John Zorich, MS CQE


 Chad Michael Cannon
 7 months ago
 Views:
Transcription
1 Used In Nonclinical Inspection & Test In the Pharma & Meddevice Industries John Zorich, MS CQE Outline of topics: Regulatory Requirements Basic Vocabulary Confidence Intervals Statistical Process Control (SPC) Confidence / Reliability Calculations Process Capability Indices Mean Time Between Failures Tests of Statistical Significance QC Sampling Plans Examples of Valid Statistical Rationale statements 1
2 Regulatory Requirements 21CFR211 (2016) Stability Testing...shall include...sample size... based on statistical criteria... FDA s Guidance for...process Validation (2011) The number of samples should be adequate to provide sufficient statistical confidence of quality both within a batch and between batches. The confidence level selected can be based on risk analysis as it relates to the particular attribute under examination. ICH Harmonised Tripartite Guideline Quality Risk Management, Q9 (2005) The following statistical tools are recommended to support and facilitate quality risk management : [statistical process] control charts process capability analysis Regulatory Requirements FDA's "GMP" (21CFR ) Sampling plans...shall be...based on a valid statistical rationale... FDA's "Medical Device Quality Systems Manual" "...all sampling plans have a builtin risk of accepting a bad lot. This sampling risk is typically determined... by...the 'operating characteristic [OC] curve [which]...can be used to determine the risk a sampling plan presents... A manufacturer shall be prepared to demonstrate the statistical rationale for any sampling plan used. ISO 13485: , Design & Development Verification 7.3.7, Design & Development Validation Records of the results and conclusions...including, as appropriate, rationale for sample size...shall be maintained. 2
3 is a mathematical summary value calculated from data taken from a Sample. All of the following are statistics: Avg thickness of every 100th cable produced last week. Range of thicknesses in that sample Median thickness in that sample. is a mathematical summary value calculated from data taken from the entire Population; that is, every data point in the entire population (e.g., average thickness of all cables produced last week). "Statistics" as a science is the mathematical analysis of "statistics", not of parameters. Statistics is the science of using "statistics" to guesstimate "parameters". is part of a The purpose of obtaining information about the sample statistic (e.g., Average, Std Deviation, or % inspec) is that we trust that it will give us information about the population parameter (which is what we really care about). 3
4 (they are the basis of most statistical methods) 95% Confidence Interval ( & Limits) Sample Mean A 95% confidence interval is an interval around the observed mean of a sample, in which interval you can expect (95% of the time) to find the true mean (= parameter) of the population from which the sample was taken. Confidence Limits are the values on either end of the interval. 4
5 Where is the true mean ( = the Parameter Mean)? Answer: Somewhere in Confidence Interval (at the chosen confidence level, typically 95%). Sample Mean Confidence interval based on a LARGE samplesize Confidence interval based on a MEDIUM samplesize Confidence interval based on a SMALL samplesize The choice of sample size is arbitrary, based upon how narrow you want the confidence interval ( i.e. how accurately you want to know the parameter). Because confidence limits are automatically adjusted based upon sample size, ANY sample size is valid!! Confidence Limits for Sample Average Standard Error of the (sample) Mean ( estimated from 1 sample ) Sample Standard Deviation Sample Size. 95% confidence interval for the Population Mean can be estimated using this equation: Sample Average + / " t " x (Std Error of Mean) 5
6 This table provides the t used to calculate Confidence Intervals; for 2sided, use A = 1 Confidence For 1sided confidence interval limits, use A = 2 x ( 1 Confidence) " v " or "d.f. (= degrees of freedom ) is commonly equal to the denominator in the calculation of the relevant standard deviation (e.g., the classic sample standard deviation has n 1 = Sample_Size 1 in its denominator. Confidence Limits for Sample Proportions, % Over a dozen different methods exist for calculating confidence limits for sample % defective  each of those methods gives a different length & different conf. limits!!! The classic method is called the Exact binomial  it can be calculated via Excel's "Beta" function  for example: UPPER 2tailed Confidence Limit =beta.inv( 1 (1 C ) / 2, k + 1, N k ) LOWER 2tailed Confidence Limit =beta.inv((1 C ) / 2, k, N k + 1 ) C = Confidence N = Sample size (e.g., 100 ) k = observed number of items of interest, e.g. defectives) 95% conf. limits for observed 10 defects in sample size = 100 beta.inv( 1 (1 0.95) / 2, , ) = = 17.6% beta.inv( (1 0.95) / 2, 10, ) = = 4.9% 6
7 Onesided confidence limits are calculated like so: UPPER 1tailed Confidence Limit =beta.inv( C, k + 1, N k ) LOWER 1tailed Confidence Limit =beta.inv( 1 C, k, N k + 1 ) 95% 1sided limits for observed 10 defects when N = 100 upper = beta.inv( 0.95, , ) = lower = beta.inv( , 10, ) =
8 Harold F. Dodge: He helped invent SPC when he worked in the QA department at Bell Laboratories from 1917 to Of his early experiences Dodge wrote: Initially, the basic procedures for variables called for samples of four, with one [SPC] chart for the average, and another for the standard deviation... We proposed shop use of samples of five instead of four; it is easier to divide by five than by four." That text is from: SPC Xbar charts with normally distributed data (ANY sample size is valid because the Control Limits are simply 2sided 99% confidence limits on current process mean) 125 n = 2 n = 4 n =
