Sample Planner for Attributes (TP105) User Manual
Contents
1. fo 01 Fraction of good items mis classified as defective b 0 045 Fraction of defective items mis classified as good You have to check the checkbox Make allowance for inspection error to make the rest of the inspection error area appear 1 Definitions 1a False positives good items that are incorrectly classified as defective a fraction false positives probability that a good item will be classified as defective 1b False negatives defective items that are incorrectly classified as good b fraction false goods the probability that a defective item will be classified as good 2 The relative magnitude of the effects of a and b 2a The effect of a There are usually so many good items in the first place that a high percentage of them mis classified as defective has a large effect 2b The effect of b File Manual02 doc H amp H Servicco Corp Printed 01 02 06 Page 28 User Manual Sample Planner for Attributes TP105 SamplingPlans com There are usually so few defectives in the first place that a high percentage of them mis classified as good doesn t have as large an effect 3 Example of manual calculations the program does this Example of the effect of a 0 1 when b 0 1096 of goods are classified as bad a 0 1 p 0 02 true fraction defective 1 0 p 1 0 0 02 0 98 true fraction good 0 1 X 0 98 0 098 fraction that are mis cl2. v Make allowance for inspection error You have to check the checkbox Make allowance for inspection error to make the rest of the inspection error area appear Increasing a decreases p L Phar p U They can become negative If p L lt 0 reduce the confidence coefficient CC If Pbar lt 0 beyond detection File Manual02 doc H amp H Servicco Corp Printed 01 02 06 Page 29 User Manual Sample Planner for Attributes TP105 SamplingPlans com Increasing a means that more of the good items in the sample are classified as defective Thus the confidence limits must be adjusted down to reflect the fewer defectives Increasing b increases p L Pbar p U They can become gt 1 If p U gt 1 reduce cc If Pbar gt 1 beyond detection Increasing b means that more of the defective item in the sample are classified as good Thus the confidence limits must be adjusted up to reflect the additional defectives Matching Variables ISL You can determine the variables ISL sampling plan that matches the producer s point and consumer s point of any attribute plan First design the attribute plan Then from the Design menu select Match a Variables ISL Plan to Existing Attribute Plan TP105 Sample Planner for Attributes Binomial Two Point ISL Calculations TE Set the options for Known SD or Unknown SD If known enter the historical within lot standard deviation Set the options for the type of ISL
3. 24 Report5e Sequential Graphs with Excel Step by Step 24 Report6 Analyze Sample Result confidence limit report 25 Additional Information srein 5 9B encierra p e tege E aaan 26 The Producer s and Consumer s Points sene 26 Understanding the OC 26 Possible Fractions Defective esee 27 How to account for the effect of Inspection 27 Matching Variables E cte RE dieat 29 Terms anid Symbols rne et e e tete rites ede 30 Assumptions and Limitations essere 31 Getting Help sues e HIS E p 31 How to activate this program s 1 31 How to Contact amp Servicco Corp 33 LALO b PRSETER 33 Glossary ee e TEE 33 H amp H Servicco Corp Printed 01 02 06 Page 1 User Manual Sample Planner for Attributes TP105 SamplingPlans com About TP105 Sample Planner for Attributes What does TP105 do TP105 calculates and or evaluates the sample size n and acceptance number Ac for acceptance sampling plans for attribute data The user specifies the sampling requirement by entering the Consumer s Point and a Producer s Point of the desired oc curve Decision Rules TP105 performs the calculations for fixed n and sequential decision rules
4. EEE ETT EN op i Continuei TT Loo TI apt a a i i Nin id A Accept Ls dik Ae ed ie LE 0 20 40 60 80 100 120 140 160 180 200 220 240 260 Number of Defectives HR 4 Onan CO Sample Number MMM Report5e Sequential Graphs with Excel Step by Step File Manual02 doc H amp H Servicco Corp Printed 01 02 06 Page 24 User Manual Sample Planner for Attributes TP105 SamplingPlans com SEQUENTIAL DECISION RULE COMPARE THE NUMBER OF Defectives TO AC AND N ACCEPT IF REJECT IF SAMPLE SIZE Defectives Defectives GROUPING OF ARE EQUAL ARE EQUAL TP1O5s TO OR LESS TO OR MORE FROM TO THAN AC THAN RE in om NP SEO N 5 N EN e ov 1 Oy Oc ES 9 NOTES x CANNOT ACCEPT CANNOT REJECT THIS SEQUENTIAL SAMPLING PLAN MATCHES THE FIXED n PLAN n 98 Ac 4 The software will organize the sequential table into three columns as shown in the example Report6 Analyze Sample Result confidence limit report File Manual02 doc H amp H Servicco Corp Printed 01 02 06 Page 25 User Manual Sample Planner for Attributes TP105 SamplingPlans com 1 Confidence Limit Report Servicco ProductName LotName ItemName DefectiveName ANALYSIS OF SAMPLE FOR THE FRACTION OF ItemNames DefectiveName 1 SUMMARY PRODUCT OR PROCESS ProductName POPULATION SIZE INFINITE 2 SAMPLE RESULT SAMPLE SIZE 98 Ite
5. Performance Curves TP105 evaluates the performance of Fixed n and Sequential sampling plans with performance curves OC ARL AOQ ASN and AFI curves Example of Input Names that specify the project process and lot definition and disposition OC curve points like Alpha 0 05 Beta 0 05 AQL 0 01 RQL 0 91 OC curve points can be input or output Fixed n decision rule like n 98 Ac 4 The decision rule can be input or output Sample results to evaluate with confidence limits like n 98 X 4 Examples of types of output reports Performance curve tables containing OC ARL AOQ ASN and AFI curves Performance graphs of OC ARL AOQ ASN and AFI curves Fixed n decision rule report Sequential decision rule table Sequential decision rule graph Confidence limit report one sided and two sided limits Quick Start For the latest information about the most recent version see the ReadMe txt file that was distributed with TP105 V3 How to Install TP105 v3 Save the Zip file Extract the EXE file from the Zip file Double click the EXE in Windows Explorer to start the install wizard Follow the instructions How to manage your license If the Design menu does not open then you will need to activate a license that you can obtain from H amp H Servicco Corp Your license s current status is shown in the Help About menu How to Uninstall TP105 v3 Click Start Settings Control Panel Add Remove Programs Uninstall TP105
6. 0 99984 1 2 3 0 007575 0 996648 4 10 0129 0 98576 5 0 01823 0 96164 1 0 02355 0 91941 7 3 0 9 0 02888 0 8546 0 0342 0 7672 amp 08 9 0 03953 0 6628 8 07 10 10 04485 0 5508 8 si 110 05018 0 4421 12 10 0555 0 3449 2 os 13 10 06083 0 2635 g 14 10 06615 0 1985 15 0 07148 0 1482 16 10 0768 0 1103 P 02 17 10 08213 0 08191 3s EN en a HHT HHT 20 0 0981 0 03383 0 01 0 02 0 03 0 04 0 05 0 06 0 07 0 08 0 09 01 011 0 12 21 10 1034 0 02532 True Lot Fraction Defective 22 0 1088 0 019 See the step by step details Report4c Performance Graphs with Excel Step by Step These steps will produce a performance curve graph using TP105 and the Excel chart wizard Produce the curve with TP105 Make a performance table containing the desired columns Example Use the program default sampling plan n 98 c 4 On tab3 of TP105 select an OC curve p stepping default Maximum p Minimum p and step size File Manual02 doc H amp H Servicco Corp Printed 01 02 06 Page 16 User Manual Sample Planner for Attributes TP105 SamplingPlans com Make the table with TP105 Not the graph Transfer the table to Excel With the mouse cursor select the rows of the table that contain curves in columns oo 0 m m the TP105 menu select Columns In an empty Excel sheet select the sheet by clicking on the square where the row headings and the
