3 t4 t5 t6 t7 t8 Two didn t like the feel of the button press il5 t7 45 47 beneficiaries su
144. mation Apparent simplicity real complexity Au tomation and human performance Theory and applications pages 3 17 1996 KM Zackowski AW Dromerick SA Sahrmann WT Thach and AJ Bastian How do strength sensation spasticity and joint individuation relate to the reaching deficits of people with chronic hemiparesis Brain 127 5 1035 1046 2004 W Zhou T Armstrong M Reed S Hoffman and D Wegner Simulating complex automotive assembly tasks using the humosim framework SAE Technical Paper pages 01 2279 2009 153 REFERENCES REFERENCES E Zitkus P Langdon and J Clarkson Accessibility evaluation Assistive tools for design activity in product development In Proceedings on the 1st international conference on sustainable intelligent manufacturing pages 659 670 2011 154 Appendix Graphs of Designer Tests General Questions participants Yy not at all slightly moderately familiar very familiar familiar familiar familiar Figure 1 Personal knowledge of participants about design of physical products 155 APPENDIX 10 participants 0 4 I l not at all slightly moderately familiar very familiar familiar familiar familiar Figure 2 Personal knowledge of participants about Computer aided De sign 10 8 2 E B 3 E A 4k 2 0 T I l l not at all slishtly moderately familiar very familiar familiar familiar familiar Figure
145. ment By selecting a user profile environ ment and task recommendations are presented to the user The user is also able to export the current set of recommendations to a PDF file for later analysis and print ing purposes Also the selections can be exported into a VSF which can be imported in the CAD phase for the presentation of recommendations from the sketch design phase 4 2 2 Administration Tool Overview ConVic G File Edit Import Export Look amp Feel po Recommendations Knowledge Base Treeview Addnew Editltem DeleteValue Edit UserModel Rules Edit Recommendation Rules Edit Environment Rules Edit Task Rules Refresh from Local File Refresh from Database Predicate Tool VICON UserModel Component IDName Hearing2kHz Glare Nickname HearingikHz LightLevels Glasses FieldOfVision Name Buttons BackgroundNois nation VProfile1_ o 2 o 2 Y VProfile1 o 0 once 3 30 2 25 2 Y 2 Mark 2 100 P1 15 1 5 1 N j1 Trevor 1 200 HProfile1_ 30 0 25 0 10 HProfile1 0 100 PS 65 3 145 3 Y 3 Gandalf 13 0 VProfile2_ 0 3 lo 3 Y 3 vProfilez_ 0 0 MDProfile2 0 0 0 0 10 MDProfile2 3 0 HProfile2 65 0 145 0 10 HProfile2 0 0 P4 65 3 as 3 Y 3 Dorothy 1 0 P2 30 1 125 1 N u Eileen 2 100 MDProfile1 0 0 0 0 lo MDProfile1 2 0 Found 87 Recommendations Figure 4 7 Knowledge base interface of the Administration software 83 4 APPLICATION IN DEVELOPMENT PROCESS The Administratio
146. ment traceability due to the focus of each phase upon different scenario issues To obtain acceptance by designers a high degree of usability is a mandatory factor for the implementation and realization of the system If designers cannot adequately use the software included in their typical software environment the approach would not support the user but rather hinder instead The verification of this hypothesis can be separated into different issues which will be analysed and discussed in chapter 5 12 1 INTRODUCTION 1 7 Conclusion This chapter introduced the need of to create a solution for supporting inclusive design during the product development process With technological advances new functionality issues appeared regarding a balance between human capabilities and possible interactions when using a device New devices were introduced e g smart phones which allow users to perform a higher amount of functionalities resulting in a higher complexity as seen in section 1 1 This complexity can most often be reflected as a burden for the users regarding the necessity to learn how to use such a product As seen in section 1 2 demographic changes must also be considered in the de sign process as there is an increasing number of elderly users of these devices Inclu sive design describes a concept how to deal the capabilities of beneficiaries to provide interfaces which can be theoretically used by everybody Guidelines and standa
147. mposium on Engineering interactive computing systems EICS 12 pages 81 90 New York NY USA 2012 ACM ISBN 978 1 4503 1168 7 doi 10 1145 2305484 2305500 URL http doi acm org 10 1145 2305484 2309300 Juan Carlos Pe a Guevara Edmundo Berumen Nafarrete Arturo Aguirre Madrid Jorge Vallejo Ponce Ivanovish De la Riva Mu oz and Juan A N ez Valdez 150 REFERENCES REFERENCES Anatomically designed shoulder hemiarthroplasty made after 3 d models and im planted in a child with rheumatoid arthritis a case report Acta Ortop dica Mexi cana 19 1 S51 S55 2005 U Persad P Langdon and J Clarkson Characterising user capabilities to support inclusive design evaluation Universal Access in the Information Society 6 2 119 135 2007 B J Pine and S Davis Mass customization the new frontier in business competition Harvard Business School Pr 1999 Emilie Poirson and Matthieu Delangle Comparative analysis of human modeling tools 2013 E Prud Hommeaux A Seaborne et al Sparql query language for rdf W3C recom mendation 15 2008 Ralph M Reitan Trail Making Test Manual for administration and scoring Reitan Neuropsychology Laboratory 1986 RNID The Royal National Institute for Deaf People VICON Task 2 1 An overview 2010 Presentation in Workshop Meeting London November 2010 E Rosenblad Wallin User oriented product development applied to functional cloth ing design Applied Ergonomics 16
148. n Tool ConVic is distributed as an applicable JAR file and is included in the installation setup To start the application it is necessary to run the JAR file with Java installed using the Start menu of Windows It also includes the Socket Server and requires the ports 65000 to 65002 to be open ConVic consists of three parts which can be chosen by a tab panel The Recom mendations tab presents the Sketch Design View on the Ontology see next section After all changes of the Ontology a restart is required File Restart to update this view To change the Ontology e g ifyou want to add new Recommendations the Knowl edge Base presents an interface to all classes which are used to build the reasoning part see figure 4 7 The administrator can select the Ontology class on the left and modify the Ontology class on the right Each class is presented as a table including all instances and predicates It also contains the following functionality e Rule editor to change all rule sets e Predicate tool to change variables and attributes of a class e Repository interface to connect with a MySQL server for version support of the Ontology and rules e Add edit and delete instances of all classes e Import of all different class instances from Comma Separated Values Excel CSV File e Export of the Ontology File OWL after reasoning The TreeView last tab visualizes the Ontology after reasoning in a tree based de sign The right si
149. n on the right side of the panel with a IEC 5010 power symbol e Only colored button is Start Pause Cancel button e Program selection as rotary Arcelik knob in the middle with 16 Arcelik settings Washing Machine e LED information panel Panel 2 between program knob and detergent dispenser e and buttons with small gap between them e Width 590mm height 125mm at shortest point and 150mm at longest point Table 5 11 Existing and emerged washing machines used for evaluation of customer satisfaction 118 5 EVALUATION TV Remotes figure 5 10 Identify on button Typically the on button on TV Remotes is presented as a red or red labelled IEC 5010 power symbol On all evaluated TV Remotes it is placed similarly on the top right of the remote Press on button Participants were asked in this task to press the identified button Identify the volume button On the Grundig large silver remote control the volume buttons are located on the lower half of the remote on the left side presented including a triangle On o both other remote controls and buttons are used Press Volume up down key The Grundig large silver has top bottom alignment of the buttons Both other devices a left right alignment Identify the channel up button On all remote controls this button is located nearby the volume buttons Press the channel up
150. n product development process e An improvement regarding more comprehensive scenario information is advan tageous Designers do not fully comprehend the scenario of their selection and how it is processed A further development as a user manual including background in formation or an extension of meta information would allow a more sophisticated scenario selection by designers 112 5 EVALUATION 5 3 Customer Satisfaction 5 3 1 Thesis and Prediction Hypothesis 3 3 Customer Satisfaction End products which are created using the framework can be used by a wider range of customers This hypothesis was evaluated with end customers beneficiaries of the prod ucts which are elderly and impaired people With respect to the framework this evaluation mostly deals with content and output of the system itself but not func tionality Different products were evaluated which were created with and withouth the framework to review the discrepancy between answers of beneficiaries Tests with products and mild to moderate impaired people were conducted in which participants evaluated end design issues by themselves Products which were created without the framework should result in more problematic issues by the end users than products created with the framework As an optimal result products created using the framework would have no accessibility issues at all In addition to user trials an expert evaluation by an accessibility
151. nd moderate visual impairment profiles HI1 and HI2 for mild and moderate hearing impairments MD1 and MD2 for mild and moderate hearing impairments The level of no impairment can be defined an empty String EnvRule String The rule with Jena inference syntax if a recommendation should be presented related to environment selection of the user i e le lighting_level 2 TaskRule String The rule with Jena inference syntax related to task selection of the user ifa recommendation should be presented Component String A component name of recommendation directly related to a specific component e g Button ComponentRule String The rule with Jena inference syntax if a recommendation should be presented related to component functionalities and attributes Phase Integer 1 Sketch 2 CAD 3 Evaluation Application Phase definition when a recommendation should be presented Table 3 13 Ontology class data properties used for Recommendation 63 3 KNOWLEDGE MANAGEMENT Each recommendation represents one specific suggestion for designers how a prod uct can be developed for a specific target user group By using the EnvRule or TaskRule different rules can be defined using the Jena inference syntax by which input selection of the designer a specific task should be presented The Phase attribute refers to the specific product development phas
152. ng their impairments and abilities For each impairment of hearing visual and man ual dexterity a classification into three profile groups was created using different preferences and parameter descriptions using nominal or categorical values of the subjects The next chapter 3 3 will describe the attributes extracted from the user study for the creation of different Personas which were used for the inference and presentation of specific recommendations related to selections of designers of target User Models According to the definition of the reasoning process in every step rules are used to define and classify different instances as members of different classes e g User Model profiling Also typical scenario settings to perform tasks using the product were extracted from the user study and will be described in chapter 3 4 3 3 Virtual User Model The main concept of Virtual User Models VUM is the representation of all scenario related issues in a knowledge base In relation to this functionality an Ontology was used to define classes and instances including a hierarchical taxonomy In order to provide quantitative and qualitative recommendations see 1 4 as an output of user specified parameters various reasoning steps using forward chain logic were implemented 51 3 KNOWLEDGE MANAGEMENT 3 3 1 Knowledge Base The knowledge base of the Virtual User Model was implemented using ontologies for each sub model These models