9 where Reliability means % inspecification (these are a type of process capability calculation) Sample Sizes Because the output of reliability/confidence calculation is a lower 1sided confidence limit... Any sample size is valid, because the smaller the sample size, the farther away from the observed %inspec is the %inspec that you can claim. 90% 100% Confidence limit based upon small sample Confidence limit base upon large sample % reliability observed in sample 9
10 ATTRIBUTE DATA: (i.e., Pass/Fail testing, not measurements results) Reliability = beta.inv ( 1 C, N F, F + 1 ) where... beta.inv = Microsoft Excel function C = Confidence desired (expressed as a decimal fraction) N = sample size F = # of failures seen in the sample That formula outputs the lower 1tailed "exact" binomial confidence limit on % inspecification observed in the sample. If no failures in a sample of 299, then 95% confidence in... =beta.inv( , 299 0, ) = 0.99 = 99% reliability If 2 failures in a sample size of 30, then 95% confidence in... =beta.inv( , 30 2, ) = 0.80 = 80% reliability Why does a sample of 299, with zero failures, equal 95% confidence of at least 99% reliability? A reliability calculation on a binomial proportion is, in effect, a lower 1sided confidence limit on the observed proportion. It's the lowermost edge of the interval in which we predict we will find the true ("parameter") proportion. We are 95% sure that the Parameter is somewhere in this interval 98% 99% Lower 1tailed 95% Confidence Limit on Sample Statistic, when N = 299 and no failures are found in sample. 100% For reliability, we get to claim the worst value in that interval (in this case, 99%) Sample Statistic 10
11 VARIABLES DATA Normally Distributed (or transformed to Normality) Kfactor table a.k.a., Normal Tolerance Factor table, or... Statistical Tolerance Factor table Procedure is: Calculate the Observed K Compare the Observed K to the K in the Ktable. Observed K = number of Std Deviations that the Process Mean is from nearest side of a 1 or 2sided specification, i.e., (SmplAvg NearestSpecLimit) SmplStdDev Juran's QH confidence reliability K If the "observed K" is at least 3.520, & the population is "Normally Distributed, & sample size is 15, then we are 95% confident that the Lot from which the sample came has at least 99% inspec parts. 11
12 Why does a sample of 15, with whose average is 3.52 std dev s away from a 1sided QC spec, equal 95% confidence of at least 99% reliability? A reliability calculation on a sample from a Normal population is, in effect, a lower 1sided confidence limit on the observed % inspec. We are 95% sure that the Parameter % inspec is somewhere in this interval 98% 99% 99.98% For reliability, we get to claim the worst value in that interval (in this case, 99%) When sample Avg is 3.52 stdevs from the onesided spec, can claim 99% reliability at 95% confidence. Observed Sample Statistic: =NORMSDIST(3.52) = % inspec but no confidence can be claimed for this statement! (requires normally distributed data, or data that has been transformed to normality) 12
13 Capability Indices = ( NSL ProcessAvg ) (3 x StdDev) where: NSL = QC specification closest to Process Average ProcessAvg = average of the sample data being analyzed StdDev = estimate of the standard deviation of the population from which the sample was taken (based on data from all samples being analyzed for Cpk) When Cpk alone is reported to a regulatory agency, the conclusions based upon the Cpk are suspect, if a small sample size was used, because... a Cpk is just a statistic  where s the parameter? = L C L (if report LCL, then...any sample size is valid) for a lower 1sided 95% limit, replace this with... normsinv(0.95) where normsinv is an Excel function Cpk without confidence Cpk with confidence (for 1 and 2sided QC specs) n in this formula refers to total # of raw data points combined from all samples in data set being evaluated If Cpk = 1.33, then lower 1sided 95% confidence limit is for n = 100 and 1.00 for n = 25 13
14 (e.g., for capital equipment) MTBF = Mean Time Between Failures Device Failure indicated by X on line Hours in service # 1 X X 500 # 2 X 500 # # 4 X X 500 After failure, device is quickly repaired & put back into service. MTBF = ( ) / 5 failures = 400 hours But that s a statistic  where s the parameter? When reporting a statistic, impossible to justify sample size. 14
15 LOWER 1SIDED CONFIDENCE LIMITS for MTBF from studies carried out for a predetermined time period: Any sample size is valid, because using Confidence Limits: 2 x T Lower Conf. Limit = Chiinv( 1 Conf, ( 2 x F ) + 2 ) Where T = Total inservice test time (all devices combined; = LengthOfStudy x NumberOfDevicesInStudy For this calculation, these are identical: T = 500 hours x 4 devices T = 50 hours x 40 devices T = 5 hours x 400 devices Chiinv = the Excel function Conf = desired confidence (as a decimal fraction) F = Total number of failures (all devices combined) e.g., ( 2 x 2000 ) / Chiinv ( , ( 2 x 5 ) + 2 ) = 190 hours 15
16 Confidence Interval explanation of ttests (mathematically identical to classic ttests) Sample Value Null Hypothesis Value 95% conf. interval (large sample) The fact that the Null Hypothesis is outside the confidence interval means the ttest had a significant result. It is much more difficult to obtain a significant result with a small sample (= large conf. interval) than with a large sample (= small conf. interval), and therefore any sample size is valid when claiming significance. Confidence Interval explanation of ttests Sample Value Null Hypothesis Value 95% conf. interval (small sample) The fact that the Null Hypothesis is NOT outside the confidence interval means the ttest had a nonsignificant result. It is much easier to obtain a nonsignificant result with a small sample (= large conf. interval) than with a large sample (= small conf. interval), and therefore... not all sample sizes are valid when claiming nonsignificance unless justified based upon the classic method of Power. 16
