Vous êtes sur la page 1sur 97

Total Quality Management - 5.

Total Quality Management

Session 5:

Process Improvement and Basic tools for quality management

Total Quality Management - 5.1

Learning Objectives After completing this session you should be able to explain what a work process is and why it is so describe a systematic approach for improving business processes discuss common pitfalls encountered when working on a process improvement project use basic tools for improving process

Total Quality Management - 5.2

Required Readings Chapter 6, Textbook (chapters 7, 8 optional)

Suggested reading: The Golden Rules of Process Redesign by Roger Rupp and James Russell

Total Quality Management - 5.3

Session Agenda

Process introduction or understanding process Your examples on process Process improvement - TQM Your examples on improvement Improvement process models and pitfalls Your ideas on process Tools for quality management Your experience on tools

Total Quality Management - 5.4


Input Process Simple Process Supplier Work Group Customer Output

Work Process Model

Requirements Product / Service Feedback

Total Quality Management - 5.5

Process Definition

Thomas Davenport, in Process Innovation: ". . . a structured, measured set of activities designed to produce a specified output for a particular customer or market." Michael Hammer and James Champy, in Reengineering the Corporation: ". . . a collection of activities that takes one or more kinds of input and creates an output that is of value to the customer."

Total Quality Management - 5.6

Process Definition

James Harrington, in Business Process Improvement "Any activity or group of activities that takes an input, adds value to it, and provides an output to an internal or external customer." Joseph M. Juran, in Juran on Planning for Quality ". . . a systematic series of actions directed to the achievement of a goal."

Total Quality Management - 5.7

Four groups of people are involved in the operation and improvement of processes: Customers Work Group Supplier Owner

Total Quality Management - 5.8

Why process?

We are quality talking; then conformance, variation, gapsbetween gapsbetween customer expectation and what we offer. A customer focuses on value they get (along the visible line when you have to sketch service blueprint last term), he does not care of each business function. Colleagues as customers Responsiveness to customer inquiries. then...

Total Quality Management - 5.9

Why process?

with process and detailed elements we have: sequence (quality) who produce the outputs (accountability) boundary clarification (responsibility and quality) process owner (accountability) input & output those help promote quality

Total Quality Management - 5.10

Examples of processes in system view

Production process
Inputs Process
- Assembly

- Customer receives product

- Component


- machines - Operators

of components into final product - ship to customer

Information on product performance

Total Quality Management - 5.11

Hiring process Inputs

- Resume

- Synthesize

Employee to fill vacant position

- Interview - Information from references, former employees, and schools

information - Evaluate information - Make decision - Hire applicant

Inform personnel decision-maker of strengths and weaknesses of employee to improve future hiring efforts

Total Quality Management - 5.12

Identifying processes Why difficult?  We are talking more on function  We have no chance to get out of out functional department  It is easier to discuss on function, task, step, or operation. Its hard to talk about process in a business  Functions have names, processes have no names.  Disputes often arise when talking process.  Process hides in stream of activities.

Total Quality Management - 5.13

Identifying process How to?  Anchor the analysis of a business process to something, some one, tangible, or customer.  Ignore process name  Set boundaries anywhere on stream to analyse (tips: where state changes, transfer of ownership, hand-offs) hand beware of loop in process.

Total Quality Management - 5.14

SAV E-MBA intake process E  

Input Output Process identifying Anchors setting Get related people involved Problem Identifying tools Process improvement

Total Quality Management - 5.15

What are the boundaries of the process?

The boundaries of a process may not be limited in a function Intrafunctional process is a process performed within a single function (e.g. preparation of wage sheets) Interfunctional process is a process involving more than one function (e.g. product development and planning)

Total Quality Management - 5.16

Defining Key Process

Six questions help in the identification of key process:

1. Which products and services are most important to the customers? (implication-priority, general (implicationview of customer requirement) 2. What are the processes that produce these products and services? (general view) 3. What are the key ingredients that stimulate action in the organization, and what are the processes that convert these stimuli to output? (more detailed of process elements)

Total Quality Management - 5.17

4. Which processes have the highest visibility with customers? (priority) 5. Which processes have the greatest impact on customercustomer-driven performance standards? (key process identified) 6. Which processes do performance data or common sense suggest have the greatest potential for improvement? (improvement opportunities)

Total Quality Management - 5.18

Continual Improvement of Work Process

Continuous improvement is a systematic approach to the closing of gaps between customer expectations and the characteristics of process outputs.