153. ns Adaptable interface The Jena Ontology framework offers a sophisticated Ontology interface with the advantage to manipulate all resources predicates and values directly from within Java With respect to the requirement of a server client architecture the server implemented in Java is able to perform manipulations of all Ontology instances Inference support Jena contains a reasoning engine which is able to operate with different sets of ontologies RDF S OWL lite OWL full Also a very generic reasoner is in cluded which can also be manually extended by build in rules 52 3 KNOWLEDGE MANAGEMENT 3 3 2 User Model The User Model represents the mass customization class of target users Each user contains parameters and references to specific impairments described either as nominal or abstract values The used attributes were defined as an output of the user study General characteristics Predicate Datatype Description Name String The name to identify a person is the only one primary predicate Mandatory to define it in an instance IDName String The IDname is unique for each object of the Ontology class E g each user profile has a unique IDName assigned Description String Description of the user profile or Persona represented by the profile Nickname String Optional nickname for the person VirtualModel String An URI Uniform Resource Identifier see Masinter et al 2005 where to
154. nstances e g parameter UserModelAge with an integer value as seen in tables 3 5 3 8 The complexity of the final resulting model after the inference of rules is defined by ALCH D 70 3 KNOWLEDGE MANAGEMENT 3 4 6 Multiple Selection In the final application the designer can select multiple User Models environments and tasks resulting in a set of recommendations In the initial set of recommenda tions each one refers to one User Model impairment profile typical environment or task setting With respect to section 1 3 and 3 4 4 each selection of the designer results in a specific set of recommendations For instance if the designer selects the User Model Gandalf the presented recommendations are members of each impair ment profile class which the User Model Gandalf is classified to Each presented recommendation has an impact on a specific impairment profile e g class of mod erate hearing impaired A combined set is created containing all recommendations for each impairment group Ifthe user selects more than one User Model the recom mendations for all must be merged An intersection of the different sets would result in an empty set due to the connec tion of each recommendation to different aspects 3 5 Conclusion This chapter presented a knowledge modelling approach to include relevant data into a knowledge base Based on chapter 2 context modelling structures were compared with respect to different requi
155. ntify how to open 19 20 37 38 36 38 compartment Total 92 66 94 45 92 6 Table 5 16 Results of user study regarding tv remotes TV remotes do not have any emerged user interfaces to be compared to so the results can not be included as a side by side comparison The results of the user trials show only a very small difference in accessibility issues Grundig large silver 92 66 Grundig large black 94 45 Grundig small black 92 6 see 5 16 with already very high values Most of the issues when performing a task deal with problems to identify the volume or channel up button that were implemented on the remote controls with different icons For instance the increase volume button on the Grundig large black remote control is realised as a button labelled Similar as the result of the mobile phone user trial this leads to the suggestion to use universal labelling on buttons if possible 126 5 EVALUATION Washing Machines This section is structured into five parts General comments from expert evaluation of both washing machine panels recommendation checklist by expert results of the user study with beneficiaries panel comparison by beneficiaries and panel compar ison by expert The Arcelik Washing Machine Panels were not available on time so an expert accessibility evaluation was conducted on them Positive Negative Clear typeface Good colour contrast between butt
156. o test single tasks using an avatar of the beneficiary and the product see Matiouk et al 2013 Figure 2 12 presents the simulation input of VIRTEX in cluding the import of a VSF file which is used to store all input data from the first 2 phases left see also section 4 1 4 and the selection of User Profile and Environment right picture VIRTEX VIRTEX ne diia estes gt VSF file path VirtexVAssets a xml VSF file patt sktop _VICON VICON_Development Virtex User Profile Environment Load File Dorothy Bathroom Eileen Living Room Gandalf Kitchen Mark Trevor hofer Uf Figure 2 12 Simulation input of VIRTEX Source Vicon Consortium 2012b The VICON project product development life cycle consists of 3 phases Sketch de sign phase CAD phase and Evaluation phase During sketch design phase designers are using the software tool of this thesis see chapter 4 2 1 to create first product drafts In CAD phase the integrated module in the CAD software Siemens NX is used see 4 2 3 to get recommendations during the creation of a virtual prototype of the product The third phase deals with additional tests and simulations including impairments of beneficiaries In the first step of the evaluation designers need to select a user profile with an already included predefined virtual humanoid and the 40 2 STATE OF THE ART Subtasks Identify a button button_0 identity a button button 1 ed iden
157. o the sample size is reduced to 46 gl4 All beneficiaries who attempted the task 46 successfully identified the location of com partment il i2 i3 i4 i5 16 17 i8 i9 ilO ill 112 113 114 115 116 i17 118 i19 i20 gl g2 g4 g5 g6 87 g8 g9 gl0 g11 g12 813 g15 g16 g17 g18 g19 g20 tl t2 t3 t4 t5 t6 t7 t8 Continued on next page 177 APPENDIX Issue Participants 43 46 successfully identified how to open the compartment il i2 i3 i4 i5 16 17 i8 i9 ilO ill 112 113 114 115 il6 117 i19 i20 gl g2 g4 g5 g6 87 g8 g9 g10 g11 g12 g13 g15 g16 g17 g18 g19 g20 tl t2 t3 t4 t5 t8 Three had difficulty when opening the com partment il i5 t7 45 46 successfully identified how to close the compartment il i2 i3 i4 i5 i6 i7 i8 i9 i10 ill i12 113 i14 115 i16 i17 il9 i20 gl g2 g4 g5 g6 87 g8 g9 g10 g11 g12 g13 g15 g16 g17 g18 g19 g20 t1 t2 t3 t4 t5 t6 t7 t8 However 21 46 had difficulty when closing the compartment specifically with lining up the compartment cover to slide it into place il i2 i3 i6 i9 il0 i11 i12 116 117 i19 i20 g2 g4 85 87 g8 812 g13 tl t8 Two beneficiaries failed to successfully open or close the battery compartment il8 t7 25 47 beneficiaries reported that the smaller buttons on the remote control were too s
158. odel to separate objectives for each knowledge part 2 2 1 Key Value Models Key Value models define the most simple data structure for context modelling see Strang and Linnhoff Popien 2004 and Baldauf et al 2007 The main idea is to add information as a pair of information connecting one keyword with another word or nominal value Formally key value models can be defined as a set of KV Ki Vi The main advantage and disadvantage by using key value models is the unique bind ing of each key K to exactly one value V The models derived from this structure are also not able to describe relations and functions between keys directly result ing that these models would not be suitable for a representative structure in case of describing issues and recommendations as mentioned above 2 2 2 Markup Scheme Models Markup Scheme models mainly concentrate upon the representation of hierarchies upon profiles In this context especially three approaches are mentioned Compos ite Capabilities Preference Profile CC PP Kiss 2006 Comprehensive Structured Context Profiles CSCP see Held et al 2002 and User Agent Profile Forum 2001 Each describes subjects e g users components as profiles including categorical and nominal values as a Resource Description Language RDF see Lassila et al 1998 based meta language 17 2 STATE OF THE ART 2 2 3 Graphical Models Context can also be described as graphical profile
159. of applications do O 0 2 6 not have an unpredictable processing time e g start of application Table 5 5 Results of questions regarding conformity with user expecta tions The structure of the software was mostly approved by participants as seen in table 5 5 The layout was similar to expectations of designers This leads to the assumption of a positive conformity with user expectations The last statement that some features of applications do not have an unpredictable processing time was slightly approved Some participants skipped this question because they could not find any features with unpredictable processing time so they were unclear about the result Eight participants found some For instance the start of the application takes some time which is unclear from designers perspective As previously mentioned in chapter 4 during the start the reasoning is performed from an initial Ontology inferring a new resulting model which is used in the application In the current state the model is inferred dynamically by every start of the application regarding further implementations and extensions of the model itself Suitability for learning Statement skip The software requires little time O 1 5 4 1 0 to learn The software is easy to learn 1 1 2 7 0 0 without prior knowledge help or manual The software is easy to use even O 3 2 6 0
160. ogy Based Models The term Ontology originally comes from the field of philosophy meaning the study of existence Ontology based models are used in various approaches like the VUMS cluster projects VERITAS Chalkia et al 2010 VICON Kirisci et al 201 1b Kirisci et al 201 1a GUIDE Hamisu et al 2011 and MyUI Peissner et al 2011 Wang et al presented an Ontology based context model which is feasible and also includes reasoning schemas Wang et al 2004 18 2 STATE OF THE ART ORM Entity Type ORM Entity Type n ary w Fact Type is of type Device id Device Type code static fact type profiled fact type Location name sensed fact type Location name derived fact type located near Device id engaged in Person name Activity name temporal fact type Activity name 38 Person name 2 e Location name located at Figure 2 2 Contextual Extended ORM Source Strang and Linnhoff Popien 2004 19 2 STATE OF THE ART Staab and Studer Staab and Studer 2009 presented a sophisticated definition about Ontology techniques and applications Ontology in general can be formally described by O C R A where C can be defined as the context R as relations and 4 as axioms 2 3 Expert Systems One main topic of artificial intelligence Al addresses the question of how to defin
161. ogy can also create new burdens and complexity to end users see Woods 1996 Especially a merge of different functionalities into single devices can be very inefficient regarding acceptance and usability by the end users This development can be seen as a two edged sword on the one side new function ality and features increase the product value on the other side all new functionality and features must be included into a recognizable product mostly resulting in a re definition of the product Also new interface components are often used to include more functionality on a small space For instance operating the BMW iDrive inter face see figure 1 1 the user is able to control different tasks like navigation radio or phone The interface consists of a touch pad on the surface of a rotary switch which can be turned to specify a selection of the user or the user can draw on the touch 1 INTRODUCTION Figure 1 1 BMW iDrive controller and display Source BMW AG pad to perform more sophisticated tasks like browsing the internet Interfaces like the iDrive are capable realizing multiple functionalities but the user needs to adapt and learn how to operate it properly figure 1 1 This issue will most likely result in a lower acceptance by especially elderly people or people who often do not have the physical capabilities to interact with such as system Human factors like the definition of motor capabilities are used to value these exact
162. oking appliance 8 Other household appliance 9 None Vision LightingLevel Integer O Poor Estimate of lighting level in room in 1 Medium 2 which user trial took place Bright LightingType Integer 1 Estimate of type of lighting in room in Natural lighting 2 which user trial took place Artificial lighting DirectLights Integer 0 No 1 Existence of direct lights in the Yes environment direct lights and glossy surfaces are related to glare Manual dexterity Temperature Integer 0 Cool Estimate of temperature level in room 1 Comfortable 3 in which user trial took place Warm Continued on next page 59 3 KNOWLEDGE MANAGEMENT WMClearSpace Integer 1 999 Amount of clear space in cm in front Front of the washing machine WMClearSpace Integer 1 999 Amount of clear space in cm at the Left left of the washing machine WMClearSpace Integer 1 999 Amount of clear space in cm at the Right right of the washing machine Table 3 10 Ontology class data properties used for Environment Environment instances are created using outcomes of the user study see section 3 2 Each environment refers to a different surrounding of the user in his or her daily life Using these abstract representation the system is able to recommend design guides based upon the specific surroundings All environment related recom mendations are connected to environment instances directiy by a spe