17 Testing for (practical) Noninferiority ( use this instead of nonsignificance and Power ) WORSE performance C B(yours) A (other) BETTER performance Assume that A is the other company s product and that B is your company s product, and that C is the first value lower than A that is considered worse than A in a practical sense). If B is statistically significantly larger than C, then you can claim that your product B is noninferior (substantially equivalent?) to product A, in a practical sense, irrespective of whether B is statistically different from A. Therefore...Any sample size is valid, for claiming noninferiority, IF it shows a significant difference between B and C 17
18 The %AQL of an AQL sampling plan is the product quality ( = lots having that % defective) that the sampling plan will accept (=approve) almost all the time. The %LQL of an LQL (a.k.a., LTPD, RQL, or UQL) sampling plan is the product quality ( = lots having that % defective) that the sampling plan will reject almost all the time. Predicting Pass Rates 100% 80% 60% % Defective OC curve for a 4% AQL sampling plan (ANSI Z1.4) 40% 20% 0% 0% 10% 20% Lot % Defective N = 1000 n = 80 c = 7 Sample Size is mandated by Sampling Plan booklet, based upon Lot Size. 18
19 100% 80% 60% % Defective OC curve for a 4% AQL, C=0 sampling plan (Squeglia's 4 & 5th ed.) 40% 20% 0% 0% 10% 20% Lot % Defective N = 1000 n = 15 c = 0 Sample Size is mandated by Sampling Plan booklet, based upon Lot Size. Previous slides, combined: 100% 80% 60% Previous 2 slides combined Z1.4 (4% AQL) ASQCZ1.4, 4% AQL ASQCC=0, 4% AQL C=0 ( 4% AQL ) 40% 20% 0% 0% 5% 10% 15% 20% Lot % Defective 14% LQL (choose sampling plans that have LQL = LTPD that support Risk Management statements) 19
20 Are LQL levels of AQL plans consistent? 100% 80% 60% 40% 20% 0% 0% 10% 20% Lot % Defective Lot Size = 1000 Lot Size = 500 Lot Size = 100 ASQCZ1.4, general, II, single, normal, 4% AQL After making choice on previous slide, be sure to determine actual LQL for planned Lot Size; adjust AQL as needed to achieve desired LQL. 20
21 CONFIDENCE INTERVALS We are 99% confident that the Population average is not smaller than 2.35 and not larger than 3.16 (where those 2 values are the lower and upper limits of the 2sided 99% confidence interval calculated based upon the Sample data). We used a sample size of 20, which is valid because this result was based upon a calculation of confidence interval limits (any sample size is valid when using confidence limits) Sample Avg CONFIDENCE / RELIABILITY STATEMENTS We are 95% confident that the Population from which the Sample was taken has an inspecification% (i.e., a %Reliability ) that is not less than 99.9%. We used a sample size of 8, which is valid because we used Ktables for our calculation, which are based on confidence limits, and therefore any sample size is valid. 5.7 Standard Deviations between QC Spec and Sample Average Sample Avg 42 21
22 QC SAMPLING PLANS We are using a formal published Sampling Plan booklet purchased from the American Society for Quality (ASQ). Sample size is a function of lot size, as explicitly required by the instructions in the Sampling Plan booklet. The particular sampling plan we chose from that booklet is one that will control the Incoming QC rejection rate, so that the rejection rate is least 90% for lots that have defective rates of 1.5% or greater (i.e., LQL = 1.5%), which is the rejection rate required by our Risk Management Plan documents. We have chosen to limit the range of allowed lot sizes that we purchase to , in order to ensure that the rejection rate is stable no matter what lot size is received. 22
Statistical Process Control Basics. 70 GLEN ROAD, CRANSTON, RI 02920 T: 4014611118 F: 4014611119 www.tedcoinc.com
Statistical Process Control Basics 70 GLEN ROAD, CRANSTON, RI 02920 T: 4014611118 F: 4014611119 www.tedcoinc.com What is Statistical Process Control? Statistical Process Control (SPC) is an industrystandard
More informationTHE PROCESS CAPABILITY ANALYSIS  A TOOL FOR PROCESS PERFORMANCE MEASURES AND METRICS  A CASE STUDY
International Journal for Quality Research 8(3) 399416 ISSN 18006450 Yerriswamy Wooluru 1 Swamy D.R. P. Nagesh THE PROCESS CAPABILITY ANALYSIS  A TOOL FOR PROCESS PERFORMANCE MEASURES AND METRICS 
More information" Y. Notation and Equations for Regression Lecture 11/4. Notation:
Notation: Notation and Equations for Regression Lecture 11/4 m: The number of predictor variables in a regression Xi: One of multiple predictor variables. The subscript i represents any number from 1 through
More informationt Tests in Excel The Excel Statistical Master By Mark Harmon Copyright 2011 Mark Harmon
ttests in Excel By Mark Harmon Copyright 2011 Mark Harmon No part of this publication may be reproduced or distributed without the express permission of the author. mark@excelmasterseries.com www.excelmasterseries.com
More informationKSTAT MINIMANUAL. Decision Sciences 434 Kellogg Graduate School of Management
KSTAT MINIMANUAL Decision Sciences 434 Kellogg Graduate School of Management Kstat is a set of macros added to Excel and it will enable you to do the statistics required for this course very easily. To
More informationShainin: A concept for problem solving
Shainin: A concept for problem solving Lecture at the Shainin conference Amelior 11 December 2009 Willy Vandenbrande www.qsconsult.be 1 Dorian Shainin (1914 2000) Aeronautical engineer (MIT 1936) Design
More informationNormality Testing in Excel
Normality Testing in Excel By Mark Harmon Copyright 2011 Mark Harmon No part of this publication may be reproduced or distributed without the express permission of the author. mark@excelmasterseries.com
More informationBinary Diagnostic Tests Two Independent Samples
Chapter 537 Binary Diagnostic Tests Two Independent Samples Introduction An important task in diagnostic medicine is to measure the accuracy of two diagnostic tests. This can be done by comparing summary
More informationBusiness Statistics. Successful completion of Introductory and/or Intermediate Algebra courses is recommended before taking Business Statistics.
