EIEIOTECHNOVATION
← Training programmes

EIEIO / AUTOMATION & EMERGING TECHNOLOGY

MEASUREMENT SYSTEM ANALYSIS (MSA)

Course overview and objectives

2 days
Discuss this programme →

Duration

MEASUREMENT SYSTEM ANALYSIS (MSA)

2 days

Overview

The purpose of measurement is to obtain information on entities and make decisions based on the information contained in the measurement result (as a tool for decision-making).

The measurement system analysis (MSA) is a study of the process of being measured that uses statistical methods to determine the amount of variation that exists. Variation in the measurement system is noise to the measurement outcome. If this noise is too large that will be causing unable to determine the actual measurement results. Thus, the risks exist in the decisions that cause rejecting a good item by the producer and accepting a bad item by the consumer respectively.

In any measurement process, accuracy and precision are associated with it. In order to know the quality of measurement results, the variation of the accuracy and precision have to be known through MSA that covers location and width variation analysis respectively.

The location variation analysis is related to the accuracy of measurement which includes bias, stability, and linearity. These studies will be using reference value as the target to compare the indicator value shown in the measuring devices and the difference between the average measurement value and the reference value is called the bias. If the bias increases with the size gradually called a linearity problem, whereas the problem of stability has a random bias with time.

The width variation analysis links with the precision that contains repeatability and reproducibility. This analysis is also called gauge repeatability and reproducibility study with the abbreviation GR&R. The repeatability is associated with measuring equipment in which the variation comes from repeated applications for the same measurand under the same conditions of measurement. Whereas reproducibility is the variation that exists when producing another set of repeatability under different conditions (If the different condition is the operators, then reproducibility is used to attribute the operator-to-operator variation). The statistical study for this width variation using continuous data is called a Gage R&R study, and the tool used for discrete data (attributes) is called an Attribute Agreement Analysis.

The magnitude of measurement variations depended on the measurement system used for making the measurement. Using MSA information about the components of variation will help properly select the suitability of the measurement system used for a given process and provide aid in the correct direction for measurement system improvement.

Scope

This course mainly focused on three areas to conduct the measurement system study on variables data (width variation), bias analysis (location variation), and attributes data (attribute agreement). These studies are based on simulated (or appropriately collected) variables and attribute data, then analyze further by using statistical software or manual calculation. The results of the studies are interpreted comprehensively. The width variation studies include X̅ & R and ANOVA (cross design) methods, whereas location variation studies cover the variability on bias, stability, and linearity. The attribute agreement analysis is based on attribute data (accept or reject) by using Cohen’s Kappa Statistic. Besides that, the probabilities and statistics concept, and the activities of performing a measurement system analysis are explained thoroughly. The results include the various component of variations, ANOVA table, graphical, indices, and misclassification rates that were used to determine the suitability of the measurement system. The methods and formats used in this program are based on the AIAG approach.

Training Objectives

The measurement system analysis is studying variations of the parameters such as bias, stability, linearity, repeatability, and reproducibility in a measurement system. The MSA methods are based on AIAG for the bias, stability, and linearity (location variation), whereas the study of repeatability and reproducibility (width variation) is using X̅ and Range and ANOVA methods. This analysis included a study on attribute data called Attribute Agreement Analysis.

This training, with the materials and guidance provided that will assist the trainees, can prepare for planning, developing, and performing a measurement system analysis by using statistical software. This training will also provide thorough guidance for the X̅ & R and ANOVA methods on repeatable measurements, attribute studies, and bias studies, which are commonly used in the industry to ensure the suitability of a measurement system. Thus, the results obtained from the measurement system are usually used for quality judgment. These studies will provide useful information for the observers on how well inspection or measurement systems are and reduce the risks of wrong decision-making. Thus, this training will provide the detail for these studies including case studies, work examples, and questions and answers that will help to ensure effective implementation. Using statistical software that provided step by step to perform these studies will help the trainees better understand application and evaluation. Finally, measurement system analysis is to qualify a measurement system for use by quantifying its accuracy, precision, and stability. All these will help the trainees to be confident to handle the situation when performing a measurement or/and answer the questions related to the measurement system raised by the concerning auditors or customers.