163. on 7 18 22 Manual Dexterity 27 14 6 Table 5 8 Participants of the user study by impairment levels 5 3 3 Execution Methodology According to the hypothesis this evaluation focuses upon end users of customer products A study with people over 65 years of age who have mild to moderate hearing vision and or manual dexterity impairments was executed in which the participants should access different products and perform predefined tasks Two different categories of products were evaluated 1 Existing User Interfaces This category involves the evaluation of different products without the use and application of the framework Resulting issues should be similar to recommen dations implemented in the framework 2 Emerged User Interfaces This part of the study focuses upon the use of products which were created with focus to inclusive design and usability by elderly people The result is compared to results of the first category After both evaluations issues and problems regarding the usability and accessibility were collected using interviews with participants Examined Products Regarding existing user interfaces the following tables 5 9 5 10 and 5 11 present the used products Both industrial partners of the VICON project provided emerged products with focus lLarger images in appendix 6 2 114 5 EVALUATION to issues of elderly customers Regarding washing machine panels Arcelik panel 1 was created fo
164. on and Thimbleby directed the concept even more to demographic terms The field of inclusive design relates the capabilities of the population to the design of products by better characterising the user product relationship Inclusion refers to the quantitative relationship between the demand made by design features and the capability ranges of users who may be excluded from use of the product because of those features Langdon and Thimbleby 2010 Various definitions of this concept are available e g Clarkson et al Clarkson et al 2003 Persad et al Persad et al 2007 Keates et al Keates et al 2000 or Coleman and Lebbon Coleman and Lebbon 2005 but all refer to the same concept to adapt demographic changes of our society into the product development process 29 tiresias org Search Coogle Making ICT accessible Sitemap EJ Home Aboutus ICT accessibility Research resources Project tools News and events HERE Checklist for Audio Output Researchers Research projects Researchers publications ICT devices ICT standards Design guidelines Scientific technological reports Serial publications gt Sources of funding Selecting partners for research projects Links Add update information No significant benefits O Vinor benefits Major benefits Volume Background noise is minimised There is a physica
165. ongly Colour Integer 0 No Colour indicates if the user is colour 1 Yes blind NearFocus Integer 0 No Ability to clearly focus on objects at 1 Slightly 2 near distance can be measured as Moderately 3 Amplitude of Accommodation in Strongly centimetres DepthPerception Integer 0 Ability to judge distance Normal 1 Mild 2 Moderate ContrastSensitivity Integer 0 Pelli Robson Score as a measure of Normal 1 contrast sensitivity value of normal Mild 2 1 6 2 mild 1 1 1 5 moderate Moderate 1 1 1 5 Glare Integer 0 No Glare indicates if the user is sensitive 1 Yes to light and glare Glasses Integer 0 No Indication if the user has glasses or 1 Yes contact lenses GlassesWithProd Integer 0 No Will the user wear glasses or contact uct 1 Yes lenses when using a product Table 3 7 Ontology class data properties used for User Model Vision 55 3 KNOWLEDGE MANAGEMENT Manual dexterity Predicate Datatype Description Arthritis Integer 0 No Answer to the question Did the user 1 Yes report Arthritis Grip Integer 0 No Grip describes difficulty by holding 1 Slightly 2 small items for example a pen or the Moderately 3 handle of a cup or items made of Strongly slippery material Buttons Integer 0 No Buttons estimates difficulty when using 1 Slightly 2 buttons or keys for example when Moderately 3
166. onomics and Human Factors Engineering CRC Press Inc Boca Raton FL USA 1st edition 2008 ISBN 0805856463 9780805856460 Inge E Eriks Hoogland Sonja de Groot Marcel WM Post and Lucas HV van der Woude Passive shoulder range of motion impairment in spinal cord injury during and one year after rehabilitation Journal of Rehabilitation Medicine 41 6 438 444 2009 European Commission Commission communication the demographic future of europe from challenge to opportunity In Europe in figures Eurostat year book 2011 2011 URL http eur lex europa eu LexUriServ LexUriServ do uri C0M 2006 0571 FIN EN PDF 145 REFERENCES REFERENCES M R Fine Beta testing for better software Wiley 2002 Charles L Forgy Rete A fast algorithm for the many pattern many object pattern match problem Artificial intelligence 19 1 17 37 1982 Wireless Application Protocol Forum Wag uaprof version 20 oct 2001 wireless application protocol wap 248 uaprof 20011020 a In WAP Forum October vol ume 10 2001 B Funke and H J Sebastian Knowledge based model building with KONWERK In ternat Inst for Applied Systems Analysis 1996 J C Giarratano and G Riley Expert systems PWS Publishing Co 1998 John Gill Access prohibited Information for Designers of Public Access Terminals Royal National Institute for the Blind 224 1997 J Goodman H Dong P Langdon and P Clarkson Factors involved in industry s r
167. ons and their surrounding Matt finish on buttons Visual appearance of program selection knob is good Visually clear and tactile marking on program selection knob Audible and tactile click from all major and minor controls on activation except the temperature button although this may be the result of damage to the display Good large size to the detergent drawer with enough space for any sized hand to fit in It is difficult to know what to do first Do you press On Or do you select a program Why is an On button necessary at all What is the difference between Start Pause Cancel and On Off Avoid bold and italics in labels It is not clear why some program labels are in bold italics and purple text while others are in regular grey text Glossy finish on button surround No obvious audible or tactile click from the temperature button on activation Parallax issues the user has to kneel in front of the display in order to read the full program guide the program selection knob blocks the view of the bottom programs For a quick wash does the user select Express 39 from the program guide or Quick Wash from the minor controls What is the difference There is no home setting for the program selection knob This means that the starting point for the knob might be different every time the user puts on a wash For users who count the turns in order to find the
168. op Keys should be raised above the body of the phone Pass preferably by 5 mm The pressure to activate a key should be between 0 5 and N A 1 Newton There should be auditory and tactual feedback of key N A activation Function keys should be tactually discernible from the Pass numeric keys There should be a tactual indication on the 5 key orona Pass QWERTY keyboard on the F and J keys A voice mode selection that announces all key presses N A One touch buttons are provided for ease of calling Pass telephone numbers stored in the memory Provide rotational or linear stop controls N A For keys that do not have any physical travel audio or N A tactile feedback should be provided so the user knows when the key has been activated e g a toggle switch or a push in pop out switch There is the ability to switch on or off any buttons on the Fail side of the telephone Where timed responses are required allow the user to N A adjust them or set the amount of time allocated to the task Physical Characteristics The phone should be easy to hold by someone with a weak Fail grip There should not be parts which can easily come off Pass The phone should be able to lie on a table and be operated N A one handed non slip material on the underside of the phone would help to hold the phone in place if it is used while lying on a table Any external antenna shou
169. ound colour combination should have Pass high contrast A clear open typeface font should be used for text Pass Table 5 21 Recommendation Checklist of Arcelik Washing Machine Panel 1 Part A 132 5 EVALUATION Recommendation Result Labelling continued from Part A Text should not be placed over a background image or over Pass a patterned background White or yellow type on black or a dark colour is more Fail legible The typeface weight and size are suitable Pass Room for improvement Upper and lower case is used Pass Washing Machines Minimum strength is needed to open and close the door N A Controls are easy to grip and turn Fail easy to grip but stiff to turn The door opens flat or as wide as possible for maximum N A access The dome in the door does not provide an obstruction to N A access Wheels are added for ease of moving top loading machines N A The door handle or button is easily activated N A The drawer for the soap powder is fairly large Pass Noise emission is at a minimum level N A Result 20 25 Table 5 22 Recommendation Checklist of Arcelik Washing Machine Panel 1 Part B Recommendation Result Controls Good visual contrast between the keys and the appliance Fail Major controls Good visual contrast between the keys and the appliance Fail Minor controls Key top
170. p 68 10 VE BE Dro Sy IStart 8 Index of Miconinstall t NX 7 5 Gateway P Figure 4 12 Annotation of a CAD object as a press button 90 a fo Ei mama SIEMENS 2 R Ed PM mm 4 APPLICATION IN DEVELOPMENT PROCESS TPHONEL STP _stp prt izli L Elle Edit View Format Tools Assemblies Information Analysis Preferences Window Help Ey star gt k 9 PH Command Finder Hh m e 9 00 Bee BA I lt Part Navigator Name lt Model History a EME JE art Navigator dy 1 M8 Body 2 M8 Body 3 M8 Body 4 M8 Body 5 28 Body 6 M8 Body 7 M8 Body 8 M8 Body 9 M8 Body 10 28 Body 11 CH o ls gt j yo p amp Annotation Jatum Coordinar y mM step presents all recommendations based upon your annotated types of components Recommendations Component Recommendations 05 For better tactlity keys should be raised above the body of the phone 10 R 05 Importance Level High Summary Keys should be raised above the body of the phone preferably by 5 mm B Body 12 AG Body 13 M8 Body 14 HM Body 15 M8 Body 16 AB Body 17 f 28 Body 18 HA Body 19 M8 Body 20 M6 Body 21 M8 Body 22 28 Body 23 Body 24 AG Body 25 M8 Body 26 M8 Body 27 26 Body 28 E Body 29 28 Body 30
171. pair ment group H10 lessThan hearing500hz 20 lessThan hearinglkhz 25 lessThan hearing2khz 30 lessThan hearing4khz 40 greaterThan backgroundnoise 100 x rdf type VICON HProfileO HI1 equal hearing500hz 20 equal hearinglkhz 25 equal hearing2khz 30 egual hearing4khz 40 equal backgroundnoise 100 x rdf type VICON HProfilel H12 equal hearing500hz 30 equal hearinglkhz 45 equal hearing2khz 65 equal hearing4khz 75 equal backgroundnoise 0 x rdf type VICON HProfile2 Figure 3 6 Recommendation Rules to create User Model Recommenda tion for impaired groups Figure 3 6 presents e g rules which were used to add a new membership to each User Model instance based on their predicates which are related to hearing impairments As mentioned in 3 3 2 these predicates define targeted WHO ICF im pairment groups After the reasoning step new classes are created e g HProfile1 for mild hearing impaired target users describing a classification of each User Model by lFor instance a membership of an instance to the User Model class is defined by x rdfitype Vi con UserModel 68 3 KNOWLEDGE MANAGEMENT Predicate Value Name For better tactility keys should be raised above the body of the phone Profile VI1 VI2 MD1 MD2 Summary Keys should be raised above the body of the phone preferably by 5 mm ID R 5 Source NCBI
172. pants were experienced in CAD a strong familiarity with in clusive design was there All participants were at least slightly familiar with Virtual User Models The group of participants of this evaluation is thus appropriate All described methods of customer involvement were applied e Quality function deployment QFD 7 11 e User oriented product development 1 11 e Concept testing 3 11 e Beta testing 3 11 e Consumer idealized design 1 11 e Lead user method 3 11 e Participatory ergonomics 1 11 As expected QFD is the most used method see also section 2 4 or Akao 2004 The group of participants is suitable as the modification of the product development process does have a strong impact on this method see section 4 3 Both Concept testing and beta testing were used by the same amount one participant described both together Lead user method concept and beta testing were applied in combina tion with QFD as they do not cover the complete product development process lAnswers only differ in applied customer involvement methods as researchers selected none 23 11 participants from research selected none applied customer involvement method 107 5 EVALUATION The coverage of involvement methods is positive with respect to the inclusion of the software framework into different product development processes Each method was applied by at least one participant of this study Suitability for the task