Business Course Text Bowerman, Bruce L., Richard T. O'Connell, J. B. Orris, and Dawn C. Porter. Essentials of Business, 2nd edition, McGrawHill/Irwin, 2008, ISBN: 9780073319889. Required Computing
More informationRecall this chart that showed how most of our course would be organized:
Chapter 4 OneWay ANOVA Recall this chart that showed how most of our course would be organized: Explanatory Variable(s) Response Variable Methods Categorical Categorical Contingency Tables Categorical
More informationData Mining Techniques Chapter 5: The Lure of Statistics: Data Mining Using Familiar Tools
Data Mining Techniques Chapter 5: The Lure of Statistics: Data Mining Using Familiar Tools Occam s razor.......................................................... 2 A look at data I.........................................................
More informationGuide to Microsoft Excel for calculations, statistics, and plotting data
Page 1/47 Guide to Microsoft Excel for calculations, statistics, and plotting data Topic Page A. Writing equations and text 2 1. Writing equations with mathematical operations 2 2. Writing equations with
More informationOneWay Analysis of Variance
OneWay Analysis of Variance Note: Much of the math here is tedious but straightforward. We ll skim over it in class but you should be sure to ask questions if you don t understand it. I. Overview A. We
More informationAugust 2012 EXAMINATIONS Solution Part I
August 01 EXAMINATIONS Solution Part I (1) In a random sample of 600 eligible voters, the probability that less than 38% will be in favour of this policy is closest to (B) () In a large random sample,
More informationSelecting SPC Software for Batch and Specialty Chemicals Processing
WHITE PAPER Selecting SPC Software for Batch and Specialty Chemicals Processing Statistical Process Control (SPC) is a necessary part of modern chemical processing. The software chosen to collect quality
More informationInstruction Manual for SPC for MS Excel V3.0
Frequency Business Process Improvement 2813049504 20314 Lakeland Falls www.spcforexcel.com Cypress, TX 77433 Instruction Manual for SPC for MS Excel V3.0 35 30 25 LSL=60 Nominal=70 Capability Analysis
More informationCourse Text. Required Computing Software. Course Description. Course Objectives. StraighterLine. Business Statistics
Course Text Business Statistics Lind, Douglas A., Marchal, William A. and Samuel A. Wathen. Basic Statistics for Business and Economics, 7th edition, McGrawHill/Irwin, 2010, ISBN: 9780077384470 [This
More informationMINITAB ASSISTANT WHITE PAPER
MINITAB ASSISTANT WHITE PAPER This paper explains the research conducted by Minitab statisticians to develop the methods and data checks used in the Assistant in Minitab 17 Statistical Software. OneWay
More informationBowerman, O'Connell, Aitken Schermer, & Adcock, Business Statistics in Practice, Canadian edition
Bowerman, O'Connell, Aitken Schermer, & Adcock, Business Statistics in Practice, Canadian edition Online Learning Centre Technology StepbyStep  Excel Microsoft Excel is a spreadsheet software application
More informationChapter 7 Section 7.1: Inference for the Mean of a Population
Chapter 7 Section 7.1: Inference for the Mean of a Population Now let s look at a similar situation Take an SRS of size n Normal Population : N(, ). Both and are unknown parameters. Unlike what we used
More informationHow To Run Statistical Tests in Excel
How To Run Statistical Tests in Excel Microsoft Excel is your best tool for storing and manipulating data, calculating basic descriptive statistics such as means and standard deviations, and conducting
More informationStatistical Functions in Excel
Statistical Functions in Excel There are many statistical functions in Excel. Moreover, there are other functions that are not specified as statistical functions that are helpful in some statistical analyses.
More informationsensr  Part 2 Similarity testing and replicated data (and sensr) p 1 2 p d 1. Analysing similarity test data.