Training Outcomes

After seriously committed to this course, the trainee should be able to do the following:

  • Understand the probabilities and statistics concepts that are used in measurement system analysis (MSA).
  • Identify MSA parameters in a measurement system with the location variations (bias, stability, and linearity) that will affect the accuracy of measurement whereas width variations (repeatability and reproducibility) will impact the precision of measurement.
  • Learn how to estimate and evaluate the variation components from the measurement system, there are EV, AV, R&R, PV, TV, and ndc.
  • Understand the X̅ and R (Range), and ANOVA methods are based on range and standard deviation approaches respectively, and appropriate application for these methods.
  • Plan, perform, analyze, evaluate, and report the GR&R studies and attribute study (attribute agreement analysis) that results provide criteria for accepting the measuring equipment.
  • Understand the attribute study (Cohen’s Kappa Statistic) to ensure the effectiveness of the inspection system.
  • Know that bias studies (width variations) are based on a t-test to judge acceptability. If the result is unacceptable, recalibration of the measuring gauge is required.
  • Use the measurement and process capability indices to determine the suitability of the measurement system to be used.
  • Plot and explain the control and bar charts to show the pattern of a GR&R study.
  • Interpreted misclassification rates are based on joint and conditional probabilities that are better used as acceptance criteria for a measurement system.
  • Use software programs or manual calculations for the MSA and interpret the results obtained.
  • Improve equipment maintenance, selection, calibration, use, and the observer's training and skill.
  • Know the consumer and producer risks due to the measurement variation.
  • Understanding MSA is to qualify a measurement system for use by quantifying its accuracy, precision, and stability.
  • Interpret and justify the measurement system analysis results and decide on what follow-up action to be taken, and be able to handle the situation during measuring and auditing related to the measurement system.

Methodology

Training slides, programed software worksheets (Based on the formulas and functions) for the Gauge R&R Studies, Review Questions, Exercises, Case Studies, Data Simulation, Q&A and Test Before Training & After Training (This training requires Microsoft MINITAB Software).

Who Should Attend?

The person works related to the measurement system that included the gauges, instruments, and operators in a measurement process. It can serve as a refresher for experienced technicians; or it can be used in orientation for new hires. The persons who have experience and knowledge in GR&R study are encouraged to attend.

The company shall have statistical software (Microsoft Excel can be used). Participants must have basic statistical knowledge, fundamental GR&R background, comfortable using a statistical software and simple algebra formulas.

Type of Industrial

Manufacturing, Laboratories, institutions, inspection agents and the industries, those involving calibration or measurements as part of their activity.

Summary Course Outline

  • Basic Probability and Statistics Used in Gauge R&R Study
  • Statistical Testing
  • Measurement System Parameters
  • Planning for the Gauge R&R Studies
  • Gauge R&R Studies (Width Variation)
  • Bias Studies (Location Variation)
  • Gauge R&R Studies by Using a Software
  • Performing an Attribute Gauge R&R Study
  • Evaluation of Gauge R&R Studies Results
  • Review Questions (Fifty Multiple Choice Questions)
  • Appendix
  • References

Detail Course Outline

  • Basic Probability and Statistics Used in Gauge R&R Study
  • Probability
  • Basic Probability Concepts
  • Basic Probability Definitions
  • Bayes’ Theorem
  • Counting of Events
  • Binomial Probability Distribution
  • Poisson Probability Distribution
  • Normal Probability Distribution
  • Statistics
  • Definition
  • Sample and Population
  • Type of Data
  • Statistics and Parameters
  • Sample Measurement
  • Measures of Central Tendency
  • Measures of Dispersion
  • Measures of Shape
  • Central Limit Theorem
  • Expectation, Variance and Covariance
  • Expectation
  • Variance
  • Covariance
  • Population and Sample Discrete Independent Random Variables
  • Chapter Review Questions
  • Statistical Testing
  • t – Test

Introduction

  • t-Test for Single Population
  • t-Test for Two Populations
  • Independent Samples
  • Non-Pooled t-Test
  • Pooled t-Test
  • Dependent Samples
  • Paired t-Test
  • Summary Procedures for t-Test
  • Pooled t-Test
  • Non-Pooled t-Test
  • Paired t-Test
  • Analysis of Variance (ANOVA)