173. press the on off button il i2 i3 i4 i5 i6 i7 i8 i9 i10 ill i12 113 114 i15 il6 117 i18 119 i20 gl g2 g4 g5 g6 87 g8 g9 g10 g11 g12 g13 g14 g15 g16 g17 g18 g19 g20 tl t2 t3 t4 t5 t6 t7 t8 Two beneficiaries reported that it would be better if the button could be bigger t7 t8 37 47 beneficiaries successfully identified the volume up down buttons il i4 i6 i7 i8 i11 112 113 i15 117 120 gl g2 g4 g5 g6 87 g8 g9 g10 g11 g12 g13 g14 g15 g16 g17 g19 g20 t1 t2 t3 t4 t5 t6 t7 t8 Two further beneficiaries identified the cor rect buttons on their second guess il4 i16 Eight beneficiaries failed to identify the volume up down buttons and had to be prompted by the researcher i2 i3 i5 i9 i10 i18 il9 g18 All 47 beneficiaries were able to physically press the volume up down buttons il i2 i3 i4 i5 i6 i7 i8 i9 i10 ill i12 113 i14 115 i16 i17 118 i19 i20 gl g2 g4 g5 g6 87 g8 g9 g10 g11 g12 g13 g14 g15 g16 g17 g18 g19 g20 tl t2 t3 t4 t5 t6 t7 t8 46 47 beneficiaries successfully identified buttons il i2 i3 i4 i5 i6 i7 i8 i9 ilO ill i12 113 115 116 i17 118 i19 i20 gl g2 g4 g5 g6 87 g8 g9 g10 gil g12 g13 g14 g15 g16 g17 g18 g19 g20 tl t2 t3 t4 t5 t6 t7 t8 One beneficiary failed to identify the 5 1 7 buttons and had to be prompted
174. puter interaction and interactive systems pages 200 220 Springer 2005 Robert K Lindsay Bruce G Buchanan Edward A Feigenbaum and Joshua Lederberg Dendral A case study of the first expert system for scientific hy pothesis formation Artificial Intelligence 61 2 209 261 1993 ISSN 0004 3702 doi http dx doi org 10 1016 0004 3702 93 90068 M URL http www sciencedirect com science article pii 000437029390068M A Martini European working group on genetics of hearing impairment infoletter 2 european commission directorate Biomedical and Health Research Programme HEAR 1996 Larry Masinter Tim Berners Lee and Roy T Fielding Uniform resource identifier uri Generic syntax 2005 Svetlana Matiouk Markus Modzelewski Yehya Mohamad Michael Lawo Pierre Kirisci Patrick Klein and Antoinette Fennell Prototype of a virtual user model ing software framework for inclusive design of consumer products and user inter faces In Universal Access in Human Computer Interaction Design Methods Tools and Interaction Techniques for elnclusion pages 59 66 Springer 2013 149 REFERENCES REFERENCES B McBride Jena A semantic web toolkit Internet Computing IEEE 6 6 55 59 2002 Brian McBride Jena Implementing the rdf model and syntax specification In SemWeb 2001 William L Moore Concept testing Journal of Business Research 10 3 279 294 1982 ISSN 0148 2963 doi 10 1016 0148 2963 82 90034 0
175. r JACK and his female counterpart JILL are already well accepted by design teams in the product development industry They are able to perform different tasks using an avatar based upon anthropometric data sets see also Naumann and Roetting 2007 1 LLE F ETA TN MAMA Figure 2 6 RAMSIS 3D CAD ergonomics tool presenting geometric kine matic digital human model Source Human Solutions GmbH 2012 25 2 STATE OF THE ART Computer based human models are currently widely used in the development of vehicle interiors aircraft cockpits passenger spaces and workplaces The functions differ from the ergonomic design of driver and passenger areas to the overall design for an efficient maintenance and repair work An avatar mannequin is used for the representation of the beneficiary in both systems see figure 2 6 for RAMSIS figure 2 7 for JACK In the first step the de signer creates a virtual environment selects the avatar specifications and defines tasks Using probabilistic posture prediction for the avatar performing these tasks analysis output can present values for reachability comfort or viewport Poirson and Delangle compared several DHM tools including RAMSIS JACK Sam mie CAD Anybody or MakeHuman see Poirson and Delangle 2013 through a list of 25 comparison criteria Most DHMs do not include capabilities of users with im pairments see Zhou et al 2009 but rather anthropometric standards Kaklanis et
176. r Models are classified into impairment groups using predefined rules These rules classify each instance of User Models by data values as presented in tables 3 5 3 8 into different impairment groups 64 3 KNOWLEDGE MANAGEMENT Table 3 14 presents an informal description of the rule syntax of the reasoning Regarding the purpose of this thesis to get new information from the initial Ontol ogy model including User Models environments etc forward chain logic rules were implemented These rules represent different parametrical thresholds to add a new membership for single Ontology instances to each model if the parameters match the rule set For instance the rule Rule Description 9x rdf type Vicon UserModel For each instance of the class Vicon UserModel x Vicon UserModelArthritis artritis Creation of the variable artritis equal arthritis Y Check if the value of the parameter is y gt x rdf type Vicon UsersWithArthritis Resulting inference here a new membership is added infers all User Model instances which have a Y as value of the UserModelArtri tis parameter as members of the class Vicon UsersWithArtritis This scheme of rules is used for all models for more detailed information about the RETE algorithm itself see Forgy 1982 and Shrobe 1993 The complete reasoning of the framework can be seen as a sequence of Ontology model inferences with the result of new classes
177. r inclusive design purposes but without the use of the framework Due to the fact that the emerged user interfaces are prototypes at an early stage and do not have the full functionality as existing user interfaces it was not possible to perform tasks as receive a call or send a text message when using the mock up table 5 9 Executed Tasks With respect to product functionality different tasks were performed by participants of the study Some tasks on the mobile phone mock up could not be performed due to non functionality of prototypes Mobile Phones figure 5 9 Identify on button In the Doro PhoneFasyR332 the on button is the same as the off button It is visible as a small IEC 5010 power symbol on the red disconnect call button In the Mock Up phone this button as all buttons is the same but without the power symbol Successfully dial a number The participants should dial a number and tell if any problems occurred Press green button to connect call The green button is marked as a telephone handset and placed similarly on both phones On the Mock Up Phone all button labels were white Identify that a call is coming in This task could not be evaluated with the Mock Up Phone due to non functionality Press green button to receive call This task could not be evaluated with the Mock Up Phone due to non functionality Identify the message button SMS On the 332 the
178. r prototype evaluations e g beta testing Moore 1982 Beta testing Using a prototype of the target product beta testing refers to a field test with customers Due to the fact that a prototype must already be available this 23 2 STATE OF THE ART method should also not be the only method for feedback by end users and supplemented by methods applied to earlier phases It is most frequently used in software engineering Fine 2002 Consumer idealized design Consumer idealized design involves end users into product development imme diately Ciccantelli and Magidson 1993 In this approach customers create a design with support by a facilitator in a group exercise Participants select first representatives of the target market or several representatives for several target groups of the product The representatives create A new design a list of articulated requirements and a record of underlying reasons for the design choices Lead user method In this approach lead users represent users who face needs that will be gen eral in a marketplace but face them months or years before the bulk and expect to benefit significantly by obtaining a solution to those needs see Herstatt and Von Hippel 1992 Due to the fact that these users also find solutions with respect to their own needs a more active involvement is possible Participatory ergonomics Participatory ergonomics involves workers users themselves actively
179. ram C tl t3 t4 t5 t6 t7 All beneficiaries had difficulty reading labels t1 t2 t3 t4 t5 t6 t7 t8 4 8 beneficiaries reported that it is hard to find and understand programmes tl t4 t6 t8 2 8 beneficiaries reported that there is so many details on the labels making it con fusing t2 t3 All beneficiaries successfully identified the Main Control panel tl t2 t3 t4 t5 t6 t7 t8 6 8 beneficiaries have mentioned that some of texts are not meaningful and not easy to use tl t2 t3 t4 t5 t7 One beneficiary reported that it might be better if some text was more colourful or was identified by shapes t7 All beneficiaries successfully identified the Minor Controls panel tl t2 t3 t4 t5 t6 t7 t8 6 8 beneficiaries reported that some of la bels are not easy to understand t2 t3 t4 t5 t6 t7 Ar elik Washing Machine Panel 2 182 APPENDIX Issue Participants All beneficiaries successfully identified the on off button tl t2 t3 t4 t5 t6 t7 t8 Two beneficiaries reported that button is hard to hold and push t3 t7 All beneficiaries successfully identified the set Program A tl t2 t3 t4 t5 t6 t7 t8 7 8 beneficiaries reported that knob is hard to hold and turn around to set Program A tl t2 t3 t4 t5 t6 t7 All beneficiaries successfully identified the set Program B
180. rate low visual acuity Table 3 3 Separation of visual ability into three different profile groups Manual dexterity descriptors Definitions of manual dexterity ability No manual dexterity impairment Mild manual dexterity impairment Moderate manual dexterity impairment The subject does not have any restrictions regarding movement or force of joints The subject does not have arthritis but has slight problems when gripping small items and using small controls such as knobs sliders buttons or keys An early to intermediate phase of arthritis results in a moderate manual dexterity impairment of the user who is not able to handle controls and items if they are too small Table 3 4 Separation of manual dexterity ability into three different pro file groups 49 3 KNOWLEDGE MANAGEMENT Due to the variability of different manual dexterity diseases like Parkinson or Arthritis a classification into different levels is necessary for further steps of User Model development Table 3 4 shows the separation of manual dexterity impair ments analogously into three different groups of no mild and moderate impairments In terms of this thesis each impairment was separated into three levels of no mild and moderate All different User Models are classified into these groups to ease the further step of abstraction for the inference of quantitative and qualitative rec ommendations Figure 3 3
181. rding personal knowledge of participants106 Results of questions regarding suitability for the task 108 Results of questions regarding self descriptiveness 109 Results of questions regarding conformity with user expectations 110 Results of questions regarding conformity with user expectations 110 Participants of the user study by age and gender 114 Participants of the user study by impairment levels 114 Existing and emerged mobile phones used for evaluation of cus tomer satisfaction 2 2 2 2 20 0002 eee eee 116 Existing and emerged remote controls used for evaluation of cus tomer satisfaction 2 0400 117 Existing and emerged washing machines used for evaluation of customer satisfaction 60000 118 General Comments Doro Mock Up Phone 121 Recommendation Checklist of Doro Mock Up Phone 122 Results of user study regarding mobile phones 123 Result comparison regarding mobile phones 123 Results of user study regarding tv remotes 126 General Comments Ar elik Washing Machine Panel 1 Part A 127 General Comments Ar elik Washing Machine Panel 1 Part B 128 General Comments Ar elik Washing Machine Panel 2 Part A 129 General Comments Ar elik Washing Machine Panel 2 Part B 130 Recommendation Checklist of Ar elik Washing Machine Panel 1 Part e e e a e e eee 132 a Part Bos me Se uy Epa