Similarity testing and replicated data (and sensr) Per Bruun Brockhoff Professor, Statistics DTU, Copenhagen August 17 2015 sensr  Part 2 1. Analysing similarity test data. 2. Planning similarity tests
More informationCHAPTER 13 SIMPLE LINEAR REGRESSION. Opening Example. Simple Regression. Linear Regression
Opening Example CHAPTER 13 SIMPLE LINEAR REGREION SIMPLE LINEAR REGREION! Simple Regression! Linear Regression Simple Regression Definition A regression model is a mathematical equation that descries the
More informationProduct Quality Management
Product Quality Management Deborah Baly, Ph.D Sr. Director, Commercial Product Quality Management, GNE/ROCHE 1 Presentation Outline: Product Quality Management Regulatory landscape and need for integrated
More informationRegression Analysis: A Complete Example
Regression Analysis: A Complete Example This section works out an example that includes all the topics we have discussed so far in this chapter. A complete example of regression analysis. PhotoDisc, Inc./Getty
More information5/31/2013. Chapter 8 Hypothesis Testing. Hypothesis Testing. Hypothesis Testing. Outline. Objectives. Objectives
C H 8A P T E R Outline 8 1 Steps in Traditional Method 8 2 z Test for a Mean 8 3 t Test for a Mean 8 4 z Test for a Proportion 8 6 Confidence Intervals and Copyright 2013 The McGraw Hill Companies, Inc.
More informationORACLE CONSULTING GROUP
ORACLE CONSULTING GROUP 9 Golder Ranch Rd., Ste. 1 Tucson, Arizona 9 Web Site: Email: 2020 2020 (FAX) CONSULTING MEMORANDUM QUALITY SYSTEM INSPECTION TECHNIQUE
More information11. Analysis of Casecontrol Studies Logistic Regression
Research methods II 113 11. Analysis of Casecontrol Studies Logistic Regression This chapter builds upon and further develops the concepts and strategies described in Ch.6 of Mother and Child Health:
More informationMaximize Software Development ROI With Quality Assurance. Showing the value of the Quality Process
Maximize Software Development ROI With Quality Assurance Showing the value of the Quality Process Thibault Dambrine Agenda Software Quality Assurance ROI  Quantifying the Cost of Quality  Justifying
More informationIntroduction to. Hypothesis Testing CHAPTER LEARNING OBJECTIVES. 1 Identify the four steps of hypothesis testing.
Introduction to Hypothesis Testing CHAPTER 8 LEARNING OBJECTIVES After reading this chapter, you should be able to: 1 Identify the four steps of hypothesis testing. 2 Define null hypothesis, alternative
More information2. Simple Linear Regression
Research methods  II 3 2. Simple Linear Regression Simple linear regression is a technique in parametric statistics that is commonly used for analyzing mean response of a variable Y which changes according
More informationActivity 3.7 Statistical Analysis with Excel
Activity 3.7 Statistical Analysis with Excel Introduction Engineers use various tools to make their jobs easier. Spreadsheets can greatly improve the accuracy and efficiency of repetitive and common calculations;
More informationChapter 7. Oneway ANOVA
Chapter 7 Oneway ANOVA Oneway ANOVA examines equality of population means for a quantitative outcome and a single categorical explanatory variable with any number of levels. The ttest of Chapter 6 looks
More informationChapter 9. TwoSample Tests. Effect Sizes and Power Paired t Test Calculation
Chapter 9 TwoSample Tests Paired t Test (Correlated Groups t Test) Effect Sizes and Power Paired t Test Calculation Summary Independent t Test Chapter 9 Homework Power and TwoSample Tests: Paired Versus
More informationGeneralized Linear Models
Generalized Linear Models We have previously worked with regression models where the response variable is quantitative and normally distributed. Now we turn our attention to two types of models where the
More informationName: Date: Use the following to answer questions 34:
Name: Date: 1. Determine whether each of the following statements is true or false. A) The margin of error for a 95% confidence interval for the mean increases as the sample size increases. B) The margin
More informationReview #2. Statistics
Review #2 Statistics Find the mean of the given probability distribution. 1) x P(x) 0 0.19 1 0.37 2 0.16 3 0.26 4 0.02 A) 1.64 B) 1.45 C) 1.55 D) 1.74 2) The number of golf balls ordered by customers of
More informationUnderstanding Confidence Intervals and Hypothesis Testing Using Excel Data Table Simulation
Understanding Confidence Intervals and Hypothesis Testing Using Excel Data Table Simulation Leslie Chandrakantha lchandra@jjay.cuny.edu Department of Mathematics & Computer Science John Jay College of
More informationASSESSMENT OF QUALITY RISK MANAGEMENT IMPLEMENTATION
PHARMACEUTICAL INSPECTION CONVENTION PHARMACEUTICAL INSPECTION COOPERATION SCHEME PI 0381 26 March 2012 AIDEMEMOIRE ASSESSMENT OF QUALITY RISK MANAGEMENT IMPLEMENTATION PIC/S March 2012 Reproduction
More informationOptimizing Quality Control / Quality Assurance Agents of a Global Sourcing / Procurement Strategy
Optimizing Quality Control / Quality Assurance Agents of a Global Sourcing / Procurement Strategy Global Pharma Sourcing Conference December 67, 2011 Philadelphia, USA Nigel J. Smart, Ph.D. Smart Consulting
More informationAP Physics 1 and 2 Lab Investigations
AP Physics 1 and 2 Lab Investigations Student Guide to Data Analysis New York, NY. College Board, Advanced Placement, Advanced Placement Program, AP, AP Central, and the acorn logo are registered trademarks
More informationCHAPTER 13. Experimental Design and Analysis of Variance