Flowcharts Cause and Effect Diagrams Control Charts

Total Quality Management - 5.19


Check Sheets

Pareto Diagrams

Scatter Diagrams

Total Quality Management - 5.20


What is a flowchart?

A flowchart is a pictorial summary of the flows and decisions that comprise a process

Flowchart Symbols

Basic input/output symbol

The general form that represents input or output media, operations, or processes is a parallelogram

Basic processing symbol

The general symbol used to depict a processing operation is a rectangle

Decision symbol

A diamond is the symbol that denotes a decision point in the process. It includes attribute type decision such as pass fail, yes no. It also includes variable type decisions such as which of several categories a process measurement falls into. A line with an arrowhead is the symbol that shows the direction of the stages in a process. The flowline connects the elements of the system. The general symbol used to indicate the beginning and end of a process is an oval.

Flowline symbol

Start / stop symbol

Total Quality Management - 5.22

Types of Flowcharts

Systems Flowchart

A system flowchart is a pictorial summary of the sequence of operations that make up a process. It shows whats being done in a process. Applications: manufacturing process, quality of design study, quality control chart, organization chart, flows of billing, accounting, purchasing, etc.

Total Quality Management - 5.23

Example: Quality of Design Study




Design of trial prototype specifications

Good Evaluation of prototype

Trial production of units

Overall Good Production evaluation of design trial accepted production units


Total Quality Management - 5.24

Layout Flowchart

A layout flowchart depicts the floor plan of an area, usually including the flow of paperwork or goods and location of equipment, file cabinets, storage areas, and so on. These flowcharts are especially helpful in improving the layout to more efficiently utilize a space.

Total Quality Management - 5.25

Constructing a Flowchart We should follow the rules:

Rule 1: The most important rule is that the right 1: people must be involved in making the chart. Rule 2: All members of the group must participate. 2: The use of an independent facilitator will be a great benefit. Rule 3: All data must be visible and 3: understandable to all the people all the time.

Total Quality Management - 5.26


Rule 4: Enough time must be allotted. 4: Rule 5: The more questions everyone asks, the 5: better. Questions are the key to the flowcharting process. What is the first thing that happens? What is the next thing that happens?

Total Quality Management - 5.27

Benefits of a flowchart

It functions as a communication tool It functions as a planning tool It provides an overview of the system It defines the roles It demonstrates interrelationships It promotes logical accuracy It facilitates trouble-shooting troubleIt documents a system

Total Quality Management - 5.28

2. CAUSE AND EFFECT DIAGRAM ~ Fishbone diagrams ~ Ishikawa diagrams

Information is the key to improving processes A cause and effect diagram is the key to gathering information

The technique consists of defining an occurrence or problem (effect). Once the effect is defined, the factors that contribute to it (causes) are delineated Relationship between causes and effects

Total Quality Management - 5.29

Constructing a C&E Diagram

1. State the problem

Effect (Quality Problem)

2. Identify major causes

Personnel Machine Measurement

Quality Problem
Environment Materials Method

Total Quality Management - 5.30

3. Brainstorm subcauses
Personnel Knowledge Absenteeism Machine Measurement Repeatability Capacity Sample size

Capability Safety

Quality Problem
Temperature Light Quality Working Environment Materials Effectiveness Method Standardization

Advantages and disadvantages of C&E diagram

Total Quality Management - 5.31

Advantages: Forces preparation of the process flow diagram Forces consideration of the entire process as potential causes of a problem Identifies alternative work procedures Educational for members not familiar with the entire process Easy to use since most group members will be familiar with the process Can be used to anticipate process problems by focusing on sources of variability.