7. 3 0 How to Use TP105 v3 8 usage tips Design a sampling plan 1 Start TP105 To start designing a sampling plan select Design from the main menu Then Attribute Binomial File Manual02 doc H amp H Servicco Corp Printed 01 02 06 Page 2 User Manual Sample Planner for Attributes TP105 SamplingPlans com Sample Planner for Attributes File Edit View Design Tools Document Window Help Attribute Binomial Attribute with Inspection Error Match a Variables ISL Plan to Existing Attribute Plan A window will open with six Tabs across the top Select Tab 2 Design Plan You will see that the program has automatically added default input data for you to modify in order to design your own sampling plans ff TP105 Sample Planner for Attributes Binomial ioj x File Edit View Design Tools Document Window Help 1 Application 2 Design Plan 3 Performance Tables 4 Performance Graphe 5 Decision Rules 6 Analysis gt Two Points on the OC Curve Producer s Point Consumer s Point AQL p 02 R L n 091 Calculate AGL and ROL Alpha 0 05 Beta n 05 Calculate Alpha and Beta r Fixed n decision rule ne 38 c 4 Calculate n and Ac Ac n 0 040816 Maximum sample fraction defective to accept the lot Actual Alpha 0 047348 Actual Beta 0 049896 Input your consumer s and producer s points AQL RQL Alpha Beta or your decision rule n Ac Push the appropriate Calcul
8. 40 49 1 065 0 93937 0 09278 0 04533 0 004190 37 01 1 047 0 95484 0 09810 0 03383 0 003308 33 97 1 035 0 96628 0 1034 0 02532 0 002611 aL 31 1 026 0 97476 0 1088 0 01900 0 002061 23 98 1 019 0 98105 DESIGN FACTORS THAT INFLUENCE THE PERFORMANCE CURVES For all curves Lot size gt 10xn for best accuracy For AOQ Curve Rectify Yes Lot Size 10000 For ASN Curve Decision rule Sequential For AFI Curve Decision rule Sequential Rectify Yes Lotsize 10000 By program Sample Planner for Attributes TPLOS V3 00 on 09 26 2004 File Manual02 doc H amp H Servicco Corp Printed 01 02 06 Page 14 User Manual Sample Planner for Attributes TP105 The stepping variable always appears as the leftmost column in this case p SamplingPlans com The AOQL can be observed as the maximum of the AOQ curve In this case AOQL 0 02604 fraction defective Below the curves is an explanation of the assumptions and relationships that are affecting the performance characteristic curves The footer of the report documents the program version and run date Report4a Graphs of performance curves Tab4 enables you to view print or save text graphs of the OC AOQ ASN ARL AFI curves Compary Name and Division QUALITY CHARACTERISTIC FRACTION Defective Items PROGABILITY OC CURVE OPERATING CHARACTERISTIC CURVE Shows Performance of Sequential Sampling Plan ACCEPTANCE 1 0000 4 2 0 9000 0 8000 0 7000 0 60
9. As a Rule of Thumb the binomial distribution is very accurate when N n gt 10 That is when the lot is 10 times larger than the sample File Manual02 doc H amp H Servicco Corp Printed 01 02 06 Page 31 User Manual Sample Planner for Attributes TP105 SamplingPlans com Getting Help Context sensitive help help F1 Sampling plan help www samplingplans com How to activate this program s license TP105 will startup by presenting a form with drop down menus much like a word processor would All of the menus will work except for the possible exception of the Design menu If the Design menu does not open then you need to obtain a license from H amp H Servicco Corp If you have purchased the program you need to obtain a permanent license If you have a beta version or a trial version then you need to obtain a time limited license In either case you must send us an Installation Code Then we will send you a Liberation Key to activate the license The procedure to activate a license is as follows STEP 1 of 2 Send us an Installation Code 1 Use menu Help Manage your license 2 Enter yourname As you type your name an Installation Code will display in the textbox below your name 3 Copy the Installation Code to the Windows clipboard 4 Push the Close button to close the form 5 Use menu Help Internet Email to samplingplans com This will start your email program 6 Paste the Install
10. Corp Printed 01 02 06 Page 34 User Manual Sample Planner for Attributes TP105 SamplingPlans com AQL The Acceptable Quality Level AQL is the fraction defective of a lot that would have a high probability of acceptance AQL is often associated with 0 95 probability of acceptance Pa AQL and Alpha define the Producer s Point ARL The ARL Average Run Length is the average number of lots that will be accepted between until a rejection occurs The ARL curve of a sampling plan tells you for any lot quality p the average number of lots accepted until until there is a rejection ASN ASN Average Sample Number is the average number of sample units inspected per lot in reaching decisions to accept or reject The ASN curve is a plot of ASN versus true lot quality p attribute data Attribute data is go no go data like defective nondefective broken whole etc Attribute data is to be distinguished from variables data which is in the form of numerical measurements Beta The Consumer s Risk Beta is the risk of accepting a rejectable lot that is a lot that contains RQL fraction defective A typical value for the Beta risk is 0 05 beta version A beta version of a program is for testing of some feature and has a timed license C 0 A C 0 sampling plan is one in which the lot is only accepted if zero defectives occur in the sample consumer s point The Consumer s Point is the point on the oc curve defined by RQL o
11. Decision Rule Reports This Decision Rule tab is where you make the decision rule reports The fixed n and sequential plans have matched oc curves that is both have the same producer s and consumer s points File Manual02 doc H amp H Servicco Corp Printed 01 02 06 Page 11 User Manual Sample Planner for Attributes TP105 SamplingPlans com ffl TP105 Sample Planner for Attributes Binomial OF x File Edit View Design Tools Document Window Help 1 Application 2 Design Plan 3 Performance Tables B Analysis Sampling Plan Design from Tab 2 AQL 0 02 R L 0 091 Fixed 98 Alpha 0 05 Beta 0 05 Ac 4 i Fixed n Decision Rule Report n Ac Include Extra Statistical Details Make Fixed n Report r Sequential Decision Rule Report Truncation Sample Size 3 times ASNmax C 12timesn 229 Truncation Sample Size Make Table of n Organize the table by n sample number Sequential C X Organize the table by failure number Decision Rule 229 Truncation Sample Size Make Graph of 20 Y axis maximum Sequential 234 CAE mS EE Decision Rule Make a fixed n sampling plan report The fixed n decision rule consists of a sample size n and acceptance number Ac You can include the slope intercepts and other properties of the sequential plan that matches the oc curve of the fixed n plan by checking Include Extra Statistical Details For this exa