CHAPTER 13 Experimental Design and Analysis of Variance CONTENTS STATISTICS IN PRACTICE: BURKE MARKETING SERVICES, INC. 13.1 AN INTRODUCTION TO EXPERIMENTAL DESIGN AND ANALYSIS OF VARIANCE Data Collection
More informationStats for Strategy Fall 2012 FirstDiscussion Handout: Stats Using Calculators and MINITAB
Stats for Strategy Fall 2012 FirstDiscussion Handout: Stats Using Calculators and MINITAB DIRECTIONS: Welcome! Your TA will help you apply your Calculator and MINITAB to review Business Stats, and will
More informationCommon Tools for Displaying and Communicating Data for Process Improvement
Common Tools for Displaying and Communicating Data for Process Improvement Packet includes: Tool Use Page # Box and Whisker Plot Check Sheet Control Chart Histogram Pareto Diagram Run Chart Scatter Plot
More informationProcess Quality. BIZ212104 Production & Operations Management. Sung Joo Bae, Assistant Professor. Yonsei University School of Business
BIZ212104 Production & Operations Management Process Quality Sung Joo Bae, Assistant Professor Yonsei University School of Business Disclaimer: Many slides in this presentation file are from the copyrighted
More informationIndividual Moving Range (IMR) Charts. The Swiss Army Knife of Process Charts
Individual Moving Range (IMR) Charts The Swiss Army Knife of Process Charts SPC Selection Process Choose Appropriate Control Chart ATTRIBUTE type of data CONTINUOUS DEFECTS type of attribute data DEFECTIVES
More informationQUIZ MODULE 1: BASIC CONCEPTS IN QUALITY AND TQM
QUIZ MODULE 1: BASIC CONCEPTS IN QUALITY AND TQM These questions cover Sessions 1, 2, 5, 6, 7. The correct answer is shown in bold A fundamental attribute of TQM is Drawing control charts Having team meetings
More informationGuidance for Industry
Guidance for Industry Q8, Q9, and Q10 Questions and Answers(R4) U.S. Department of Health and Human Services Food and Drug Administration Center for Drug Evaluation and Research (CDER) Center for Biologics
More informationAssessing Measurement System Variation
Assessing Measurement System Variation Example 1: Fuel Injector Nozzle Diameters Problem A manufacturer of fuel injector nozzles installs a new digital measuring system. Investigators want to determine
More informationTHE BASICS OF STATISTICAL PROCESS CONTROL & PROCESS BEHAVIOUR CHARTING
THE BASICS OF STATISTICAL PROCESS CONTROL & PROCESS BEHAVIOUR CHARTING A User s Guide to SPC By David Howard ManagementNewStyle "...Shewhart perceived that control limits must serve industry in action.
More informationCase Study in Data Analysis Does a drug prevent cardiomegaly in heart failure?
Case Study in Data Analysis Does a drug prevent cardiomegaly in heart failure? Harvey Motulsky hmotulsky@graphpad.com This is the first case in what I expect will be a series of case studies. While I mention
More informationConsider a study in which. How many subjects? The importance of sample size calculations. An insignificant effect: two possibilities.
Consider a study in which How many subjects? The importance of sample size calculations Office of Research Protections Brown Bag Series KB Boomer, Ph.D. Director, boomer@stat.psu.edu A researcher conducts
More informationHOW TO WRITE A LABORATORY REPORT
HOW TO WRITE A LABORATORY REPORT Pete Bibby Dept of Psychology 1 About Laboratory Reports The writing of laboratory reports is an essential part of the practical course One function of this course is to
More informationLifecycle Management of Analytical Procedures; What is it all about? Jane Weitzel Independent Consultant
Lifecycle Management of Analytical Procedures; What is it all about? Jane Weitzel Independent Consultant 2 USP Stimuli Article Lifecycle Management of Analytical Procedures: Method Development, Procedure
More informationRegulatory Expectations of Executive Management
Regulatory Expectations of Executive Management Steven Lynn, MS, CMQ/OE Director Office of Manufacturing and Product Quality Office of Compliance CDER/US FDA PDA ICH Q10 Executive Management Workshop PDA
More informationFOOD FOR THOUGHT Topical Insights from our Subject Matter Experts UNDERSTANDING WHAT IS NEEDED TO PRODUCE QUALITY DATA
FOOD FOR THOUGHT Topical Insights from our Subject Matter Experts UNDERSTANDING WHAT IS NEEDED TO PRODUCE QUALITY DATA The NFL White Paper Series Volume 7, January 2013 Overview and a Scenario With so
More information2013 MBA Jump Start Program. Statistics Module Part 3
2013 MBA Jump Start Program Module 1: Statistics Thomas Gilbert Part 3 Statistics Module Part 3 Hypothesis Testing (Inference) Regressions 2 1 Making an Investment Decision A researcher in your firm just
More informationRegression stepbystep using Microsoft Excel
Step 1: Regression stepbystep using Microsoft Excel Notes prepared by Pamela Peterson Drake, James Madison University Type the data into the spreadsheet The example used throughout this How to is a regression
More informationGOOD DOCUMENTATION AND QUALITY MANAGEMENT PRINCIPLES. Vimal Sachdeva Technical Officer (Inspector), WHO Prequalification of Medicines Programme
GOOD DOCUMENTATION AND QUALITY MANAGEMENT PRINCIPLES Vimal Sachdeva Technical Officer (Inspector), WHO Prequalification of Medicines Programme Contents 1. Why good documentation is essential? 2. What constitutes
More informationCDC UNIFIED PROCESS PRACTICES GUIDE
Document Purpose The purpose of this document is to provide guidance on the practice of Quality Management and to describe the practice overview, requirements, best practices, activities, and key terms
More informationModule 4 (Effect of Alcohol on Worms): Data Analysis