Total Quality Management - 5.32

Disadvantages Easy to overlook potential causes (e.g., materials or measurements) since people may be too familiar with the process. Hard to use on long, complicated processes

Total Quality Management - 5.33

Investigation of Potential Causes The C&E diagram is not intended to identify which potential causes are root causes. Thus, it is necessary to identify whether a potential cause is a root cause of a problem

C&E diagram


Hypothesis generation

Design studies

Simple statistical tools

C&E diagram

Select most likely potential causes. Evaluate each against problem specification

Determine which data would identify potential cause as a root cause

Collect appropriate data

Analyze data


Potential cause determined to be a root cause

Yes No
Problem eliminated Correct problems and monitor results

Stop Process for identifying whether a potential cause is a root cause of a problem

Total Quality Management - 5.35

3. CONTROL CHARTS The major kinds of control charts: Attribute Data: Charts for nonconforming units ( p chart) Charts for counts of nonconformities (c chart) Variable Data: R chart X-bar chart Process Capability refers to the inherent variability of process output relative to the variation allowed by the design specifications.



Total Quality Management - 5.36

Histograms are a graphic display of the number of times a measurement has occurred at a particular value or within an interval of the measurement scale. scale.

Typical Patterns of Variation

Total Quality Management - 5.37


Double peaked




Comb (a) (b)



Isolated peaked

Edge peaked

Constructing a Histogram

Total Quality Management - 5.38

Number of measurements (N) < 50 50 100 100 250 > 250

Number of scale intervals 57 6 10 7 12 10 20

Total Quality Management - 5.39

a. Quick Method A histogram can be constructed using the following steps: 1. Find the largest and smallest measurement 2. Construct the measurement scale between the largest and smallest measurement
Smallest measurement Largest measurement

Total Quality Management - 5.40

3. Place a slash \ on the scale in step 2 for each occurrence \ of a measurement. Every fifth occurrence should use a longer slash ( \ \ \ \ ) . A space should be skipped between group of five

4. Evaluate whether there are too many intervals in the main body of the histogram. If there are too many intervals, the interval method should be used.

Total Quality Management - 5.41

b. Interval Method A histogram can be constructed using the following steps: 1. Find the largest and smallest measurement 2. Choose a value for the number of intervals using the preceding guidelines for the number of measurements to be plotted in the histogram 3. Compute the interval size:
l ! interval size ! ( largest measuremen t - smallest measuremen t ) number of scale intervals

Total Quality Management - 5.42

4. Round l to a convenient number. For example, number. 0.512 becomes 0.5 5. Choose a convenient number slightly smaller than or equal to the smallest measurement; call it measurement; Xlow. Compute the following intervals: intervals: low.  Xlow Interval 1  Xlow + l

Xlow + 2l Xlow + 3l

Interval 2 Interval 3

Total Quality Management - 5.43

6. Construct a scale to be used in plotting that corresponds to the intervals computed in step 5:
Smallest measurement Largest measurement

Xlow Xlow + 1l

Xlow + 2l Xlow + 3l

7. Place a slash \ in the interval in step 6 for each occurrence of a measurement. measurement.


Xlow + 1l

Xlow + 2l

Xlow + 3l

Total Quality Management - 5.44

For either the quick or the interval method, shaded bars are often drawn around the slashes. A slashes. frequency (number of occurrences) scale is typically used. used.


Xlow + 1l

Xlow + 2l

Xlow + 3l

Total Quality Management - 5.45

Histogram from Control Charts There are three types of histograms that can be obtained form typical X-bar and R chart data: data:
Sample 1 2 3 N 1 Subgroup (n = 5) 2 3 4 5 X R Individuals histogram

means histogram ranges histogram

Total Quality Management - 5.46


Check sheets are used for collecting or gathering data in a logical format. The most important purpose of check sheet is to enable the user(s) to gather and organize data in a format that permits efficient, easy analysis.