182. rds exist referring to approaches and recommendations presenting design issues with respect to devices and user impairments but are not directly included in the product development process section 1 3 Section 1 4 presented the approach how to separate these recommendations into quantitative and qualitative for later use Research questions were defined to state the topic of this thesis section 1 5 resulting in three hypotheses section 1 6 re garding the possibility to use Ontology based models for storage of data a suitable reasoning to describe scenarios and designer acceptance as designers are the end users of the framework Chapter 2 will present the relevant state of the art for the issue of supporting inclusive design during the product development process 13 Chapter 2 State of the Art An objective of this thesis is the storage and management of knowledge needed for the design process The title of this thesis already raised one possible answer defining Ontology based models The following section deals with the question of how and which kind of context aware systems can be used for the representation integration and inference of knowledge This includes the manipulation of data and the creation of results based upon rules relations and constraints 2 1 Current Product Development Process The product development process covers the product design from first creative ideas to the creation of the final prototype most oft
183. reate objects in a virtual environment including simple boxes spheres cubes etc without specific functional context With respect to the aim to support the de signer functionality of a component must be annotated previously see 4 2 3 The Component Model defines all currently available functional types see 3 3 5 to get qualitative and quantitative recommendations for each annotated CAD object and is used in the annotation tab of the CAD module So the designer can set up context to the model Afterwards the CAD module provides a set of recommendations based on annotation selections This can also be applied directly by the module if it is a quantitative recommendation To summarize the framework includes the following applications and services e Administration User Interface The Administration User Interface provides different tools to change and manip ulate the initial Ontology and rule sets Also the Sketch Application is included for preview purposes of the final VUM e Sketch Application This application connects to the socket server and provides an interface to dis play different recommendations based on the selections of user profile typical environment and typical task 74 4 APPLICATION IN DEVELOPMENT PROCESS CAD Siemens NX Siemens NX is used as a CAD software solution for the approach presented in this thesis API NX Open Siemens NX includes an API called NX Open to access the virtual environment which is
184. rements and led to the conclusion that especially On tology based models see section 3 1 are suitable with respect to requirements of partial validation level of formality and applicability to existing environments as an answer to hypothesis 1 Hypothesis 1 Ontology based model application Ontology based models can be used to store and manipulate various data con cerning requirements especially of elderly people for the use of products Various models e g User Model were structured based on a user study with beneficiaries resulting in a definition for each part User Model Environment Model Task Model Component Model and Recommendation Model were defined and com bined for scenario representation The combination Virtual User Model including reasoning is able to connect recommendations by input selection of target benefi ciary group typical environment in which the product can be used and typical task 71 Chapter 4 Application in Development Process The following chapter focuses upon the application of the knowledge base presented in the previous chapter in a development process Using inclusive design guidelines personal expertise of designers and user studies an Ontology was defined includ ing recommendations All resulting data including context information and recom mendations qualitative and quantitative see section 3 3 6 use graphical front end applications and an integrated module in CAD software 4 1 Sys
185. rments was created see VUMS White Paper Finally it targets at helping designers and developers to maximize the level of us ability and accessibility of products and services by providing appropriate user models Moreover they are intended to be used for the generation and adap tation of user interfaces during runtime It presents general definitions and a ISourceForge website http sourceforge net projects convic 2VUMS White Paper can be found at http veritas project eu 2012 02 vums white paper 142 6 DISCUSSION AND FUTURE WORK concept of generic interoperable user models that describe the relevant charac teristics of users interacting with products and user interfaces These include physical cognitive and sensory attributes habits preferences and accessibility capabilities Extending to functionality as top hierarchy A hierarchical superstructure of target functionalities as n m relations would be preferable especially if the design is not restricted to a device If design is not restricted interaction recommendations between human and computer would be possible For instance the simple interaction of reading and writing emails would infer a display keypad and a touch display for the same purpose Both solutions could handle the task Designers could select human computer in teractions and choose from a set of solutions In addition if there are new technological advancements new devices could be incl
186. ruppe t Fj o g Objekte ausw hlen und MT3 verwenden oder auf ein Objekt do Rollen o Melik O Industriespezifisch a Systemvoreinstellungen 85 Erweitert N Erweitert mit vollst ndigen Men s amp z Wesentlich Empfohlen a BP Wesentlich mit vollst ndigen Men s P I DEAS E E Power Drafting Advanced 7 Power Drafting Essentials ViconToolset A 2 TE MS M o Wels BS PL OG Figure 4 10 Special Vicon Role selection in Siemens NX 4 Get textual recommendation and or apply recommendation Based on semantic information about the annotated objects qualitative rec ommendations can be applied to an object immediately Figure 4 13 presents the recommendation view of the CAD module with the Apply Recommendation Button Designers receive as output all recommendations from the first phase by VSF im port and component related recommendations annotated to the virtual components Latter instances can have the relation to different component parameters e g spe cific attributes of the component including nominal values e g size of a component For the application of qualitative recommendations these parameters must be pre 89 4 APPLICATION IN DEVELOPMENT PROCESS Star D ll 9 Cx Y Command Finder Th le Edit View Format Tools Assemblies Information Analysis Preferences Window Help e eteT sr ak Pek SIEMENS amp Name sP
187. s and relations e g using Unified Modeling Language UML Rumbaugh et al 2004 UML diagrams combine elements focusing upon the direct representation of relational data For instance Hendricksen et al Henricksen et al 2005 presented a context ex tension to the object role modelling ORM approach by Haplin et al Halpin et al 2008 as presented in figure 2 2 in which different facts of context information is described as entities and facts Strang and Linnhoff Popien 2004 2 2 4 Object Oriented Models Object oriented models like the cues as presented by Schmidt and Van Laerhoven Schmidt and Van Laerhoven 2001 mainly focus upon encapsulation and fusion of data Baldauf Dustdar and Rosenberg Baldauf et al 2007 described these models to offer the full power of object orientation e g encapsulation re usability inheritance Accordingly these factors to be able to divide all kind of information and build relations between them is one main requirement for the topic of this thesis object oriented models define a possible solution 2 2 5 Logic Based Models Logic based models offer a very high degree of formality see Baldauf et al 2007 and Strang and Linnhoff Popien 2004 including a possibility to use information to infer results based upon rules or relations This reasoning step is able to add update or delete context information automatically with the requirement of a strong formalisation 2 2 6 Ontol
188. s should be convex or flat with a raised edge Pass Keys should be as large as possible without reducing the Pass distance between the keys to less than half the key width Ideally the keys should be internally illuminated but the Fail internal illumination should not reduce the legibility The visual markings on the keys should be high contrast Fail clear and as large as is possible on the key top The pressure to activate a key should be between 0 5 and N A 1 Newton Table 5 23 Recommendation Checklist of Arcelik Washing Machine Panel 2 Part A 133 5 EVALUATION Recommendation Result Controls continued from Part A There should be auditory and tactual feedback of control Fail activation For controls that do not have any physical travel audio or Fail tactile feedback should be provided so the user knows when the control has been activated e g a toggle switch or a push in pop out switch There is a clearly labelled reset control N A Buttons or keys have tactile markings Fail Buttons or keys including touch screen buttons are large Pass and easily identifiable from each other Buttons or keys are operable with one hand Pass Glare is minimised on the surface of the product Fail Instructions Program Guide Use simple clear concise language Fail Be task orientated Pass Use a typeface with good legibility Pass but bold and italics should not
189. section 2 4 seven customer involvement methods were reviewed resulting in the need of Virtual User Model to include as much information about beneficiaries as possible This section describes the motivation for the next chapter DHMs were described in 2 5 as virtual product prototype evaluation tools which allow designers to simulate tasks performed by virtual avatars indicating ergonomic issues but for the simulation a virtual prototype must already be available With respect to results of the first and fourth section of this chapter see section 2 1 and 2 4 a support at an early stage is advantageous and will be further focused Related projects of this field with similar approaches were analysed in section 2 6 with different purposes The projects MyUl and GUIDE focus on software devel opment issues regarding requirements of elderly people while VERITAS and VICON relate to a supporting framework for designers All projects conducted user trials for a scenario definition by an XML Ontology approach The next chapter will present the knowledge management approach used in this thesis 43 Chapter 3 Knowledge Management 3 1 Context Modelling With respect to the state of the art of context aware systems Strang and Linnhoff Popien Strang and Linnhoff Popien 2004 presented a survey based upon demands on context modelling approaches The conclusion of the survey indicates that Ontol ogy based models fulfil most of the requirements to u
190. ssible ui design and multimodal interaction through hybrid tv platforms towards a virtual user centered design framework Universal Access in Human Computer In teraction Users Diversity pages 32 41 2011 F Hayes Roth D Waterman and D Lenat Building expert systems Addison Wesley Reading MA 1984 A Held S Buchholz and A Schill Modeling of context information for pervasive computing applications Procceding of the World Multiconference on Systemics Cy bernetics and Informatics 2002 K Henricksen J Indulska and T McFadden Modelling context information with orm In On the Move to Meaningful Internet Systems 2005 OTM 2005 Workshops pages 626 635 Springer 2005 C Herstatt and E Von Hippel From experience Developing new product concepts via the lead user method A case study in a low tech field Journal of product innovation management 9 3 213 221 1992 Human Solutions GmbH Documentation of the RAMSIS Software http www appliedgroup com ramsis 2012 Accessed 2012 08 28 Robert J Ivnik James F Malec Eric G Tangalos Ronald C Petersen Emre Kokmen and Leonard T Kurland The auditory verbal learning test avlt norms for ages 55 years and older Psychological Assessment A Journal of Consulting and Clinical Psychology 2 3 304 1990 C Jung and V Hahn Guide adaptive user interfaces for accessible hybrid tv appli cations In Second W3C Workshop Web amp TV 2011 Nikolaos Kaklanis Panag
191. st table 5 15 presents the result comparison regarding mobile phones Positive Negative e Button size is good e Space between buttons 1 2 3 is too far apart causing the phone to be too wide As a result it would be diffi cult to hold and use the phone in one hand e Numeric labels are clear e Colour contrast is good There is no obvious on off button The function of A B and C buttons is not obvious e The A B and C buttons are in a par ticularly prominent location It is not clear why they need to be given a prime location when not every user will use them It seems they could be easy to accidentally hit off It is not clear how you would select a menu option on the screen There is an up down button in the top centre of the keypad but there is no OK or select button s Table 5 12 General Comments Doro Mock Up Phone 121 5 EVALUATION Recommendation Result Keypad Good visual contrast between keys and body of the phone Pass Key tops should be convex or flat with a raised edge Pass Keys should be as large as possible without reducing the Pass distance between the keys to less than half the key width Ideally the keys should be internally illuminated but the N A internal illumination should not reduce the legibility of the numbers in daylight The visual markings on the keys should be high contrast Pass clear and as large as is possible on the key t