Introduction

  • The Logic of Analysis of Variance
  • ANOVA Notation and Formulas
  • Conclusion of ANOVA
  • Two-Factor (Factor-A and Factor-B) ANOVA
  • One-Way ANOVA
  • Two-Way ANOVA
  • Chapter Review Questions
  • Measurement System Parameters
  • Measurement Variation
  • Measurement Process
  • Measurement Parameters
  • Source of Variation
  • Check Standard
  • Reference Value
  • True Value
  • Discrimination and Number of Distinct Categories (ndc)
  • Measurement Process Control Charts
  • Measurement Process Variation
  • Gauge Stability
  • Gauge Bias
  • Gauge Linearity
  • Gauge Repeatability
  • Gauge Reproducibility
  • Gauge Repeatability and Reproducibility
  • Chapter Review Questions
  • Planning for the Gauge R&R Studies
  • The Techniques to Perform Gauge R&R Study
  • Repeatability and Reproducibility Variations
  • Sample Size for Gauge R&R Study
  • The Gauge R&R Study Awareness
  • Relation of the Procedure to Control Charts
  • Chapter Review Questions
  • Gauge R&R Studies (Width Variation)
  • Variation Within Sample
  • Performing a Traditional GR&R Study
  • Example of a Traditional GR&R Study
  • The Number of Distinct Categories (ndc)
  • Interpretation of the Results
  • Chapter Review Questions
  • Bias Studies (Location Variation)
  • Bias Study

Introduction

  • Conducting the Study
  • Analysis of Results
  • Numerical Example
  • Stability Study

Introduction

  • Conducting the Study
  • Analysis of Results
  • Numerical Example
  • Linearity Study

Introduction

  • Conducting the Study
  • Analysis of Results
  • Numerical Example
  • Chapter Review Questions
  • Gauge R&R Studies by Using a Software (ANOVA Crossed Design)
  • Introduction to X̅ & Range Method
  • Introduction to ANOVA Method
  • Study Preparation
  • Data Work Sheets
  • Formulas and Tables
  • Example of Gauge R&R Study
  • Study Information
  • Data and Work Sheets
  • Specification Control Indications
  • Data Arrangement for Minitab
  • Performing a Gauge R&R Study Using Excel
  • X̅ & Range Method
  • ANOVA Method
  • Performing a Gauge R&R Study Using Minitab
  • Data Transferring from Excel to Minitab
  • Data and X̅ & Range/ANOVA Methods
  • Setup for X̅ & Range/ANOVA Methods
  • The Outputs
  • Reporting the Results for X̅ & Range Method
  • The Graphs for X̅ & Range Method
  • Reporting the Results for ANOVA Method
  • The Graphs for ANOVA Method
  • Saving MINITAB Project in Desktop Directory
  • Select and Copy the Results from Minitab
  • Chapter Review Questions
  • Performing an Attribute Gauge R&R Study

Introduction

  • The Short-Form Attribute Study
  • The Study without Standard
  • The Study with Standard
  • The Long-Form Attribute Study with Standard

Introduction

  • Procedure and Calculation
  • Acceptance Criteria
  • Attribute R&R Study Using Cohen’s Kappa Statistic

Introduction

  • Crosstab Method
  • Analyse the Agreement Among the Appraisers
  • Analyse the Reliability of the Appraisers
  • Chapter Review Questions
  • Evaluation of Gauge R&R Studies Results
  • Graphical Analysis
  • Bar Chart
  • X̅ Chart
  • Range Chart
  • Run Chart
  • Box Plot
  • Interaction Chart
  • Misclassification Rate
  • Definition
  • Calculation of Misclassification Rates
  • Comparison of the Results by using Excel
  • Equipment Variation, EV
  • Part (P), Appraiser (A) and Interaction (P×A) Factors
  • The Study Method and Interaction (Part × Appraiser) Effect
  • Chapter Review Questions
  • Review Questions
  • Before-Thirty Multiple Choice Questions-(Hand-out)
  • After-Fifty Multiple Choice Questions
  • Appendix
  • Appendices
  • Appendix A - The Standard Procedure
  • Appendix B - Factors for Varying Sample Sizes
  • Appendix C - Derivation of Repeatability and Kl and K2 Factors
  • Appendix D - Derivation of the Calculation of Reproducibility
  • Appendix E - How to Address Variation Within a Sample
  • Appendix F - Use of Control Charts
  • Appendix G - Relationship of Process Capability and Measurement Capability Indices
  • Appendix H - The Effect of R&R on Process Capability
  • Appendix I - Factors for Control Chart
  • Appendix J - Table of Bias Factors (Attribute Gage Studies)
  • Appendix K - Area Under the Normal Curve
  • Appendix L - Repeatability calculation: Value of d2 and K1
  • Appendix M - Reproducibility calculation: Value d2 and K2
  • Appendix N - Part-to-part variations: Value of d2 and K3.
  • Appendix O - Values of A2, D3 and D4 factors for computing control chart limits
  • Appendix P- Percentage Points of the t Distribution
  • Appendix Q- Percentage Points of the F Distribution
  • Terminology
  • Definitions of common terms In Gage R&R Studies
  • Other important terminology
  • Glossary of Symbols
  • References

Programme scope, delivery arrangements and suitability are confirmed during enquiry.