Total Quality Management - 5.47

Types of check sheets

Attribute check sheet. Gathering data about sheet. defects in a process is necessary for stabilization, improvement, and innovations of the process. Variables check sheet. Gathering data about sheet. a process also involves collecting information about variables, such as size, length, weight, temperature, etc. These data are best etc. represented by organizing the measurements into a frequency distribution on a variables check sheet. sheet.

Total Quality Management - 5.48

Defect location check sheet. Another way to sheet. gather information about defect in a product is to be used a defect location check sheet. This is a picture of a product or a portion of it on which an inspector indicates the location and nature of the defect. Checklist. Checklist. Checklist is used under operational conditions to ensure that all important steps or action have been taken. Its purpose is for guiding taken. operations, not for collecting data. The preflight data. checklist that a commercial airline pilot must complete is a good example. example.

Total Quality Management - 5.49

Checklist for Evaluating a Check Sheet

Check if true S S Questions 1. Is an appropriate title given? Are department or operation names and numbers specified? Is the part name and number given? 2. Is the individual who completed the check sheet identified (who)? If more than one person completed that check sheet, is each person identified? 3. Is the appropriate method of collecting the data understood (what)? Are brief instructions for completing the check sheet written on the collection form? 4. Is the process location where parts are collected identified (where)? Is the sampling method defined and understood? 5. Is the time in which parts are sampled identified (when)? 6. Do individuals who are collecting the data understand why the data are being obtained (why)? 7. Is the evaluation or gagging method understood (how)? Have gage studies have been performed to evaluate the bias and repeatability of gages? Is a gage identification number required?



8 Are all stratification variables identified, such as machine, spindle, pallet and fixture? Are such factors as vendor, test hand, furnace, die, tool and material identified? Are any lot identification numbers records? 9 9Will total number of parts evaluated be recorded (how many)? Is the number of parts produced during the evaluation period indicated? 10 Do individuals understand how to record multiple defects? Will all defects on a part be evaluated? Do individuals understand how multiple occurrences of the same defect on a unit will be recorded (e.g., is 1 missing screw or 10 missing screws on a unit recorded as a single defect)? 11 Are data recorded on check sheets in a manner analogous to the gage measurement scales? 12 Is coded gage data used if appropriate? Is the relationship between the coded gage data and the actual measurement state as: Actual value = offset + scale x gage value


13. Has a trial run of the check sheet been attempted? Has a trial analysis of the data been attempted? 14. Is there sufficient space for the user of the check sheet to write in and make comments concerning unusual occurrences?

Total Quality Management - 5.51


Pareto focused attention on the concept of the vital few versus the trivial many. Pareto is a tool used to identify and prioritize problems for solution. The 80-20 rule is based on an application of 80the Pareto principle. For example, 80 percent of sales come from 20 percent of customers, or 80 percent of dollar inventory is accounted for by 20 percent of the items.

Total Quality Management - 5.52

Constructing a Pareto Diagram 1. Establish categories for the data being analyzed. 2. Specify the time period during data will be collected. 3. Construct a frequency table arranging the categories from the one with the largest number of observations to the one with the smallest number of observations.

Total Quality Management - 5.53




Relative %

(1) (2) (3) (4) 1. XX Y1 C1 C1/C x 100 = P1 2. XX Y2 C2 C2/C x 100 = P2 3. XX Y3 C3 C3/C x 100 = P3 Total C 100% N = Total number of parts or items inspected

Cumulative % (5) P1 P1+P2 P1+P2+P3 100%

% of Total Inspected (6) C1/N x 100 C2/N x 100 C3/N x 100


Total Quality Management - 5.54

4. Construct a Pareto diagram. Part A, Pareto bars: Select the most appropriate scale to plot the Pareto bars: Frequency = Column 3 Relative % = Column 4 % of total inspected = Column 6

Part B, cumulative percentages: Select the line plot or stacked bar format to graph the cumulative percentages and plot the data in column 5.