12. Performance Tables tab Tab3 makes tables of performance curves for the sampling plan that you designed on Tab2 TP105 calculates and displays up to five kinds of performance curves in the performance tables OC AOQ ASN ARL and AFI File Manual02 doc H amp H Servicco Corp Printed 01 02 06 Page 9 User Manual Sample Planner for Attributes TP105 SamplingPlans com ff TP105 Sample Planner for Attributes Binomial i OF x File Edit View Design Tools Document Window Help 1 Application 2 Design Plan 3 Performance T ables 4 Performance Graphe 5 Decision Rules 6 Analysis Sampling Plan Design from Tab 2 0 02 ROL 0 091 Fixed n 98 Alpha 0 05 Beta 0 05 4 Select the columns of the table Column of constant interval IV p Lot fraction defective e p 20Rows Pa Probability of Acceptance Pa 40 Rows Example ASN Average Sample Number 0 10875 ARL Average Run Length Maximum p 0 10343 00 Average Outgoing Quality 0 00225 0 03810 AFI Average Fraction Inspected m Meu Average Fraction Inspecte Minimum p 0 08745 0 08213 Sequential 0 00532 C Fixed n Step Size v Rectify Reiected Lots Infinite Lot Size Make a Table of the C Finite Lot Size Performance Curves The sampling plan itself is always displayed at the top of Tab3 and Tab4 and Tab5 To make a table with this Tab3 check the performance curves that you want to
13. TPLOS V3 0 6 on 01 31 2005 The decision rule matches those of standards Mil Std 105 and ANSI ASQ 21 4 This report completely documents the sampling plan the producer s point and the consumer s point the fixed n decision rule the date that it was designed and the software used and its version number If you check Include Extra Statistical Details on tab 5 Decision Rule the report would also display properties of the sequential plan whose oc curve matches this fixed n plan File Manual02 doc H amp H Servicco Corp Printed 01 02 06 Page 19 User Manual Sample Planner for Attributes TP105 SamplingPlans com The slope and intersept of the lines of the sequential chart match with those in the standard ISO 8422 Report5b Tables of a Sequential Decision Rule From the 5 Decision Rules tab you have the option of organizing tables of a sequential decision rule in two ways The two kinds of table are equivalent they are just organized differently The ntable is appropriate for most applications The first two columns are the beginning and end of groups of sample numbers n as you take them sequentially File Manual02 doc H amp H Servicco Corp Printed 01 02 06 Page 20 User Manual Sample Planner for Attributes TP105 SamplingPlans com 1 Sequential Decision Rule n table Servicco ProductName LotName ItemName DefectiveName SEQUENTIAL ATTRIBUTE SAMPLING PLAN ProductName FRACTION DefectiveName ItemNames PRODUCERS P
14. Testing A R A R 04 R A R R Accept PAD A 02 A A A OO t t tttttttt A tttttt4t rt rbj ttt ttt ttt ttttttt tt 1 39 78 117 156 195 234 SAMPLED ItemNames gt gt gt CRITERIA FOR SAMPLING ProductName FOR DefectiveNames PRODUCERS POINT CONSUMERS POINT MINIMUM AND MAXIMUM SAMPLE SIZES AQL fractn 0 02 R L fractn 0 091 MIN nTo A c 40 MAXIMUM N 229 DESIGN ALPHA 0 05 DESIGN BET 0 05 MIN nToRe 2 MAX ASN 76 THIS SEQUENTIAL SAMPLING PLAN MATCHES THE FIXED n PLAN n 98 Ac 4 LOT DEFINITION AND DISPOSITION Lot Definition LotName Disposition of accepted lots provide to customer Disposition of rejected lots scrap the lot By program Sample Planner for Attributes TP105 V3 0 6 on 01 31 2005 File Manual02 doc H amp H Servicco Corp Printed 01 02 06 Page 23 User Manual Sample Planner for Attributes TP105 SamplingPlans com You can make a better looking graph by selecting the rows of an ntable that contain just the columns of data Then Edit Copy columns to the Windows clipboard and paste them in to a graphic program Most spreadsheet programs can make graphs like MS Excel or the spreadsheet in Open Office etc Their Chart wizards enable you to tailor make graphs Report5d Sequential Graphs with Chart Wizard This sequential decision chart was made by exporting columns to Excel Then using the Excel chart wizard
15. be in the table for the sampling plan that you have designed In addition to the OC curve you may include the ASN ARL AOQ AND AFI curves Here is an example of the performance curve report Curves that require additional information or restrictions The AOQ curve reflects the results of rectification when you check Rectify Rejected Lots For non rectification plans the AOQ table will show that AOQ p The AFI curve requires that you specify a finite lot size The lot is the population N from which the sample is to be drawn Effects of decision rule type on the performance curves The sequential fixed n choice effects the ASN and AFI curves but not the OC ARL or AOQ curve Tab4 Make Graphs of Performance Curves The Performance Graph tab Tab4 is where you make graphs of the performance curves File Manual02 doc H amp H Servicco Corp Printed 01 02 06 Page 10 User Manual Sample Planner for Attributes TP105 SamplingPlans com Select a type of graph Choose one curve type at a time OC ASN ARL AOQ or AFI Select whether the decision rule is sequential or fixed n Choose whether rejected lots are to be rectified Enter the lot size Specify the limits of the Xaxis and Y axis Maximum Y axis the performance measure Minimum and maximum fraction defective p Step size of p along the Xaxis Recommended for most cases the program default limits scale the graph well Tab5 Make
16. item and characteristic This information does not have any effect on the decision rules that you will make the program calculate but it will be used to make the reports more meaningful You also can define what a lot is and specify what is to be done with a lot when it is accepted and when it is rejected The program will add default data to the Application tab at startup so you have the option to skip Tab1 until you finish designing the sampling plan and are ready to refine the documentation for printing You can add and save your own default data in the menu Tools Defaults Then whenever you open the design window the Application tab will be like the following File Manual02 doc H amp H Servicco Corp Printed 01 02 06 Page 7 User Manual Sample Planner for Attributes TP105 SamplingPlans com TP105 Sample Planner for Attributes Binomial H amp H Servicco Corp provide to customer scrap the lot 2 5 This Application tab opened containing user defaults Tab2 Design a Sampling Plan The Design Plan tab is where you design the sampling plan This is where you relate the producer s point the consumer s point and the fixed n decision rule File Manual02 doc amp Servicco Corp Printed 01 02 06 Page 8 User Manual Sample Planner for Attributes TP105 SamplingPlans com ffl TP105 Sample Planner for Attributes Binomial e File Edi View Design Tools Document
