WELCOME TO MSA TRAINING PROGRAM NATHAN & NATHAN CONSULTANT - - PDF document

12-12-2018 1

WELCOME

TO

MSA TRAINING PROGRAM

NATHAN & NATHAN CONSULTANT PVT. LTD.

12-12-2018 2

MEASURED DATA VARIABLE ATTRIBUTE

MEASUREMENT DATA

12-12-2018 3

DATA TYPES

• Variables data

Interval and proportional scale Eg: 0C, kg, N

• Attribute data

Nominal and ordinal scale Eg: good/bad, stage, rank

VARIABLE DATA

• The measurement that can be meaningfully divided into

infinite increments of precision

• Characterizes a product or process feature in terms of size,

weight, time, etc., Example Height, Weight, time, temperature, dimensions etc

12-12-2018 4

ATTRIBUTE DATA

• Counts the frequency of occurrence [No. of times of success
• r failure]
• Is not capable of being meaningfully sub divided into precise

increments

• Example – Good/bad, pass/fail, counts such as defects,
• utcomes when a coin is tossed and so on

USES OF MEASUREMENT DATA

• Decision to adjust a manufacturing process
• Determine the significant relationship between two
• r more variables using statistical studies
• The result of analytical studies depends upon the

Quality of data produced by measurement system

12-12-2018 5

MEASUREMENT SYSTEM

Process to assign a number or decision to a characteristic Process

• Appraiser
• Material
• Checking Method
• Instrument
• Environment

S Standard W Work piece I Instrument P Person/Procedure E Environment

MSA - DEFINITION

Measurement System Analysis (usually referred to as MSA) is a structured procedure which we use to assess the ability of a measurement system to provide good quality data There are several types of MSA and which type is required will depend on the type of data being measured and the influences on the system

AS 13003 : MEASUREMENT SYSTEMS ANALYSIS (MSA): The process of evaluating the fitness for purpose of a measurement system, including methods such as Gauge R&R, Attribute Agreement, Bias assessment, Stability assessment, Linearity assessment, etc.

12-12-2018 6

CONSIDERATIONS DURING MSA

• Factors that need to be evaluated include:

 Environment – temperature, humidity,

contamination, vibration, electromagnetic radiation, etc.

 Location – different buildings, sites, etc.  Part variation that will affect the measured

value (surface finish, flexibility, shape, size, etc.)

 People – shift patterns, times of the day,

experience levels

 Process – fixtures, probes, accessories, etc.

WHEN TO CONDUCT MSA

• MSA shall be conducted as part of New Product Introduction to

validate the measurement system prior to production.

• Situations where MSA should be repeated, include
• Changes to gauge design
• Refurbishment/repair
• Environment
• Product design change to the feature being measured, etc.

12-12-2018 7

CONSTITUTION OF TOTAL VARIATION

• Total variation
• Manufacturing process variation
• Part to part variation
• Within part variation
• Measurement system variation
• Equipment variation
• Appraiser variation
• Appraiser part interaction

CONSTITUTION OF TOTAL VARIATION

12-12-2018 8

STATISTICAL PROPERTIES OF MS

Classification Measurement system errors are:

Variable Measurement Attribute Measurement

Bias

Risk Assessment or Hypothesis Analysis

Linearity Stability Repeatability Reproducibility

RESOLUTION

• Ability to detect the small changes in process
• MSA Requirement
• Resolution 1/10th of Total Tolerance
• In case of high Process Capability it can be

1/10th of Process variation

12-12-2018 9

ACCURACY & PRECISION

ACCURACY Captured by

• BIAS
• LINEARITY
• STABILITY

PRECISION Captured by

• REPEATABILITY
• REPRODUCIBILITY

ACCURACY & PRECISION

ACCURACY

• Closeness to reference or

master value

• Required where two or

more MS measuring a same characteristic

• Same parameters are

checked at Suppliers end

• r at Customer end

PRECISION

• Ability of MS to repeat the

same reading

• Required where MS is

repeatedly used to assess and adjust the process

• In process inspection as

per control plan

12-12-2018 10

BIAS

ACCURACY ERROR OR BIAS:

• The difference between the observed average value of

measurements and a known true value or accepted reference value.

• Accuracy Error is measurement error not captured in most MSA

evaluations and must be considered in the overall assessment of the measurement system

12-12-2018 11

DISCRIMINATION (NUMBER OF DISTINCT CATEGORIES)

The number of groups within the process data that the measurement system can discern is used as a quality check of the measurement system

12-12-2018 12

REPEATABILITY & REPRODUCIBILITY

Repeatability

• One Appraiser
• One Equipment
• Same part
• Several trials

This Variation is represented by Equipment Reproducibility

• Same equipment
• Same Parts
• Several trials
• Different Appraiser

This Variation is represented by Appraiser

X BAR & R CHART METHOD – ACCEPTANCE CRITERIA

GRR Decision Comments Under 10 % Generally considered to be an acceptable MS Recommended, especially useful when trying to sort or classify parts or when tightened process control is required 10% to 30 % May be acceptable for some application Decision should be based upon, for example, importance of application measurement, cost of measurement device, cost of rework or repair Should be approved by the customer Above 30% Considered to be not acceptable Every effort should be made to improve the MS. NDC ≥ 5 MS is acceptable

12-12-2018 13

R CHART INTERPRETATION

ATTRIBUTE MEASUREMENT

• COMPARES EACH PART TO A SPECIFIC SET OF LIMITS AND ACCEPTS

THE PART IF THE LIMITS ARE SATISFIED

• IS DESIGNED TO ACCEPT/REJECT A SET OF MASTER PARTS
• CANNOT INDICATE HOW GOOD OR HOW BAD A PART IS,ONLY

WHETHER THE PART IS ACCEPTED OR REJECTED (PASS/FAIL)

• VISUAL STANDARDS MAY RESULT IN 5 TO 7 DISTINCT

DATA CATEGORIES. MSA MANUAL DOES NOT PRESCRIBE ANY METHOD OF EVALUATION OF THE SAME

• RISK ANALYSIS METHOD USING HYPOTHESIS TEST IS USED TO

EVALUATE THE MS WHEN IT IS NOT FEASIBLE TO GET SUFFICIENT PARTS WITH VARIABLE REFERENCE VALUES

12-12-2018 14

RISK ASSESSMENT KEY STEPS

• Take 50 parts representing entire process variation having

bad, marginally bad, good & marginally good parts

• Mark as 1 to 50
• Select 3 appraisers
• Conduct study in a random manner and record the decisions,

1 as OK and 0 as Not ok

12-12-2018 15

Thank You Thank You

magadihari@yahoo.com LinkedIn – hariprakash-3pc Phone – +91 9480202408 https://mot3xiot.org