Example: Consider the following defects for a machined part

Total Quality Management - 5.55

Defects Undersized A hole Porosity Nicks Concentricity diameter A-F Parallelism surface Z-W Oversized A hole Others
Total defects: Number inspected:

Quantity 224 52 149 46 58 5 23

C = 557 N = 2217

Total Quality Management - 5.56

Pareto Analysis Work Sheet Defects (1) Undersized Nicks Parallelism Porosity Concentricity Oversized Others Total Signal Frequency Relative Cumulative % % (2) (3) (4) (5) A 224 40 40 B 149 27 67 C 58 11 78 D 52 9 87 E 46 8 95 F 5 1 96 G 23 4 100 557 100% 100% % of Total Inspected (6) 10.1 6.7 2.6 2.3 2.1 0.2 1.0 557/2217 = 25.1

Pareto Diagram of Defects for a Machined Part

250 200 Num ber of Defec ts 150 100 50 0

Total Quality Management - 5.57

100 90 70 60 50 40 30 20 10 0 Cum lative P erc entage 80


s it y



it y





Ov er





Ot he

li s

r si


ll e

tr ic


s iz


Total Quality Management - 5.58

Pareto diagram with stacked bars











Cumulative percentage

Total Quality Management - 5.59

Pareto Diagram of defects for a machined part after the improvement




Total Quality Management - 5.60

Cost Pareto Diagram

Sometime Pareto diagram can have more impact when problems or defects are represented in terms of their dollar costs. costs.
Pareto Analysis Cost Work Sheets: Cost per Defect Work Sheet Signal (2) A B C D E F Frequency (3) 224 149 58 52 46 5 534 Cost per unit ($) (4) 0.25 0.10 1.25 0.75 1.25 1.25 Cost of defect (5) 56.00 14.90 72.50 39.00 57.50 6.25 246.15 Rank (6) 3 5 1 4 2 6

Defect (1) Undersized Nicks Parallelism Porosity Concentricity Oversized Total

Total Quality Management - 5.61

Pareto Analysis Cost Work Sheet Defect (1) Parallelism Concentricity Undersized Porosity Nicks Oversized Total Signal (2) C E A D B F Frequency (3) 58 46 224 52 149 5 534 Cost of defect (4) 72.50 57.50 56.00 39.00 14.90 6.25 246.15 Relative cost Cumulative % % (5) (6) 29 29 23 52 23 75 16 91 6 97 3 100

Pareto Diagram for Defect Cost 35 C o s t p e r c e n ta g e ($ ) 30 25 20 15 10 5 0 C E A D B

Total Quality Management - 5.62

100 80 70 60 50 40 30 20 10 0 F C u m u la tiv e p e r c e n ta g e 90

Total Quality Management - 5.63

Warning about using Pareto Analysis

y y y

y y

Ignore customers requirements Data collected from unstable processes Identify the easy-to-find problems as the easy-tovital few Ignore the trivial many problems Short time for Pareto analysis

Relationship of attribute data analysis tools

Tool Purpose

Check Sheet

Collect Data

Pareto diagram

Determine top problems

Cause and Effect Diagram

Problem A

Obtain potential causes of a problem

p chart and c chart

Assess process stability and improvement

Total Quality Management - 5.65


Scatter diagram provides a method to evaluate these interrelationships between two characteristics Understanding the relationship between characteristics increases our ability to control a process and detect problems.

Total Quality Management - 5.66

Types of Patterns
Characteristics 2 Characteristics 2 Characteristics 2

Characteristics 1

Characteristics 1

Characteristics 1

Characteristics 2

Characteristics 2


Characteristics 1

Characteristics 1



Total Quality Management - 5.67

Two mistakes are commonly made:

First, it is assumed that if correlation exists between two characteristics, one characteristic in some way causes the value of the other characteristic This is not true Correlation does not imply causation. The second mistake commonly made in using scatter plots is to consider too narrow a range of the characteristics being evaluated. evaluated.

Total Quality Management - 5.68

Constructing a scatter diagram

Sampling unit 1 2 N Characteristic 1 X1 X2 XN Characteristic 2 Y1 Y2 YN

It is a good practice to use a sample size of at least N = 30 50 sampling units. units. 1.Select the characteristic (label 1), if any, that for some reason is thought to predict the value of the other characteristic (label 2).