17. the design window File Manual02 doc amp Servicco Corp Printed 01 02 06 Page 6 User Manual Sample Planner for Attributes TP105 SamplingPlans com How to use TP105 Tab1 Application Description You start to design or analyze a sampling plan by bringing up the tabbed planning design window You do this by choosing Design from the top menu The screen shot below shows that the planning window contains six tabs across the top Each tab gives you access to one aspect of the planning process You can develop a sampling plan by selecting the tabs in sequence 1 2 3 4 5 6 The Application tab is currently shown ffl TP105 Sample Planner for Attributes Binomial ioj x d view Design Tools Document Window Help Application 2 Design Plan 3 Performance Tables 4 Performance Graphe 5 Decision Rules B2 amp nalysis General information Attribute Sampling Plan Design Project Department CompanyName Product Process optional Name Product or Process ProductName Name of the Sampled Item Name of Quality Characteristic DefectiveN ame Lot Definition and Disposition optional Lot Definition LotName Lot Disposition if Accepted provide to customer Lot Disposition if Rejected scrap the lot Show lot information on decision rule reports On the Application tab you can identify the sampling plan that you are developing by naming the company product
18. unit from 0 2 to 0 1 Change Minor unit from 0 04 to 0 05 Refinement change the scale of the p axis of the chart The p axis has increments of 0 02 Change them to 0 01 Select the p axis Right click select Format Axis Scale Tab Change Major unit from 0 02 to 0 01 Change Minor unit from 0 004 to 0 005 Refinement change the the backgroound color of the chart Select Plot Area Right click Format Plot Area example Border Select None Area color select white Use the mouse to change the width and height of the chart to your liking The chart should look like this File Manual02 doc H amp H Servicco Corp Printed 01 02 06 Page 18 User Manual Sample Planner for Attributes TP105 SamplingPlans com Report5a Fixed n Decision Rule Report This fixed n decision rule report was made from the tab 5 Decision Rule fl Fixed n Decision Rule Servicco ProductName LotName ItemName DefectiveName FIXED n ATTRIBUTE SAMPLING PLAN ProductName FRACTION DefectiveName ItemNames PRODUCERS POINT CONSUMERS POINT AQL fractn 0 02 R L fractn 0 091 DESIGN ALPHA 0 05 DESIGN BET 0 05 ACTUAL ALPHA 0 047 ACTUAL BETA 0 05 DECISION RULE n 98 ITEMS Sample Size AC 4 DEFECTIVES Acceptance Number RE 5 DEFECTIVES Rejection Number DEFINITION AND DISPOSITION Lot Definition LotName position of accepted lots provide to customer position of rejected lots scrap the lot program Sample Planner for Attributes
19. 00 0 5000 0 4000 0 3000 0 2000 0 1000 ttt t t A 0 0000 HHAH 3 H HH 4 H 44 Ht 9H 7 99 0 00225 0 02000 0 03775 0 05550 0 07325 0 09100 0 10875 ia peu saa iP niorir aoi gt gt gt LOT FRACTION Defective Items gt gt gt CRITERIA FOR SAHPLING Product FOR Defectives PRODUCERS POINT CONSUHERS POINT AQL fractnj 0 02 RQL fractn 0 091 DESIGN ALPHA 0 05 DESI GW BETA 0 05 THIS SEQUENTIAL SAHPLING PLAN HATCHES THE FIXED n PLAN n 98 Ac 4 By program Sample Plarmer for Attributes TP105 173 00 on 09 27 2004 File Manual02 doc H amp H Servicco Corp Printed 01 02 06 Page 15 User Manual Sample Planner for Attributes TP105 SamplingPlans com Here is more information about how this graph was made graphs of performance curves You can make a better looking graph by selecting the rows of a performance table that contains p Pa etc columns Then columns to the Windows clipboard and paste them in to a graphic program Most spreadsheet programs do can make graphs like MS Excel or the spreadsheet in Open Office etc Their Chart wizards enable you to tailor make graphs Report4b Performance Graphs with the Excel Chart Wizard If you have Microsoft Excel or equivalent you can copy a performance table into an Excel spreadsheet and use Excel s chart wizard to make a graph containing one or more curves p Pa 0 00225
20. 105 SamplingPlans com Producing Output Reports Report3a Table of Performance Curves The Performance Table report shows the effect of the true lot fraction defective p on the five performance characteristics curves OC AOQ ASN ARL AFI NO AOG ARL Company Name and Division Product PERFORMANCE OF SEQUENTIAL ATTRIBUTE SAMPLING PLAN WITH RECTIFICATION QUALITY CHARACTERISTIC FRACTION Defective Items PRODUCERS POINT CONSUMERS POINT ANL fractn 0 02 R L fractn 0 091 DESIGN ALPHA 0 05 DESIGN BETA 0 05 TRUE PROBABILITY AVERAGE AVERAGE AVERAGE AVERAGE FRACTION OF OUTGOING SAMPLE RUN FRACTION Defective ACCEPTANCE QUALITY NUMBER LENGTH INSPECTED ip Pa 00 ASN ARL AFI 0 002250 0 999840 0 002240 41 09 6 238 0 004269 0 007575 0 996648 0 007515 46 31 298 4 0 007367 0 01290 0 98576 0 01265 52 30 70 21 0 01940 0 01823 0 96184 0 01743 58 84 26 20 0 04382 0 02355 0 91941 0 02151 65 42 12 41 0 08661 0 02888 0 8546 0 02450 71 29 6 875 0 1515 0 03420 0 7672 0 02604 75 59 4 296 0 2386 0 03953 0 6628 0 02599 77 62 2 965 0 3424 0 04485 0 5508 0 02451 27 1l 2 226 0 4534 0 05018 0 4421 0 02202 74 34 1 792 0 5612 0 05550 0 3449 0 01901 69 96 1 526 0 6575 0 06033 0 2635 0 01592 64 71 1 358 0 7382 0 06615 0 1985 0 01305 59 20 1 248 0 8027 0 07148 0 1482 0 01054 53 86 1 174 0 8526 0 07680 0 1103 0 008427 48 90 1 124 0 8903 0 08213 0 08191 0 006697 44 44 1 089 0 91846 0 08745 0 06088 0 005302
21. 11 012 Lot Fraction Defective This example decision rule 98 Ac 4 was designed by specifying two points of the oc curve four numbers on the Design tab Producer s point AQL 0 02 Alpha 0 05 Consumer s point RQL 0 091 Beta 0 05 Every decision rule n Ac has its own oc curve which describes the sampling plan s probability of accepting lots having various fractions defective File Manual02 doc H amp H Servicco Corp Printed 01 02 06 Page 27 User Manual Sample Planner for Attributes TP105 SamplingPlans com Possible Fractions Defective TP105 does not use lot size to relate AQL and RQL to n and Ac So it can not automatically make AQL or RQL to any specific value of lot p For example with lot containing N items the possible fractions defective are 0 1 N 2 N 3 N Specifically if N 100 the possible numbers of defective items are defectives 0 1 2 3 So the true values of p are limited to p O N 0 00 1 N 0 01 2 N 0 02 3 N 0 03 Consequently it is possible to specify AQL and RAI that cannot be attained exactly How to account for the effect of Inspection Error To account for inspection error open the Design window with Attribute Binomial with Inspection Error On Tab2 the program can adjust the sample size and acceptance number for inspection error On Tab6 the program can adjust confidence limits for inspection error See later of this page On Tab2 C
22. 39 User Manual Sample Planner for Attributes TP105 SamplingPlans com File Manual02 doc amp H Servicco Corp Printed 01 02 06 Page 40
23. OINT CONSUMERS POINT AQL fractn 0 02 R L fractn 0 091 DESIGN ALPHA 0 05 DESIGN BETA 0 05 ACTUAL ALPHA 0 047 ACTUAL BETA 0 05 SEQUENTIAL DECISION RULE COMPARE THE NUMBER OF DefectiveNames TO AC AND RE iN ACCEPT IF REJECT IF SAMPLE SIZE DefectiveN DefectiveN GROUPING OF ARE EQUAL ARE EQUAL ItemNames TO OR LESS TO OR MORE FROM TO THAN THAN RE 1 2 3 og z 4 z4 3 25 39 4 40 45 4 46 60 5 61 66 i i 5 67 81 1 6 82 87 z 6 88 102 2 103 108 3 7 109 123 3 8 124 130 4 8 ISI 144 4 3 145 151 5 3 152 166 5 10 167 172 6 10 173 187 6 11 188 193 7 LE 194 208 12 209 226 8 12 27 227 3 12 228 228 10 12 229 229 11 12 NOTES CANNOT ACCEPT CANNOT REJECT THIS SEQUENTIAL SAMPLING PLAN MATCHES THE FIXED n PLAN n 98 Ac 4 LOT DEFINITION AND DISPOSITION Lot Definition LotName Disposition of accepted lots provide to customer Disposition of rejected lots scrap the lot By program Sample Planner for Attributes 105 V3 0 6 on 01 31 2005 In the Xtable below the first column is X the number of defectives in the sample so far File Manual02 doc H amp H Servicco Corp Printed 01 02 06 Page 21 User Manual Sample Planner for Attributes TP105 SamplingPlans com 2 Sequential Decision Rule X table u 2 Sequential Decision Rule X table CompanyName ProductName LotName ItemName DefectiveName SEQUENTIAL ATTRIBUTE SAMPLING PLAN ProductName FRACTION Defect