192. sults of Beneficiary Tests Source Vicon Consortium 2013b Profiles Age Gender Hearing Vision Manual Beneficiary Code Dexterity g Germany i Ireland t Turkey 65 Female Mild Mild No gl 80 Female Mild No Moderate g2 88 Male No Severe Moderate g3 73 Female No No Moderate g4 88 Female Moderate No Mild g5 94 Female Mild No No g6 82 Female Moderate No Mild g7 84 Female No Moderate Mild g8 89 Female Mild Moderate Moderate g9 70 Female Mild Mild No gl0 Continued on next page lOne beneficiary was deemed to have too severe a vision impairment to be included in the analysis the total sample size therefore for the product analyses 47 instead of 48 see 5 3 165 APPENDIX Age Gender Hearing Vision Manual Beneficiary Code Dexterity g Germany i Ireland t Turkey 83 Male No Mild No gll 80 Female No Mild No gl2 92 Female No Mild Moderate g13 90 Female No Moderate Mild gl4 82 Male No Moderate No gl5 66 Female No Moderate No gl6 81 Male No Moderate Mild g17 91 Female No Mild No g18 76 Female No Moderate No gl9 87 Female No Moderate No g20 74 Female Mild Moderate Mild il 88 Male Moderate Mild Mild i2 90 Female Moderate Moderate Mild i3 66 Male No No Mild i4 84 Female Mild Mild No i5 65 Female Moderate Moderate No i6 65 Male
193. ta is already included 75 4 APPLICATION IN DEVELOPMENT PROCESS For usability purposes and especially regarding acceptance of designers a soft ware installer is provided to install all parts of the framework During the installation process and in case Siemens NX is already installed it creates a new user role and all necessary registry values in which the CAD module is included in the toolbar of Siemens NX 4 1 2 User Input Draft Phase Qualitative recommendations and guidelines User model Environment Framework Task ConVic Module Vse impon Qualitative and quantitative adaptable recommendations CAD Phase User Interface definition Component specifications Product Development Process Designer input upporting the Development Process i Figure 4 2 User input of the designer supporting the sketch design and CAD phases of Product Development Process Figure 4 2 presents a functional diagram concerning the input of designers and the output of the framework As previously mentioned see 2 1 designers create sketches using paper drafts or software solutions To maintain a flexible support a stand alone application was implemented ConVic Hereby designers can select different possible scenarios in which the target product can be used resulting in qualitative see 1 4 textual design recommendations A multi selection of scenarios 76
194. tem architecture This section deals with the application part of the system giving a general overview required tools were developed A software back end facilitates this 4 1 1 Overview Figure 4 1 shows the system architecture with the different applications The ap proach consists of a socket server representing the Ontology interface for sketch and CAD product development phases and the backend Ontology In accordance with the proposed Virtual User Model a software framework has been implemented as a core part of the support system The aim of this framework is to support designers in a non obstructive way during the product development In the first phase designers create draft sketches of the target product This step can include different software solutions but with respect to requirements of design ers they often create these drafts on paper sheets Therefore a stand alone solution 72 4 APPLICATION IN DEVELOPMENT PROCESS suopeordde pus 1uozj pue yoeq SurureJuos YIOM9UTEJJ 9IEMYOS Adwo L 1 MSH WMA MNRAS suolo 9pow pu wwo y juauodulo gt s jny opoq a npow UYOHDLJSIUILU 9pow 9pow A6o oluO e py Asp JUSUIUOJAUJ 9poy asp P J0H1p3 46oJouO 1 pupH uojp uuo gt 4220S J uosp 2 ony 5iuSuso J9 puby sulBuz Buluosp q uollp9 uuo VNJ J i 9INPOW AV P tbiB lui XN Ssu ui is uoypo1 dd y 424245 pJi lu 195 N UO DAJSIUILUp YW u do XN IdV XN su
195. text fonts or push buttons are too small Existing recommendations and design guidelines provide this kind of information about the needs and requirements of beneficiaries with respect to impairments of the users see 1 3 and 1 4 for a detailed review Sustainable interfaces must take as many issues as possible into account to maintain a proper use of a product resulting in the question which recommendations about different aspects for a specific prod uct are important for which product and how they should be presented to designers to be accepted in the development of a product 1 2 Targeted Impairments Two challenges drive the idea of inclusive design the demographic change and the growing number of functionalities devices are able to perform 5 Men E Women Solid colour 2009 Bordered 1990 Figure 1 3 Population pyramid for EU 27 2009 excluding french over seas departments Source European Commission 201 1 Figure 1 3 from European Commission 2011 presents the population pyramid for 27 European countries for the years 1990 and 2009 Both life expectancy of women and men increased during this period This demographic change to more elderly people does have an impact upon product customers so user needs must be included in the product development process Especially consumer products like 1 INTRODUCTION mobile phones have a responsibility to act and adapt to these changes 5 Men E Women Solid colour 206
196. the Euro pean Group on genetics of hearing impairments EGGHD Separation of visual ability into three different profile groups Separation of manual dexterity ability into three different profile PLOUDS e ee doy s Wein Rm di e ee BR ede Sr Re Ee ete es Ontology class data properties used for User Model General characteristics so secde eb a a E Ontology class data properties used for User Model Hearing Ontology class data properties used for User Model Vision Ontology class data properties used for User Model Manual dex Ontology class data properties used for Environment Ontology class data properties used for Task Ontology class data properties used for Component Ontology class data properties used for Recommendation Informal description of the simplified text rule syntax of reasoner One instance of the recommendation class Used DL Expressivity of Ontology Results of statements about general concept 64 LIST OF TABLES LIST OF TABLES Tab Tab Tab Tab Tab Tab Tab Tab Tab Tab Tab Tab Tab Tab Tab Tab Tab Tab Tab Tab Tab Tab Tab Tab Tab Tab Tab Tab 5 2 5 3 5 4 5 5 5 6 5 7 5 8 5 9 5 11 5 12 5 13 5 14 5 15 5 16 5 17 5 18 5 19 5 20 5 21 5 22 5 23 5 24 5 25 5 26 5 27 5 28 5 29 Results of questions rega
197. the inclusion of recommendations is advantageous independently of involvement method applied The variety of approval about the time consumption to go through the recommen dation list can be seen very ambiguous see figure 7 While on the one side it would be time and cost saving to be able to go through the recommendations very fast on 108 5 EVALUATION the other side a deceleration of product design would be more fruitful with respect to more accessible products and designer creativity The aim of this question was to maintain this assumption This will also be an issue in the next section of the evaluation The next two statements were answered very similarly with a trend to approval figure 8 and 9 Participants of the study would need a user manual to the software and the look and feel of the software was suitable and pleasant Regarding the front end presentation the results are slightly positive but can also be improved The last question of this section described the approval to the general statement that the software is easy to use Eight participants approved this bar in table 99 5 3 while 2 disapproved and one skipped this question The strong approval concludes that the software in its current state is already easy to use by designers but can be improved One participant could not use the software directly on the own pc and commented that he was not able to install and use the software on a
198. themselves In design with representa tives of the target group participate in the design process e g within evaluation but the designer is still creating the product Design bu moves this responsibility to the target group entirely 91 4 APPLICATION IN DEVELOPMENT PROCESS The method as presented in this thesis refers to the inclusion of guidelines into the design process without the involvement of real customers beneficiaries into the product development process The involvement of the target group is handled by context information based on user studies Also a specification for only one target group or a combination of different groups is possible Concept Detailed Specification Development Design Phase of the design process Prototyping Final Product 4 User oriented Product Development Concept Testing Design Lead User Method C customer denied Dese gt 4 Participatory Ergonomics Impact field Y z N Beta SS Y ot Y Type of customer involvement Figure 4 14 Impact on different methods of involvement Figure 4 14 describes all previously mentioned methods of end user involvement in relation to product development phases The impact field represents affected phases of the product development process Referring to the theses of this work see section 1 6 it is aimed to include the presented tools into product development process phases without
199. ther interfaces In the current framework version the links can only be included as qualitative recommendations Regarding quantitative recommendations an extension would be needed to include recommendations as functions depending on more than one parameter which also requires new user trials for the definition of parameters and values The use of the envelope logo rather than the term SMS was more in line with inclusive design guidelines as the term SMS could be classified as technical ter minology or country specific terminology being unfamiliar in some countries e g Ireland The choice of a logo instead of text leads to overcome barriers relating to language or literacy A quick survey reveals that either one or a combination of three commonly used icons is used the term SMS an envelope or a speech bubble Also a combination of these is possible e g a SMS in a speech bubble An universal icon for SMS is lBy looking at ISO and ETSI Standards relating to pictograms using a Google image search using the terms text message icon and SMS icon Most commonly used mobile phone operating systems Android iOS Windows Phone have their own standard icons but can also be replaced by new themes 124 5 EVALUATION not available Also the recent evolution from mobile to smart phones as mentioned in section 1 1 increases the dilemma since icons are needed that differentiate not just between simple SM
200. tions related to customer involvement method 103 5 EVALUATION The online questionnaire starts with general information about the topic and the aim of this study The first questions relate to personal information including pre knowledge about inclusive design and Virtual User Modelling for later classification In addition and as presented in section 2 4 the type of customer involvement is asked see figure 5 2 In the next step participants used the installer mentioned above The question naire is seen divided into 2 different parts The first part deals with the explanation of both tools sketch design tool and cad module including questions related to the use and complexity The second part contains questions related to ISO Norm 9241 110 see Schneider 2008 The sketch design tool is described by an explanation of all input fields for design ers as presented in figure 5 3 Convic File Edit tmport Export LooksFeet Recommendations Knowledge Base Treeview 00 Cood vau constant 5 Ther aba Toutt 201 imorove tne tactile detection of hey tops medium importance HH 202 Larce ieys for better aitterentiation nigh importance Name For better tactility keys should be raised above the body of the phone Summary Keys shoud be raised above the body of the phone preferably by mm Text People who rely on touch to operate keypads benefit from keys that are as distinctive as possible to the touch R