Module 4 (Effect of Alcohol on Worms): Data Analysis Michael Dunn Capuchino High School Introduction In this exercise, you will first process the timelapse data you collected. Then, you will cull (remove)
More informationSPC Response Variable
SPC Response Variable This procedure creates control charts for data in the form of continuous variables. Such charts are widely used to monitor manufacturing processes, where the data often represent
More informationIMPROVING QUALITY WITH IFS QUALITY MANAGEMENT IFS CUSTOMER SUMMIT 2011, CHICAGO
www.ifsworld.com IMPROVING QUALITY WITH IFS QUALITY MANAGEMENT IFS CUSTOMER SUMMIT 2011, CHICAGO KARIN RAINESALO BUSINESS SOLUTIONS CONSULTANT karin.rainesalo@ifsworld.com IFS CUSTOMER SUMMIT 2011, CHICAGO
More informationNonInferiority Tests for One Mean
Chapter 45 NonInferiority ests for One Mean Introduction his module computes power and sample size for noninferiority tests in onesample designs in which the outcome is distributed as a normal random
More informationGetting Started with Minitab 17
2014 by Minitab Inc. All rights reserved. Minitab, Quality. Analysis. Results. and the Minitab logo are registered trademarks of Minitab, Inc., in the United States and other countries. Additional trademarks
More informationSUPPLIER QUALITY SYSTEM AUDIT
Company Name: Date: Company Address: Quality Assurance Mgr: President: Number of employees: Is your company receptive to source inspection? List all standards the quality system is based on Web Page: Company
More informationAP Statistics 2005 Scoring Guidelines
AP Statistics 2005 Scoring Guidelines The College Board: Connecting Students to College Success The College Board is a notforprofit membership association whose mission is to connect students to college
More informationCommercial Manufacturing  Qualification & Validationrelated GMP Deficiencies and Other Lifecycle Considerations
Commercial Manufacturing  Qualification & Validationrelated GMP Deficiencies and Other Lifecycle Considerations Kevin O Donnell PhD Market Compliance Manager, IMB PDA / FDA Conference Pharmaceutical
More informationWISE Power Tutorial All Exercises
ame Date Class WISE Power Tutorial All Exercises Power: The B.E.A.. Mnemonic Four interrelated features of power can be summarized using BEA B Beta Error (Power = 1 Beta Error): Beta error (or Type II
More informationANOVA ANOVA. TwoWay ANOVA. OneWay ANOVA. When to use ANOVA ANOVA. Analysis of Variance. Chapter 16. A procedure for comparing more than two groups
ANOVA ANOVA Analysis of Variance Chapter 6 A procedure for comparing more than two groups independent variable: smoking status nonsmoking one pack a day > two packs a day dependent variable: number of
More informationBest Practices In Ensuring Quality Standards When Outsourcing To Contract Manufacturers, Licensees And Consultants
Best Practices In Ensuring Quality Standards When Outsourcing To Contract Manufacturers, Licensees And Consultants Alex D. Kanarek, PhD BioProcess Technology Consultants, Inc. Strategic Institute Quality
More informationUsing Statistical Process Control (SPC) for improved Utility Management
Using Statistical Process Control (SPC) for improved Utility Management Scott Dorner Hach Company Manage and Transform data into information to gain efficiencies 1 Data, Data Everywhere We track enormous
More informationSurvey, Statistics and Psychometrics Core Research Facility University of NebraskaLincoln. LogRank Test for More Than Two Groups
Survey, Statistics and Psychometrics Core Research Facility University of NebraskaLincoln LogRank Test for More Than Two Groups Prepared by Harlan Sayles (SRAM) Revised by Julia Soulakova (Statistics)
More informationPharmaceutical Product Quality, Quality by Design, cgmp, and Quality Metrics
Pharmaceutical Product Quality, Quality by Design, cgmp, and Quality Metrics Lawrence X. Yu, Ph.D. Deputy Director Office of Pharmaceutical Quality Center for Drug Evaluation and Research Food and Drug
More informationStatistics Review PSY379
Statistics Review PSY379 Basic concepts Measurement scales Populations vs. samples Continuous vs. discrete variable Independent vs. dependent variable Descriptive vs. inferential stats Common analyses
More informationCONTENTS OF DAY 2. II. Why Random Sampling is Important 9 A myth, an urban legend, and the real reason NOTES FOR SUMMER STATISTICS INSTITUTE COURSE
1 2 CONTENTS OF DAY 2 I. More Precise Definition of Simple Random Sample 3 Connection with independent random variables 3 Problems with small populations 8 II. Why Random Sampling is Important 9 A myth,
More informationStatistical Quality Control
Statistical Quality Control CHAPTER 6 Before studying this chapter you should know or, if necessary, review 1. Quality as a competitive priority, Chapter 2, page 00. 2. Total quality management (TQM) concepts,
More informationHow to Conduct the Method Validation with a New IPT (InProcess Testing) Method
How to Conduct the Method Validation with a New IPT (InProcess Testing) Method Weifeng Frank Zhang QA Engineer BMS Ernest Lee Senior Manager, Facilities & Engineering Medarex, a fully owned subsidiary
More informationGE Fanuc Automation CIMPLICITY. Statistical Process Control. CIMPLICITY Monitoring and Control Products. Operation Manual
GE Fanuc Automation CIMPLICITY Monitoring and Control Products CIMPLICITY Statistical Process Control Operation Manual GFK1413E December 2000 Following is a list of documentation icons: GFL005 Warning
More informationGUIDANCE FOR ASSESSING THE LIKELIHOOD THAT A SYSTEM WILL DEMONSTRATE ITS RELIABILITY REQUIREMENT DURING INITIAL OPERATIONAL TEST.