24. SL k is related to Cpk by the relationship Cpk k 3 Known SD Variables sampling plans have a known SD when the standard deviation is known in advance and has been determined to be in statistical control from lot to lot This as concluded from using an S Chart Past in control conditions cannot demonstrate that the current lot has not changed Therefore good practice is to test the current lot SD with a statistical F test or for small samples a statistical range test With the SD Known procedure the sample size is smaller than with the SD Unknown procedure license TP105 uses a licensing system based on RSA public key encryption and digital signatures See the current license status of this program with the menu Help About File Manual02 doc amp H Servicco Corp Printed 01 02 06 Page 36 User Manual Sample Planner for Attributes TP105 SamplingPlans com To change the status of the current license or to activate a new license for this program Help Manage your License Mil Std 105 Military Standard 105E Sampling Procedures and Tables for Inspection by Attributes This standard is based on AQL Now out of print it replaced by ANSI ASQ 21 4 n n Lower case n is the symbol for the sample size to be inspected N Upper case N is the symbol for the lot size Number of items in the lot ntable The ntable organizes a sequential decision rule into four columns 0 3 4 Sample Size n Accept Reject From To
25. Upper Lower or Both Click the calculate button The sample size the k value and the decision rule is displayed Unfinished Currently the program cannot send this plan to a printer except that you could do a screen capture with Alt PrtScr and paste it to a program like Paint File Manual02 doc amp H Servicco Corp Printed 01 02 06 Page 30 User Manual Sample Planner for Attributes TP105 SamplingPlans com Terms and Symbols Nomenclature Acceptance Sampling Terms and Symbols used by TP105 SYMBOL NAME Pa probability of acceptance of a lot p fraction defective of a lot AQL Acceptable Quality Level RQL Rejectable Quality Level Alpha risk of rejecting a lot if p ZAQL Beta risk of accepting a lot if p RQL n Sample Size Exact n Sample Size before rounding to whole number N Lot Size Ac Acceptance Number Same as C C Acceptance number same as Ac X The number of defective items in a sample Assumptions and Limitations Range of Input variables RQL 0 000001 to 0 999999 decimal fractions Beta 0 000001 to 0 999999 decimal fractions For Ac gt 0 option Ac 0 to 700 Binomial Distribution TP105 calculates the sample size using the binomial distribution The binomial does not use the lot size in calculating n As such the binomial is an approximation to the true distribution the hypergeometric It is a very accurate approximation and is commonly used when small samples are taken from large lots
26. User Manual File Manual02 doc Sample Planner for Attributes TP105 SamplingPlans com Sample Planner for Attributes TP105 User Manual About TP105 Sample Planner for 2 Mhat does TP105 d0 5a iem e e aic eee ni 2 Quick Start e ONE e a e Ee en REESE DUE 2 Menu ated t S eR Md ea 4 Howto use tt HO eR e e tH 6 Tabl Application 1 7 Tab2 Design a Sampling Plan essere 8 Tab3 Make Tables of Performance Curves 9 Tab4 Make Graphs of Performance Curves 10 Tab5 Make Decision Rule Reports 11 Tab6 Make a Confidence Limit 15 esses 12 Producing Output Reports sess enne 13 Report3a Table of Performance 13 Report4a Graphs of performance 15 Report4b Performance Graphs with the Excel Chart Wizard 16 Report4c Performance Graphs with Excel Step by 16 Report5a Fixed n Decision Rule Report sese 18 Report5b Tables of a Sequential Decision 20 Report5c Graph of Sequential Decision 22 Report5d Sequential Graphs with Chart 17
27. Window Help 3 Performance Tables 4 Performance Graphe 5 Decision Rules b Analysis Two Points on the OC Curve Producer s Point Consumer s Point AQL 0 02 R L fo 091 Calculate AQL and ROL Alpha n 0 05 Beta fo 05 Calculate Alpha and Beta r Fixed n decision rule ne 38 Ac 4 Calculate n and Ac Ac n 0 040816 Maximum sample fraction defective to accept the lot Actual Alpha 0 047348 Actual Beta 0 049896 Sampling Plan Calculations Depending on what information you start with you can perform three types of calculation with the Design Plan tab You specify the two points on the oc curve AQL Alpha RQL and Beta Then calculate the decision rule Calculate n and Ac You specify the fixed n decision rule n Ac and the AQL amp RQL Then calculate the risks Calculate Alpha and Beta risks You specify the fixed n decision rule n Ac and the Alpha amp Beta risks Then calculate AQL amp RQL Calculate AQL and RQL Additional information displayed e The Design Plan tab also displays the maximum fraction defective of the sample that allows acceptance of the lot Ac n The Design Plan tab also displays the actual Alpha and Beta risks which may differ from your chosen values due to the necessity to round n and Ac to whole numbers The Design Plan tab also transfers its sampling plan to tabs 3 4 and 5 Tab3 Make Tables of Performance Curves This
28. assified as bad Apparent p 0 02 0 098 0 118 a relatively large effect 3b Example of the effect of b 0 1 when a 0 10 of defectives are classified as good b 1 p 0 02 true fraction defective 0 1 X 0 02 0 002 fraction that are mis classified as good apparent p 0 02 0 002 0 018 a relatively small effect 4 The need to compensate for the effect of inspection error on the OC curve When n constant increasing a moves the oc curve to the left That is the false positives make the sample appear to have more defectives than it actually does So the additional apparent defectives make it less likely to accept a lot The program compensates for this when you calculate the n and Ac with inspection error When n constant increasing b moves the oc curve to the right That is the false good items make the sample appear to have fewer defectives than it actually does So the fewer apparent defectives make it more likely to accept a lot The program compensates for this when you calculate the n and Ac with inspection error 5 Effects of a amp b on confidence limits You can account for inspection error in Tab6 which will cause an adjustment to the sample confidence limits so that they reflect the fraction of items that are truly defective On Tab6 Analysis of Sample Results with Confidence Limits Sample Data Inspection Errar Sample Size n 135 0 01 Number Defective X b 0 045