201. tity a buton button 7 h identity a button bution 6 Identify a bution bution 2 _ 2012b Evaluation Results Doro VICON Prototype Dorothy Kitchen Test Label font size Test Internal illumination Subtask Reach On Off button Test E grip of controls Test Grasping facilitation Subtask Press On Off button Test Activation pressure for buttons con 7 For educational use only Figure 2 14 Simulation output of VIRTEX Source D 41 2 STATE OF THE ART virtual environment in which the product should be used The simulation performs a predefined set of tasks related to a specific device type During the simulation see figure 2 13 the virtual humanoid performs each subtask resulting in a classification if the task was successful or a failure marked with a green and red background for each task Currently processed tasks are marked yellow The output of the system also includes recommendations that are related to each task Figure 2 14 presents such an output testing a mobile phone prototype 2 6 5 Comparison MyUI Creation of software adaptive user interfaces with respect to end user impairments GUIDE Creation of a software framework for designers to create adaptive TV interfaces for elderly people VERITAS Support designers in product development by a complex simulation framework including end user impairments VICON Support designers by giving recommendations in early phases and vir
202. to define a right construction methodology which is valid for all problems The requirements are too different On the one hand different product groups result in differ ent requirements for construction methodology on the other hand engineering departments already have experience based upon the operational field of the company Additionally each designer has his own personal experiences and preferences which he prefers translation for original text see Berthold 2002 p 35 15 2 STATE OF THE ART specification decomposes the product into parts and sub parts referring to functional requirements of the product As a result and based upon the form and surface of previous drafts a concrete functional model can be seen in a virtual environment Figure 2 1 presents these phases in the product development process according to VDI Gesellschaft Entwicklung Konstruktion Vertrieb 1993 translation see p 16 ff As already mentioned in this thesis especially the phases Draft Phase and CAD Phase are focused upon by giving support for the creation of inclusive designed products 2 2 Context Modelling All aspects dealing with issues related to inclusive design must be representable It is also necessary to ease the inclusion of facts and issues derived from guidelines and personal experience of designers into the database to gain the advantage of ma nipulation of constraints directly The following example recommendat
203. tual simulation for evaluation of virtual prototype Table 2 5 Focus of related projects Table 2 5 presents the focus of each project like software development for different scenarios MyUI GUIDE as well as physical user interface development VERITAS VICON including aspects of impaired end users to create more customer oriented products In each project user trials were conducted and a XML Ontology approach was driven to reflect scenarios 2 7 Conclusion In this chapter the current state of the art with respect to the topic as presented in chapter 1 was defined In the first section 2 1 product development process from a general point of view was elaborated resulting in the specification of phases in which design support is possible and needed The phases Draft Phase and CAD Phase were identified as suitable for a quantitative and qualitative support during product development process In section 2 2 several context modelling approaches were pre sented including concise definitions of each method Based upon the investigations conducted in chapter 2 chapter 3 will present a survey in which each method will be 42 2 STATE OF THE ART compared using a selection of requirements Next to context modelling approaches expert systems were introduced as an alter native approach in which the user is able to configure concepts as a representation of a specified scenario as seen in section 2 3 In
204. tual user concept can help to involve the user s perspective into the development process earlier Pm more hesitant for this statement The problem is that the model is put rather late So that is maybe too late or for some parts of the design maybe too late Labelling and textures can still be adapted but in order to change some forms there are not enough time and money usually available This is my opinion for this version of VICON If VICON is developed and in cluded my opinions above it will be strongly agree The focus is not on the time but on the complex information context which I as a designer get Things I have to consider are good packed it s good platform where I could inform myself and get an overview also before starting the design 98 5 EVALUATION Yes it can but a good designer should always think about the user s perspec tive first before beginning sketching The question is Is the data from the database really the user s perspective This shouldnt solely rely on the virtual concept instead of going to the real people Inclusive design is not a group of people you cannot summarize all the individuals Of course there are Personas and categories but everyone is different The sketch tool could have some help it provides a list with recommendations But we already have it One answer to the third statement is interesting in particular The statement issued
205. tware framework would add the possibility for designers to create a virtual model of their product press one big Start button and afterwards the product is inclusive The software would change the complete design to end user needs itself On the one side it would be great to have such a solution but this would also result in a smaller variety of product designs Each product would only focus on the connection of technological and human factors by guidelines and conditions This extreme scenario results in inclusive design but would also destroy creativity during product development process The amount of different designs would be narrowed by the target device resulting in very similar end results The presented framework can be seen as a first supportive concept regarding inclu sive design and was proven that it already can help during product development 141 6 DISCUSSION AND FUTURE WORK processes But for a seamless integration without restrictions even to devices a hierarchical superstructure about various kinds of interactions is necessary As seen in section 1 1 during the last years technology evolved due to new functional possibilities but also new kinds of interaction speech gesture etc For instance buttons are increasingly replaced by touchscreens both reflecting the same func tionality Different devices can provide the same functionality It is possible to include a higher stage of hierarchy into the framework pres
206. ubtasks the task is Subtasks composed of Subtasks String Hierarchically numbered list of subtasks The numbering scheme is as follows lt Number gt lt SubtaskNumber gt e g 2 4 for the fourth subtask of the task number two The subtasks in the list are separated by comma Table 3 11 Ontology class data properties used for Task These abstract values are used to represent an abstract relation between the tasks and the problems if the target user fulfils this task The recommendation definition of the TaskRule attribute see table 3 13 of each recommendation connects each recommendation to a specific task see 3 3 6 and 3 4 4 for a detailed review 3 3 5 Component The component model is used in the CAD phase of the project It defines the annota tion options during the annotation step in the CAD module Each component refers to a different set of recommendations which can also be optionally applied to a CAD object 61 3 KNOWLEDGE MANAGEMENT General characteristics Predicate Name Datatype String Description The annotated component name will be presented in the CAD Annotation Form State String How many states can the component perform e g switch with 2 states Function String Description of the functionality of each component e g binary state change for press button Table 3 12 Ontology class data properties used for Component As already m
207. uded by provided func tionality Extension of recommendations An extension of recommendations would always be preferable especially regard ing new devices In the current version recommendations focus on the interac tion used by mobile phones tv remotes and washing machine panels There is a need to include new instances in the task model when extending the knowledge base to new devices Inclusion of structured and annotated CAD objects Regarding the CAD environment the aim would be to describe exemplary CAD objects which could appear in the integrated module Designers would be able to select single objects from a predefined and already inclusively created set of objects for new products The reasoning would be able to present a subset of possible objects based on different conditions of the target product 143 References C Abbott tiresias org information resource for people working in the field of visual disabilities Journal of Assistive Technologies 1 1 58 59 2007 Y Akao Quality function deployment integrating customer requirements into product design Productivity Press 2004 Franz Baader The description logic handbook theory implementation and applica tions Cambridge university press 2003 M Baldauf S Dustdar and F Rosenberg A survey on context aware systems International Journal of Ad Hoc and Ubiquitous Computing 2 4 263 277 2007 A Berthold Der fertigungsorientierte Modellierer FERMOD a
208. used to read and manipulate all virtual objects Siemens NX integrated CAD module The CAD module connects to the Socket Server to provide different support to designers while creating and manipulating a product in CAD Software Siemens NX JENA Connection Handler The Handler is used for the direct connection to the Ontology by parsing and translating commands into SPARQL to access the Ontology Prud Hommeaux et al 2008 Equal commands are also implemented in the Socket Connection Handler Socket Server This part of the software is not visible to the end users designers It provides a middle layer between all applications to the Ontology data Generic Rule Reasoner Using the reasoner inferred from the initial Ontology the final construct is cre ated as presented in 3 4 Socket Connection Handler Similar as the JENA Connection Handler this part parses and translates com mands as get users into an equivalent SPARQL command to access parame ters and data properties of each model Prud Hommeaux et al 2008 Ontology Editor and Administration Module Using SPARQL commands generated by the JENA Connection Handler this module also uses JENA to read and manipulate the Ontology data directly Prud Hommeaux et al 2008 Ontology Data User Model Environment Model Task Model Component Model Rules and Recommendations This part represents the initial model which is used for the reasoning All con text related da
209. ush a button Additionally these values can be set and reset automati cally by recommendations e g to minimum values With respect to the analysis of the product design processes on behalf of indus trial partners and the expectations provided by respective designers and developers it turned out to be a basic requirement that the shape of a product and or its user interface should not become dictated by a recommendation system Moreover de signers usually prefer to start with sketches and drawings on paper from scratch In order to achieve this the system provides templates for component names that can be imported into an existing product model This way the product developer has the alternative either to compare his own model with the loaded component templates or to use the template according to his ideas In other words in order to support creativity for the product developers parameters and dimensions of a CAD model are only manipulated within the predefined templates The utilization of component templates provides a further advantage that the ge ometrical dimensions of the components can be reduced to a subset of core param 86 4 APPLICATION IN DEVELOPMENT PROCESS Sjuauoduwo 2 ejouuy lt sapnpul gt suonmepu uiuio35 tl yu uoduio2 N IA suolepu9ulu 029Y ASEL MAJA suoljepU9Uu lu029Y JU9UWUOJ AUJ MOLA suomnmepu uiulio3 ti luoid 9PON 13SN N IA AE lt saphpul gt lnpoutu AVI
