Statistical Process Control and Six Sigma

Question 1

PDCA

Answer 1

Plan Do Check Act

DO - {P_i}_i=1,2,…

ACT - {X_i}_i=1,2,…

Sequence of processes {P_i}

and Resulting Products {X_i}_i

such that P_i-1 < P_i implies X_i-1 < X_i

Question 2

Statistical Process Control

Answer 2

• Understanding the process
• Understanding the causes of variation
• Eliminiation of the sources of special cause variation

Question 3

Usage Control Charts

Answer 3

1 Select process

2 Identify product or process characteristics that describe process performance

3 Select the appropriate type of control chart

4 Measure process performance over a period of time

5 Use appropriate calculations based on measurement data to determine center lines and control limits for performance characteristics

6 Plot measurement data on control charts

7 Are all measured values within limits and distributed randomly around centerlines?

8 Process is stable; continue measuring

9 Process is not stable

10 Identify and remove assignable causes (Back to 4)

Question 4

Common questions for investigating an out-of-control process

Answer 4

• Are there differences in the measurement accuracy of instruments/medhotds used?
• Are there differences in the methods used by different personnel?
• Is the process affected by the environment?
• Has there been a significant change in the environment?
• Is the process affected by predictable conditions?
• Were any untrained personnel involved in the process at the time?
• Has there been a change in the soruce for input to the process?
• Is the process affected by employee fatigue?
• Has there been a change in policies or procedures?
• Is the process adjusted frequently?
• Did the samples come from different part of the process? Shifts? Individuals?
• Are employees afraid to report ‘bad news’?

Each team should address each ‘Yes’ answer as a potential source of a special cause.

Question 5

Fishbone Chart

Answer 5

Question 6

Six Sigma

Answer 6

• A business philosophy focusing on continuous improvement
• Methodology for improving key processes
• 2 key methodologies
  
  • DMAIC
  • DMADV
• DMAIC is used to improve an existing business process
• DMADV is used to create new product or process designs for predictable, defect-free performance
• Is based on the following key underling principles of statistical thinking:
  
  • Everything is a process
  • All processes have inherent variability
  • Data is used to understand variation and to drive decisions to improve the process

Question 7

Defects per Million Opportunities

Answer 7

A six sigma process is one in which 99.99966% of the products manufactured are statistically expected to be free of defects (3.4 defects per million)

Question 8

DMAIC

Answer 8

A general-purose problem solving methodology

Problem or goal statement (Y)

Define

• Refine problem and goal statements
• Define project scope and boundaries

Measure - Analyze - Improve - Control

• An improvement journey to achieve goals and resolve problems by discovering and understanding relationships between process inputs and outputs, such as.

Y = Effectiveness of inspections
= f(size, complexity, duration, reading technique)

Question 9

DMAIC Roadmap

Answer 9

Define:

• Define Project Scope
• Establish formal project

Measure:

• Identify needed data
• Obtain data set
• Evaluate data quality
• Summarize & baseline data

Analyze:

• Exploure data
• Characterize process & problem
• Update improvement project scope & scale

Improve:

• Identify possible solutions
• Select solution
• Implement (pilot as needed)
• Evaluate

Control:

• Define control method
• Implement
• Document

Question 10

Revising and Updating Control Limits

Answer 10

• Revising and updating control charts both involve recalculating the control limits but for different reasons and somtimes in different ways.

Question 11

Revising Control Limits

Answer 11

• You use the initial or trial limits for an onoing chart to omit the following from calculations:
  
  • Unrepresentative data
  • Special causes

Question 12

Updating Control Limits

Answer 12

• You use additional, more recently collected data to re-compute the limits based on the following changes:
  
  • More data is available
  • The process has shifted
  • A deliberate change ahs been made to the process

Question 13

Quartiles

Answer 13

A quartile is any of the three values which divide the sorted data set into 4 equal parts, so that each part represents 1/4th of the sample or population.

i.e. Sorted data can be scatter plot, bell curve, box plot etc

Question 14

Box Plot

Answer 14

Lower tail, Upper tail

Median

Lower quartile, Upper quartile

Scale

|_____|_____|_____|_____|_____|