29. ate button Generate performance reports 2 Make reports of the sampling plan performance curves from tabs 3 and 4 OC AOQ ARL ASN AFI curves Generate fixed n and sequential decision rules 3 Make reports of the sampling plan decision rules from tab 5 Fixed n Sequential Generate a confidence limit report for a lot 4 Enter values for sample data into tab 6 to make a confidence limit report one sided or two sided limits Handling multiple reports on the screen 5 When you have multiple reports on screen you can arrange them for easy examination Window Arrange For simplicity the right mouse button over a report will invoke the Window menu 6 Toggle any individual report between full size and normal size by double clicking on it File Manual02 doc H amp H Servicco Corp Printed 01 02 06 Page 3 User Manual Sample Planner for Attributes TP105 SamplingPlans com 7 For easier viewing increase the viewing area to full screen by maximizing the overall program to full screen Use click the upper right full screen symbol on the title bar Save time with user defaults 8 You can replace the startup default values such as Alpha Beta AQL RQL with your own values by Tools Defaults Save Updates to TP105 v3 will be free The main unfinished tasks are 1 Graphs are drawn with text characters to be replaced with line graphs or Excel charts 2 The help file is not finished 3 To add sampling plans
30. ation Code into the message area of the email 7 Add the title Send Liberation Key and add any appropriate message 8 Send the email to support samplingplans com STEP 2 of 2 We send you a Liberation Key to activate the license 9 You will receive back a Liberation Key We prepare the liberation key manually so if you are in a hurry phone us at 651 777 0152 to expedite the service 10 Copy the liberation key from our email to the Windows clipboard File Manual02 doc H amp H Servicco Corp Printed 01 02 06 Page 32 User Manual Sample Planner for Attributes TP105 SamplingPlans com 11 Use menu Help Manage your license and paste the Liberation Key into the textbox labeled Liberation Key 12 Push the Register button You can now use the Design menu to access all of the program s sampling plan design capabilities NOTE 1 If you want to use the program both at work and at home you may use the same liberation key to activate the program on both computers You name will always be on the program in Help About NOTE 2 If you want to use an alternative method of obtaining registering a license please contact us 651 777 0152 How to Contact amp Servicco Corp Website www samplingplans com Discussion forum www samplingplans com forum forum htm E mail service samplingplans com Phone 651 777 0152 Mail PO Box 9340 North St Paul MN 55109 0340 USA Author of TP105 Stan Hilliard Attr
31. atistical range test With the SD Known procedure the sample size is smaller than with the SD Unknown procedure SD Unknown Variables sampling plans have an unknown SD when the standard deviation is not known prior to taking the sample An estimate of the SD is then calculated from the sample and used in the accept reject decision These plans do not require that lots be in statistical control With the SD Unknown procedure the sample size is larger than with the SD Known procedure sequential A sequential sampling plan is a technique by which we build up our sample one item at a time and after inspecting each item ask ourselves Can we be sure enough to accept or reject this batch on the information so far collected Its value is in enabling reliable conclusions to be wrung from a minimum of data This was deemed sufficient to require that it be classified Restricted within the meaning of the Espionage Act during the war of 1939 45 sigma Sigma and SD are both used as symbols for the Standard Deviation TP105 File Manual02 doc amp H Servicco Corp Printed 01 02 06 Page 38 User Manual Sample Planner for Attributes TP105 SamplingPlans com TP105 is the short name of this software program Sample Planner for Attributes TP105 allows input of both consumer s point and producer s point or n Ac to calculate the Alpha and Beta risk or to calculate AQL and RQL TP105 also calculates in addition to the complete OC curv
32. column headings join On the Excel menu select edit paste or control V Create an OC graph with Chart Wizard Select the data example A1 B20 Select the chart wizard from the tool bar Step 1 of Chart Wizard Select Chart Type XY Scatter Select Chart sub type Scatter with data point connected by smoothed Lines without markers Note For sequential decision charts you would use chart sub type Scatter with data points connected by lines without markers Next gt File Manual02 doc amp H Servicco Corp Printed 01 02 06 Page 17 User Manual Sample Planner for Attributes TP105 SamplingPlans com Step 2 of Chart Wizard Option series in columns default Next gt Step 3 of Chart Wizard Chart Title Tab example OC Curve for n 98 C 4 Value X axis example True Lot Fraction Defective p Value Y axis example Probability of Acceptance Pa Gridlines Tab Example Value X axis check Minor Gridlines Example Value Y axis check Minor Gridlines Legend Tab uncheck Show legend Unless you are plotting more than one curve Next gt Step 4 of Chart Wizard Place Chart example As object in sheet Finish View the chart It should look pretty good Refinement change the scale of the Pa axis of the chart The Pa axis is 0 to 1 2 Change it to 0 to 1 0 Select the Pa axis Right click select Format Axis Scale Tab Change Maximum from 1 2 to 1 0 Change Major
33. e the AOQ ARL ASN and AFI curves TP105 also calculates for any n Ac the matched sequential sampling plan Or from two points on the oc curve both fixed n and sequential sampling plans trial version A trial version of the program is a means for a potential user to evaluate its sutibility for his her application A trial version is a full version but has a timed license Unknown SD Variables sampling plans have an unknown SD when the standard deviation is not known prior to taking the sample An estimate of the SD is then calculated from the sample and used in the accept reject decision These plans do not require that lots be in statistical control With the SD Unknown procedure the sample size is larger than with the SD Known procedure variables ISL sampling plan A variables sampling plan uses a decision rule calculated from numerical measurements of items in a sample An ISL variables sampling plan has specifications that define the conformance of the individual items X X is a symbol for the actual number of defectives in a sample You would reject a lot if X is greater than Ac and accept the lot if X is equal to or less than Ac Xtable The Xtable organizes a sequential decision rule into three columns Number of Rejectif Accept if Defectives n lt Rn n gt An Design menu The Design menu opens the window where you to design sampling plans File Manual02 doc amp H Servicco Corp Printed 01 02 06 Page