2. Plot the values for each sampling unit. If repeated unit. values occur in a particular location, use the following plotting convention: convention:
Single value Two values Three values Four values

3. Use the types of patterns above to evaluate the relationship between the characteristics. The next characteristics. two procedures can be used to test for the presence of correlation

Total Quality Management - 5.70

Procedures for testing correlation

Sign Test for correlation

1. Rank from lowest to highest the values of each characteristic. characteristic. For characteristic 1, rank the X values, and for characteristic 2 rank the Y values. values. 2. Find the median value for each characteristic. characteristic. 3. Prepare a scatter plot. Draw the dashed median plot. lines on the scatter plot, and label the lines and four quadrants. quadrants.

Total Quality Management - 5.71

Median 1 Characteristic 2

2 3

1 4

Median 2

Characteristic 1

4. Count the number of points in each quadrant, and compute the following quantities: A = number of points in quadrants 1 and 3 B = number of points in quadrants 2 and 4 Q = number of points on either median line Total N = A + B + Q

Total Quality Management - 5.72

5. Compute the test value C, where C = smaller of A and B = min (A, B) Sample size for the test SS = N Q If C is no greater than the table value c for the given value of SS, the characteristics are correlated. The maximum value c of computed C needed to conclude that a correlation exists:

Total Quality Management - 5.73

SS 20-22 23-24 25-27 28-29 30-32 33-34 35-36 37-39 40-41 42-43 44-46 47-48 49-50 51-53

c 5 6 7 8 9 10 11 12 13 14 15 16 17 18

SS 54-55 56-57 58-60 61-62 63-64 65-66 67-69 70-71 72-73 74-76 77-78 79-80 81-82 83-85

c 19 20 21 22 23 24 25 26 27 28 29 30 31 32

SS 86-87 88-89 90-91 92-93 94-96 97-98 99-101 110 120 130 140 150 200

c 33 34 35 36 37 38 39 43 48 52 57 61 84

Total Quality Management - 5.74

Correlation Coefficient (r)

Correlation Coefficient (r) shows the linear relationship between characteristics:

r = 0.8 1.0 r=0 r = -0.8 -1.0

high positive correlation no correlation high negative correlation

Total Quality Management - 5.75


Total Quality Management - 5.76

 The seven-step process improvement model sevenintroduces a systematic approach for applying quality management to any type of process. The approach helps build a fundamental understanding of the business processes before attempting to improve them.  The seven steps are described in a straightforward, linear fashion for clarity. Actual application will often be complicated by the need to recycle back to earlier steps as new information is gained and earlier hypotheses are found to be incomplete or incorrect.

1 2 3 4 5 6 7

Define the Problem

Study the present system

Analyze the potential causes

Plan and implement a solution

Evaluate the results

Standardize the improvement

Reflect on process and develop future plans

Process Improvement Model

Total Quality Management - 5.78

Step one: Define the Problem one:


Identify the gap between what is and what should be Document the importance and relevance of the problem List the customers key quality characteristics State how closing the gap will benefit customers Identify the outputs Identify processes producing these outputs Identify the process owner

Total Quality Management - 5.79

Step two: Study the present system two:


Develop flowchart of the processes Identify the process participants Determine what data to be used to measure progress Customer satisfaction Cost of quality Output delivered Process variables (what, where, to what extent and who)

Total Quality Management - 5.80

Design data collection instrument, develop any operation definitions needed to collect data Collect data, summarize the information learned about variables effect on the problem

Total Quality Management - 5.81

Step three: Analyze the potential causes three:

Use the data collected and the experience of the process participants to identify conditions that might lead to the problem Analyze the conditions of interest and decide on most likely causes If possible verify the cause through observation or by directly controlling variables

Total Quality Management - 5.82

Step four: Plan and implement a solution four:


Develop a list of solutions to be considered Assess the feasibility of each solution, likelihood of success and potential adverse consequences Indicate clearly why a particular solution is chosen Plan the implementation (Any pilot project? Who is responsible? Any training required?) Implement the preferred solution

Total Quality Management - 5.83

Step five: Evaluate the results five:

Collect data on measurements used to measure progress decided in step two Analyze the results. Determine if the solution tested was effective Repeat prior steps as necessary

Total Quality Management - 5.84

Step six: Standardize the improvement six:

Develop a plan for institutionalizing the improvement and assign responsibilities Implement the plan, check to see that it has been successful

Step seven: Reflect on process seven: and develop future plans

Total Quality Management - 5.85

Determine if the remaining gap should be narrow further Identify related problems that should be addressed Summarize lessons learned during the process improvement project, make recommendations for future teams Celebrate the contributions of the team.

Total Quality Management - 5.86

Illustration of the Seven-step Method to E-MBA SevenE

7-step method problem solving for E-MBA is a Esystematic one for improvement rather than using the tools alone and intermittently. In these steps, using the 7 tools of improvement for finding causes and solutions Where and which tools to use?

There are other methods of problem solving but the tools are useful there as well.

Total Quality Management - 5.87

Other approaches to Process Improvement

The PDCA Cycle

Steps 1,2 & 3

P (Plan) A (Act)

D (Do) C (Check)

Step 4

Steps 6 & 7

Step 5


Total Quality Management - 5.88

Plan-Do-CheckPlan-Do-Check-Act Procedure

Plan. Recognize an opportunity and plan a change. Do. Test the change. Carry out a small-scale smallstudy. Check. Review the test, analyze the results and identify what youve learned.

Total Quality Management - 5.89

Plan-Do-CheckPlan-Do-Check-Act Procedure

Act. Take action based on what you learned in the study step: If the change did not work, go through the cycle again with a different plan. If you were successful, incorporate what you learned from the test into wider changes. Use what you learned to plan new improvements, beginning the cycle again.

Total Quality Management - 5.90

When to Use Plan-Do-Check-Act Plan-Do-Check  


As a model for continuous improvement. When starting a new improvement project. When developing a new or improved design of a process, product or service. When defining a repetitive work process. When planning data collection and analysis in order to verify and prioritize problems or root causes. When implementing any change.

Total Quality Management - 5.91

The Philip Crosby Associates Method for eliminating Non-conformances Non1 Define Situation

Fix Situation

Identify Root Causes

Corrective Action

Evaluate and Follow Up

Xeroxs Quality Improvement Process

1 2 3 Identify Output Identify Customers Identify Requirements

4 Develop Specifications 5 Identify Process Steps 6 7 Select Measurements Determine Capability OK ? Produce Output 8 Evaluate Results OK ? 9 Recycle Problem Solve Problem Solve

Total Quality Management - 5.93

Organizations have customized the process improvement model for their own use. Some modified the terminology. Some reduced the number of steps by combining elements, and others increased the number of steps to explicitly consider certain elements. These variants of the process improvement model are all somewhat different from one another but in essence they are all very similar.

Checklist of common Pitfalls to process improvement

Total Quality Management - 5.94

Although numerous problems will be encountered along the seven steps, those included in the checklist impact the early steps and often cause irreparable damage. Tampering Incomplete ownership Lack of expertise Imposed solution Constraints World hunger
y y y y y y

Total Quality Management - 5.95

Concluding remarks

Systematic process improvement relies on building a fundamental understanding of customers requirements, process capability and the causes for the gaps between them. Hypotheses are developed and tested and improvement gained through the continuous cycle of plan-do-check-act. This systematic approach plan-do-checkbears striking contrast to the classic short cut of problem detection and subsequent solving, an approach resembling plan-do-plan-do. plan-do-plan-

Total Quality Management - 5.96

The seven-step model was developed as a sevenuniversal guide for systematic improvement. It can be used to improve manufacturing as well as nonnonmanufacturing processes and helps assure application of the plan-do-check-act strategy. plan-do-check-