GUIDANCE FOR ASSESSING THE LIKELIHOOD THAT A SYSTEM WILL DEMONSTRATE ITS RELIABILITY REQUIREMENT DURING INITIAL OPERATIONAL TEST. 1. INTRODUCTION Purpose The purpose of this white paper is to provide guidance
More informationStatistical Process Control OPRE 6364 1
Statistical Process Control OPRE 6364 1 Statistical QA Approaches Statistical process control (SPC) Monitors production process to prevent poor quality Acceptance sampling Inspects random sample of product
More informationNCSS Statistical Software Principal Components Regression. In ordinary least squares, the regression coefficients are estimated using the formula ( )
Chapter 340 Principal Components Regression Introduction is a technique for analyzing multiple regression data that suffer from multicollinearity. When multicollinearity occurs, least squares estimates
More informationSPSS/Excel Workshop 2 Semester One, 2010
SPSS/Excel Workshop 2 Semester One, 2010 In Assignment 2 of STATS 10x you may want to use Excel or SPSS to perform some calculations, that is, finding Normal probabilities and Inverse Normal values in
More informationMeasurement and Metrics Fundamentals. SE 350 Software Process & Product Quality
Measurement and Metrics Fundamentals Lecture Objectives Provide some basic concepts of metrics Quality attribute metrics and measurements Reliability, validity, error Correlation and causation Discuss
More informationMULTIPLE LINEAR REGRESSION ANALYSIS USING MICROSOFT EXCEL. by Michael L. Orlov Chemistry Department, Oregon State University (1996)
MULTIPLE LINEAR REGRESSION ANALYSIS USING MICROSOFT EXCEL by Michael L. Orlov Chemistry Department, Oregon State University (1996) INTRODUCTION In modern science, regression analysis is a necessary part
More informationRandomized Block Analysis of Variance
Chapter 565 Randomized Block Analysis of Variance Introduction This module analyzes a randomized block analysis of variance with up to two treatment factors and their interaction. It provides tables of
More informationCertified Quality Engineer
Certified Quality Engineer Quality excellence to enhance your career and boost your organization s bottom line asqmena.org/certification.php www.asqmena.org Certification from ASQ is considered a mark
More informationNCSS Statistical Software
Chapter 06 Introduction This procedure provides several reports for the comparison of two distributions, including confidence intervals for the difference in means, twosample ttests, the ztest, the
More informationMultiplexer Software. www.elcometer.com. Multiplexer Software. Dataputer DATAXL Software
Multiplexer Software Multiplexer Software There are two ways that data can be collected Electronically, where there is no human intervention, and Manually, where data is collected by the User with the
More informationThe G Chart in Minitab Statistical Software
The G Chart in Minitab Statistical Software Background Developed by James Benneyan in 1991, the g chart (or G chart in Minitab) is a control chart that is based on the geometric distribution. Benneyan
More informationTotal Quality Management TQM
Total Quality Management TQM What is quality? Why is quality important? How can quality be improved? How is a process controlled? How can products be controlled? Contents Definition of Quality...1 Evolution
More informationEXCEL Tutorial: How to use EXCEL for Graphs and Calculations.
EXCEL Tutorial: How to use EXCEL for Graphs and Calculations. Excel is powerful tool and can make your life easier if you are proficient in using it. You will need to use Excel to complete most of your
More informationDATA VALIDATION, PROCESSING, AND REPORTING
DATA VALIDATION, PROCESSING, AND REPORTING After the field data are collected and transferred to your office computing environment, the next steps are to validate and process data, and generate reports.
More informationChapter Study Guide. Chapter 11 Confidence Intervals and Hypothesis Testing for Means
OPRE504 Chapter Study Guide Chapter 11 Confidence Intervals and Hypothesis Testing for Means I. Calculate Probability for A Sample Mean When Population σ Is Known 1. First of all, we need to find out the
More information1. How different is the t distribution from the normal?
Statistics 101 106 Lecture 7 (20 October 98) c David Pollard Page 1 Read M&M 7.1 and 7.2, ignoring starred parts. Reread M&M 3.2. The effects of estimated variances on normal approximations. tdistributions.
More informationSite Quality Metrics OTC perspective. Veronica Cruz, PhD Executive VP/COO. May 13, 2014 Quality Session 3
1 Site Quality Metrics perspective Veronica Cruz, PhD Executive VP/COO May 13, 2014 Quality Session 3 2 1 Quality Management System A comprehensive Quality Management System. Establishes controls that
More informationOPTIONS TRADING AS A BUSINESS UPDATE: Using ODDS Online to Find A Straddle s Exit Point
This is an update to the Exit Strategy in Don Fishback s Options Trading As A Business course. We re going to use the same example as in the course. That is, the AMZN trade: Buy the AMZN July 22.50 straddle
More informationTwosample ttests.  Independent samples  Pooled standard devation  The equal variance assumption
Twosample ttests.  Independent samples  Pooled standard devation  The equal variance assumption Last time, we used the mean of one sample to test against the hypothesis that the true mean was a particular
More information