34. for defects based on the poisson distribution How to get help or to inform us of problems or bugs Email service samplingplans com Telephone 651 777 0152 Website www samplingplans com Menu Bar TP1 05 menu bar LE zaen planner for Attributes The File menu will print a copy of the report that has the focus to the default printer copy it to the Windows clipboard or exit the program The edit menu will modify reports G Sample hanner for Cut will remove the selected area and store it on the Windows clipboard for further processing with other software such as word processors File Manual02 doc H amp H Servicco Corp Printed 01 02 06 Page 4 User Manual Sample Planner for Attributes TP105 SamplingPlans com Copy will copy the selected area to the Windows clipboard but will not remove it from the report Copy Columns Use this option to transfer data to other software to make graphs charts It will copy selected columns of data from performance tables and sequential decision tables to the Windows clipboard For further details see the instructions on making use of the Excel chart wizard for performance graphs and sequential decision graphs Delete will delete remove one or more report The View menu controls the menu bar and the status bar Sample Planner for Attributes fie Edt View Design Tools Do dow Heb The Design menu is the main starting point to design a sampling plan You
35. ibutions TP105 uses the following products VB Helpwriter Teletech Systems To design this Help system ActiveLock To design the digital signature based copy authentication system Visual Basic V6 Microsoft To design and compile the program code Glossary Ac The Acceptance Number is the maximum allowable number of defective items in the sample and still accept the File Manual02 doc amp H Servicco Corp Printed 01 02 06 Page 33 User Manual Sample Planner for Attributes TP105 SamplingPlans com lot If the number of defectives in the sample exceeds Ac then the lot is rejected The symbols Ac and C are commonly used to represent the acceptance number acceptance number The acceptance number symbol Ac or C is the maximum allowable number of defectives in the sample for a lot to be accepted For the acceptance number of zero the lot is rejected when a sample contains one or more defective items acceptance sampling Acceptance sampling has as its purpose to decide whether to accept or reject a population based on the results of the inspection of samples taken from that population In manufacturing applications of acceptance sampling the population is typically incoming raw material lots in process sublots and finished product lots Other areas involve administrative processes inventory etc activate If you have purchased the program you need to activate a permanent license If you have a beta version o
36. iveName ItemNames PRODUCERS POINT CONSUMERS POINT AQL ifractn 0 02 R L fractn 0 091 DESIGN ALPHA 0 05 DESIGN BETA 0 05 ACTUAL ALPHA 0 047 ACTUAL BETA 0 05 SEQUENTIAL DECISION RULE COMPARE THE SAMPLE SIZE TO En AND REJECT IF ACCEPT IF SAMPLE SIZE SAMPLE SIZE NUMBER OF IS EQUAL TO IS EQUAL TO DefectiveNames OR LESS OR GREATER X THAN Pr THAN An B m 40 l NN 61 2 2 24 103 4 45 124 5 66 145 6 87 167 108 188 8 130 209 9 151 229 10 172 229 11 133 229 12 229 NOTES CANNOT ACCEPT THIS SEQUENTIAL SAMPLING PLAN MATCHES THE FIXED n PLAN n 98 4 LOT DEFINITION AND DISPOSITION Lot Definition LotName Disposition of accepted lots provide to customer Disposition of rejected lots scrap the lot By program Sample Planner for Attributes 105 V3 0 6 on 02 22 2005 Report5c Graph of Sequential Decision Rules From the tab 5 Decision Rules you can make graphs of sequential decision rules File Manual02 doc H amp H Servicco Corp Printed 01 02 06 Page 22 User Manual Sample Planner for Attributes TP105 SamplingPlans com 1 Sequential Decision Rule Graph Servicco ProductName LotName ItemName DefectiveName ProductName FRACTION DefectiveName ItemNames NUMBER SEQUENTIAL DECISION RULE DIAGRAM OF DefectiveNames 20 28 16 l4 12 RRRRR R z R A R A 10 R Reject R R A R 08 R R R R A 06 R A 8 R Continue
37. mNames RESULT 4 DefectiveNames POINT ESTIMATE 4 98 0 040816 FRACTION DefectiveName 3 INTERVAL ESTIMATE POPULATION FRACTION OF ItemNames DefectiveName COULD BE AS HIGH AS 0 101218 UPPER CONFIDENCE LIMIT BEST ESTIMATE 0 040816 SAMPLE AVERAGE COULD BE AS LOW AS 0 011230 LOWER CONFIDENCE LIMIT 4 THE BASIS OF THESE LIMITS IS THE EXACT BINOMIAL DISTRIBUTION B TWO SIDED 95 CONFIDENCE LIMITS By program Sample Planner for Attributes TP105 V3 0 6 on 01 31 2005 File Manual02 doc H amp H Servicco Corp Printed 01 02 06 Page 26 User Manual Sample Planner for Attributes TP105 SamplingPlans com Additional Information The Producer s and Consumer s Points The producer s and consumer s points on an oc curve are Producer s point AQL Alpha The producers point protects against rejecting good p AQL lots Consumer s point RQL Beta The consumer s point protects against accepting off grade p lots TP105 calculates from tab3 and tab4 all the points on the one and only oc curve that goes through those two points TP105 calculates from tab2 the one and only fixed n decision rule that has its oc curve going through those two points Understanding the OC Curve Using the OC Curve Plot of Pa versus p This oc curve describes the performance of a decision rule For this example the decision rule is n 98 Ac 4 OC Curve Fa 001 0 02 0 03 0 04 0 05 0 06 0 07 0 08 009 01 0
38. mple compare the matched fixed n and sequential plans by comparint the n 98 of the fixed plan to the sequential s Minimum n to Reject 2 Minimum n to Accept 40 Maximum ASN 76 Truncate a sequential sampling plan report The sample size of sequential plans do not have a maximum even though random sampling variability causes the plans to always stop and accept or reject The program offers three ways to truncate a plan e Truncate n at 3 times the maximum of the average sample number ASN curve e Truncate n at 1 2 time fixed n e Manually enter the truncation n Make a table of the sequential decision rule The program can organize the table by sample number or defect number Make a graph of the sequential decision rule The program calculates the maximum Y axis and Xaxis but you can change them manually Tab6 Make a Confidence Limit Analysis Tab6 is wher you analyze a sample result to make a confidence statement File Manual02 doc H amp H Servicco Corp Printed 01 02 06 Page 12 User Manual Sample Planner for Attributes TP105 SamplingPlans com 3 PRCCHAS 77 ES TP105 Sample Planner for Attributes Binomial e Enter the sample data n and X Choose the type of confidence limit s one sided upper or lower or two sided Choose the confidence coefficient percent confidence File Manual02 doc amp H Servicco Corp Printed 01 02 06 Page 13 User Manual Sample Planner for Attributes TP
39. n the X axis and Beta on the Y axis Example beta 0 05 probability of accepting a lot if its fraction defective is RQL 0 10 10 Cpk Cpk is a quality index Cpk UISL Xbar 3 SD or Xbar LISL 3 SD whichever is smaller discriminating The ability of a sampling plan to discriminate tell the difference between an AQL lot and RQL lot A sampling plan can discriminate well when its AQL and RQL are sufficiently close together If one sampling plan is more discriminating than another its oc curve is steeper more verticle exact n The exact sample size shows the value of n before rounding to a whole number that exactly satisfies the consumer s point that you entered The exact n can be useful to know when the sample size is small and the direction of rounding can have economic consequences False negatives A false negative is an inspected item which is defective but is incorrectly classified as a good item due to inspection error File Manual02 doc H amp H Servicco Corp Printed 01 02 06 Page 35 User Manual Sample Planner for Attributes TP105 SamplingPlans com The symbol b represents the probability of making such a wrong classification calling a defective good False positives A false positive is an inspected item which is of good quality but is incorrectly classified as a defective due to inspection error The symbol a represents the probability of making such a wrong classification calling a g