210. ut remote controls Publications Aremote control can also be referred to as a remote or controller Itis known ICT projects by many other names as well such as the clicker channel changer etc Commonly remote controls are used to issue commands from a distance to E televisions or other consumer electronics such as stereo systems and DVD players Remote controls for these devices are usually small wireless handheld objects with an array of buttons for adjusting various settings such as Checklists television channel track number and volume In fact for the majority of modern devices with this kind of control the remote contains all the function controls while the controlled device itself only has a handful of essential primary Standards controls Most ofthese remotes communicate to their respective devices via infrared IR signals and a few via radio signals News 8 events Universal remote controls Onks Auniversal remote is a remote control that can be programmed to operate various brands of one or more types of consumer electronics devices Low end universal remotes can only control a set number of devices determined by their manufacturer COST 219 Archive while mid and high end universal remotes allow the user to program in new control codes to the remote Many remotes sold with various electronic devices include universal remote capabilities for other types of device which allow the remote to control other devices
211. w avo 1 p 1 pp W 194195 pug og 1ua gt pu3 juoly 73 4 APPLICATION IN DEVELOPMENT PROCESS was desired The resulting recommendations are exported for later use in following design phases The software application called ConVic is used in this phase and consists of two different front ends The sketch application front end see section 4 2 1 presents an interface to the designer to get qualitative recommendations based on different user input scenarios For maintenance manipulation and extension purposes of the VUM before reasoning an administrator interface directly communicates with the Ontol ogy using a connection handler as a middle layer ConVic connects to the Ontology in the back end left side of figure 4 1 Also the sketch application included in the ConVic connects to the Ontology using a socket server middle layer In this middle layer the reasoning is implemented as presented in section 3 4 to access the VUM af ter reasoning This separation of both connection types before and after reasoning was necessary due to the reasoning steps The socket server provides access to the final construct the administration module to the initial model In this first step a VSF Vicon Status File see 4 1 4 is created for export includ ing the current input scenario selection of the designer It can be imported in the integrated CAD module This is used in the second phase CAD in which design ers c
212. w a e ee 40 2 6 5 Comparison Z ses ee e a a ie e yasa SQ DE a de 42 2 7 COMCIUSI N cs io ws aa a ia e a 42 3 Knowledge Management 44 3 1 Context Modelling 60 44 3 2 User Study e scs a as ala aus WU k amaa ea 46 3 2 1 Impairments s s c a k s w w w s dada w s Ua css 47 3 2 2 Methodology vece sassa was v4 ve uu ee os se w dp N 50 3 2 3 OUICOM S ap y e a op da E e ier e oy e ice e el aa 51 3 3 Virtual User Model o SUN WR W Se 51 3 3 1 Knowledge Base eee 52 3 3 2 User Model s ra A AR y ke 53 3 3 3 Environment o sa eses essek esasan 58 OA Task dida A AAA AAA sade sahi 60 3 3 5 Component sas 0 ee ae ee ee ms ae 61 3 3 6 Recommendation 00000 62 9 4 Reasoning ik Se e e EE e elek A e ek ilm e ee 64 SA T RUES AN 64 3 4 2 Reasoning Engine sas sas sns sa sns ns ea sa 65 3 4 3 ReasoningApproaceh e 66 3 4 4 Ontologylnference rr r rr rr s rs s s 68 3 4 5 Description Logic Expressivity 70 3 4 6 Multiple Selection a 71 3 0 CONCIUSION 4 s e e sa seo ehh ee AR ache de de 71 4 Application in Development Process 72 4 1 System architecture Q Q GW s W 72 All OVERVIEW a Res te s S A de 72 4 12 User In putin ise e edim eke ib dt tl S SUN 76 4 1 3 Server Tool lt sua z sa ee casar a 77 4 1 4 Vicon Status Files s s eos o 78 4
213. were implemented to generate a realistic virtual scenario Kaklanis et al 2010 A database of target users including nominal and categorical values for impairments and characteristics of elderly users was created to maintain a precise simulation Moschonas et al 2012 Table 2 3 exemplifies values for different impairment profiles which are included in the Virtual User Model These values are used for different measurements in vir tual environments to create results as presented in table 2 4 The table displays different attributes related to the virtual humanoid which are included in the User Model physical characteristics Normal values represent nominal range values if the virtual humanoid represents a user that does not have any impairment The other values define degrees of freedom for rheumatoid arthritis Pe a Guevara et al 2005 spinal cord injury Eriks Hoogland et al 2009 adhesive shoulder capsuli tis Kazemi 2000 hemiparesis Zackowski et al 2004 or an average elderly man between 75 and 79 years For instance a User Model with rheumatoid arthritis has a major impairment in the ranges for shoulder flexion shoulder abduction and shoul der external rotation Modelling Framework The modelling framework in VERITAS consists of 3 different parts All parts consist of context information that are necessary for the Simulation Platform UsiXML was used to implement preferences and attributes see Limbourg et al 2005
214. x Vicon UserModelVisualAcuity visualacuity is added to the term Each of predicate value can be compared using the syntax presented in Table 3 14 Each User Model instance is classified by parameter values For instance the clas sification of mild manual dexterity impaired user groups is made using the following rule equal arthritis N equal grip 2 equal controls 2 equal buttons 2 equal discomfort 2 gt x rdf type Vicon MDProfile1 By using build in commands like equal x y values are compared to each other The right arrow defines the state if all axioms are true forward chaining Usually all variables starting with a must be defined first before the first comparison For instance to get the value of the predicate if the User Model suffers from arthritis the first axioms should be x rdf type Vicon UserModel x Vicon UserModelArthritis arthritis In the first axiom an instance of the User Model class is selected x The value of the arthritis predicate analogue other predicates can be set afterwards by using the direct predicate name always beginning with Vicon and the class name All stages of reasoning are compiled analogue to this scheme resulting in new inference models until the final Virtual User Model Using the knowledge base as initial model a new model is deducted including new classes for the separation of selection states While the knowledge
215. y Moderately difficult Rigidity No difficulty Mild difficulty Moderately difficult Table 2 1 GUIDE Manual Dexterity related k means Cluster Centres as a result of user survey Source Guide Consortium 2011 33 2 STATE OF THE ART Cognition TMT seconds 30 49 136 AVLT series 1 Short Term Memory 10 15 words 7 15 words 5 15 words trial 1 AVLT series 2 Short Term Memory 11 15 words 9 15 words 6 15 words trial 2 AVLT series S Short Term Memory 13 15 words 9 15 words 6 15 words trial 3 AVLT series 4 Short Term Memory 14 15 words 10 15 words 7 15 words trial 4 AVLT series 5 Short Term Memory 7 15 words 5 15 words 3 15 words trial 5 WAIS digit symbol test symbols 75 30 20 written in 2 minutes Table 2 2 GUIDE Cognition related k means Cluster Centres as a result of user survey Source Guide Consortium 201 1 Regarding cognitive tests different learning tests were executed During the AVLT Auditory verbal learning test 15 words had to be learned during 5 different trials After each trial participants were asked to recall as many words as possible In the WAIS digital symbol test of table 2 2 participants were asked to combine single characters from different rows with each other Each number from one row belongs to a character in the second row The final score presented in the table is the amount of character combinations written in 2 minutes Simulation Plat
216. y correctly interacting with off but ton i3 i6 i7 i9 111 112 i18 Successfully dialled number il i2 i3 i4 i5 i6 i7 i8 i9 ilO ill 113 il4 115 il6 i17 i18 i19 i20 gl g2 g4 g5 g6 87 g8 g9 gl0 gil g12 g13 g14 g15 g16 g17 g18 g19 g20 Successfully pressed green button to con nect call il i2 i4 i5 i6 i7 i8 i9 110 i11 il3 i14 il5 il6 il7 i18 il9 i20 gl g2 g4 g5 g6 g7 g8 g9 g10 g11 g12 g13 g14 g15 g16 g17 g18 g19 g20 Continued on next page 171 APPENDIX Issue Participants Failed to press green button to connect the call i3 Made a typing mistake with the numbers g2 83 g6 87 g8 g9 g12 g15 817 Pressed red button to connect call i8 g19 Pressing buttons harder than necessary ill il4 Pressed two buttons at the same time to con i14 nect the call fingers too large Confused by the A speed dial key gl0 Successfully identified that a call was com ing in il i2 i3 i4 i5 16 17 i8 i9 ilO ill 113 il4 115 il6 il7 118 i19 i20 gl g2 g4 go g6 87 g8 g9 g10 gil g12 g13 g14 g15 g16 g17 g18 g19 g20 Successfully pressed green button to answer the call il i2 i3 i4 i5 i6 i7 i8 i9 i10 ill 113 il4 115 116 i17 118 i19 i20 gl g2 g4 g5 g6 87 g8 g9 g10 gil g12 g13 g14 g15 g16 g17 g18 g19 g20 Pressed up
217. y pro viding recommendations to include end user needs VICON is a part of the VUMS cluster VUMS is a cluster that includes the projects VICON GUIDE MyUI and VERITAS All projects work on improving the accessibility of several products and application areas taking into account different impairments 2 6 1 Myul Project The MyUI Project Mainstreaming Accessibility through Synergistic User Modelling and Adaptability aims to create adaptive software user interfaces based on multi 1See http www veritas project eu vums Email Inbox 1 Before adaptation Permanent ac cess to user profile and user inter face profile via adaptation area bottom right 2 During adaptation Pulsing icon here chameleon indicates on going adaptation Email Inbox HA Christina a Cinema 24 09 2011 3 Dek 3 After adaptation The user can fa e300 undo the adaptation via button with A ESOS 2011 a curved backwards arrow L Christina Cinema Figure 2 8 Automatic adaptation with implicit confirmation Source Peissner et al 2012 28 2 STATE OF THE ART modal design patterns Peissner et al 2012 29 2 STATE OF THE ART A framework was implemented divided into 3 stages 1 Ul Parametrization In the first stage parameters and variables valid for the output Ul are defined Variables include e g the font size parameters e g the need for voice input The data used in this stage is derived from the
218. y to design a product as accessible and usable by an as wide range group of people as possible see Newell and Gregor 2000 This approach is also driven by the demographic change resulting from low birth rates and a higher life expectancy due to better medical treatment see 1 2 Regarding the design process and from realization perspective Personas based upon ethnographic user studies are used to describe beneficiaries Goodwin 2002 Figure 1 2 Beneficiary user with mobile phone Source RNID 2010 Figure 1 2 represents one main factor of inclusive design in general The person in the picture holds a mobile phone and derived from the facial expression of this person she is not sure what to do or how to perform a specific task This leads to the question of who is responsible for a proper interpretation of a product interface The designer who should be aware of the user or the user who needs to learn how to operate the product interface properly With respect to the demographic change see next section 1 2 this question be comes more important With more elderly users of product interfaces different user requirements must be considered while designing a product interface Also differ ent diseases of ageing especially hearing visual and manual dexterity impairments must be included in the product development process For instance do users with low 1 INTRODUCTION visual acuity have problems to find and press buttons if
219. y to start the JAR file The Sketch Design tool includes the following specific functionality e Device selection mobile phone washing machine or TV remote e Selection of User Model Persona typical environments and typical tasks e Output of textual recommendations e List of recommendations e Export current list of recommendations as RTF or PDF 80 4 APPLICATION IN DEVELOPMENT PROCESS 3 qIssod ose st uono l s ad nur y SYSEL Teord4j pue S JU9UIUOMAUJ E91d SPPON ISSN PAPS 0 qe SI ISN dy YI 39 uo UoTeoTdde U31S9 YPAS 3 p MSH suolqepu ujuuo2 w 84 PUNoy uo33nq 3x 3 3ll uolgeBlAeu 1934 O JEDIALU SS Jd 103309 1161p 1e10 Yoz ng pu3 42349 1031Nq pu3 SS3Jd Keydsip 1934 Yozng Ne 1934 uo3nnq 1 2 ss Jd ne eu MENTRE 4284 uoud Jo 31275 1934 Yozang JJO UO ss Jd dad se uado 414 Se vado ay n ep Jo 3u6n eye u s qe1 ay330 1 q163 ay a3NpaJ JOU p noys uoljeultuun7 i 1euJ 3ul aya 3nq paje un AeuJeqU 3g PINOYS say ay ATe pi q ym sKay JO UO JEUIWNT euJ 3u 1e Yauaq ajo ay Bu 3u6n quaiqwe 347 J 3Jep ul vo euIWNy euJj 3u Buey 07 aDeJuenpe 3u6ns e aq few ayayz Huyn qua que 34B11q u 3seJ3Uo2 ul JU335Ix3 UOU JO MO ag Aew Guyn y JE 3J8YM JO SUOIJIPUO Dulay n yesnjeu yep ul uul3 1Y JU 3e suolyIpuo32 BUAY N qua que Jamo u 5 X3 sK 3 JO UONEYILUNMN J2UJ33UI OF 1Y3U3Q Y AL 3u uu 13 uol32 S ON 345n4ep Jo 34517 n

An ontology-based approach to achieve inclusive design - E-LIB

Contents

Download Pdf Manuals

Related Search

Related Contents