40. on Ac on Re OC The oc curve Operating Characteristic Curve of a sampling plan is a graph of Pa versus p where Pa probability of acceptance p true lot fraction defective of a lot The oc curve tells you how the sampling plan will perform in making accept reject decisions oc curve The oc curve Operating Characteristic Curve of a sampling plan is a graph of Pa versus p where Pa probability of acceptance p true lot fraction defective of a lot The oc curve tells you how the sampling plan will perform in making accept reject decisions p is the symbol for the true fraction of items in a lot that are defective p L p L is a symbol for the lower confidence limit p U p U is a symbol for the upper confidence limit Pa Pa is the probability that a sampling plan will accept a lot Pa depends on the decision rule n C and the lot s fraction defective p Pbar Pbar is a symbol for the sample fraction defective population The population is the decision unit from which samples are taken for inspection In manufacturing applications of acceptance sampling the population is typically incoming raw material lots in process sublots and finished product lots probability statement The probability of a type II error File Manual02 doc H amp H Servicco Corp Printed 01 02 06 Page 37 User Manual Sample Planner for Attributes TP105 SamplingPlans com The probability of accepting a lot if it contains RQL fracti
41. on defective The probability of a type error The probability of rejecting a lot if it contains AQL fraction defective producer s point The producers point of the OC Curve is defined by AQL and Alpha rectification To rectify a rejected lot is to sort 100 and replace rework the defective items with good ones Rectify To rectify a lot is to sort it 100 and fix or replace all defectives A rectification plan specifies that all lots that the sampling plan rejects be rectified register See the topic License Management in the menu Help Program Help Contents RQL The Rejectable Quality Level RQL is the fraction defective of a lot that would have a low probability of acceptance RQL and Beta define the consumer s Point of the oc curve Other names that have the identical meaning as RQL are LQ and LTPD sample size The sample size is number of items drawn from the lot for inspection The symbol for sample size is n whereas the symbol for the lot size is N SD SD and Sigma are both used as symbols for the Standard Deviation SD Known Variables sampling plans have a known SD when the standard deviation is known in advance and has been determined to be in statistical control from lot to lot This as concluded from using an S Chart Past in control conditions cannot demonstrate that the current lot has not changed Therefore good practice is to test the current lot SD with a statistical F test or for small samples a st
42. ood item defective fixed n A fixed n sampling plan has a decision rule that specifies the same sample size n for each lot This differs from a sequential decision rule for which the sample size varies from lot to lotdepending on lot qualtiy p and random variation Inspection Error With attribute data inspection error is the misclassification of individual items good items as defective and defective items as good The error of misclassification of items is similar to the error of mis dispositioning lots Similarities Whereas alpha is the probability of rejecting an AQL lot a is the probability of classifying a good item as defective Whereas beta is the probability of accepting an RQL lot b is the probability of classifying a defective as a good item ISL ISL is an acronym for Individual Specification Limit ISLs defines when a measured individual item is defective on the high side UISL the low side LISL of both With measured data ISLs are used to calculate the a variables sampling plan s decision rules calculate Cpk and with rectification plans to sort nonconforming units from rejected lots ISO 8422 International Organization for Standardization 8422 Sequential sampling plans for inspection by attributes k value The k value is a decision limit used in variables ISL sampling plans to make the accept reject decision k is the number of SD s between the average of a sample and the nearest I
43. r a trial version then you need to activate a time limited license In either case you must send us H amp H Servicco Corp an Installation Code Then we will send you a Liberation Key that activates the license To activate a license Menu Help Manage Your License For see detailed instructions in activating a license for this program Help Program Help Contents License Management LicenseManagement txt AFI AFI Average Fraction Inspected is the fraction of items out of a series of lots that will be inspected by the plan For non rectifying fixed n plans AFI n N For non rectifying sequential plans AFI ASN N Plans that rectify rejected lots can make AFI increase as p increases because of 100 inspection of rejected lots Alpha The Producer s Risk Alpha is the risk of rejecting a good lot that is a lot that contains AQL fraction defective ANSI ASQ 21 4 This is the current AQL based standard Sampling Procedures and Tables for Inspection by Attributes AOQ AOQ is the Average Outgoing Quality the average fraction defective of lots outgoing to customers Useful to evaluate sampling plan applications that rectify rejected lots The AOQ Curve is the AOQ as a function of various fixed values of p AOQL AOQL Average Outgoing Quality Limit is the maximum AOQ over all possible values of incoming product quality p for a given acceptance sampling plan File Manual02 doc amp H Servicco
44. typically choose the sub menu Attribute Binomial TE Sample Planner for Attributes File Edit View Tools Document Window Help Attribute with Inspection Error Match a Variables ISL Plan to Existing Attribute Plan The feature Attribute with Inspection Error is explained here The feature Match a Variables ISL Plan to the Existing Attribute Plan is explained here The Tools menu provide control of user defaults ffi Sample Planner for Attributes File Edi Tools Document Window The Document menu creates a new note document Sample Planner for Attributes dit View Design Tools Do Window The Window menu controls the windows on the interface The Show option will display reports if they are hidden under the design form window Window Help Arrange Reports Best fit 5 how The Close option remove all reports from the screen or will remove the design form File Manual02 doc H amp H Servicco Corp Printed 01 02 06 Page 5 User Manual Sample Planner for Attributes TP105 SamplingPlans com Right Click menu you can also invoke the Windows menu and sub menus by right clicking report or on the design form The Help menu provides help in using TP105 L3 Sample Planner for Attributes The Internet sub menu of Help The Right mouse button pop up menu is the same as the Window menu You can right click over any sampling plan report window or over
Download Pdf Manuals
Related Search