PDCA(slides)
- 1. PRACTICAL HANDBOOKPRACTICAL HANDBOOK for Quality Improvement Team Dr Ibrahim Nor 1
- 2. COMPETENCY IMPROVEMENT TEAM USING PDCA CYCLEUSING PDCA CYCLE Dr Ibrahim Nor 2
- 3. THE PDCA CYCLE PLAN CHECK DO ACTION PDCA Dr Ibrahim Nor 3
- 4. SELECT THE THEME From:- • Customer complaints. • Department’s objectives. • The big ‘Q’ • The small ‘q”PLAN Dr Ibrahim Nor 4
- 5. PLAN THE SCHEDULE • List all the 8 steps in the PDCA cycle. • Indicate the expected time frame. • Use the milestone chart. PLAN Dr Ibrahim Nor 5
- 6. GRASP THE CURRENT STATUS Obtain and review data using checksheet. Study the effects of the problem from various perspectives. Present available data in graphs, and Pareto charts.PLAN Dr Ibrahim Nor 6
- 7. SET THE TARGET • Indicate clearly the statement of result to be achieved. • Target must be reasonable and realistic. • Use SMART concept. PLAN Dr Ibrahim Nor 7
- 8. ANALYSE THE CAUSE & DETERMINE CORRECTIVE ACTION • Prepare cause and effect diagram. • Prepare hypothesis and verify most probable causes. • Determine corrective action. PLAN Dr Ibrahim Nor 8
- 9. IMPLEMENT CORRECTIVE ACTION • Take immediate action by following exactly as planned. • Collect data on results and record any deviations. • Provide adequate training. DO Dr Ibrahim Nor 9
- 10. EVALUATE THE RESULT • Check the effect of the corrective action. • Compare results with target • Use paired Pareto Chart. CHECK Dr Ibrahim Nor 10
- 11. STANDARDISE THE PROCESS • Document as standard operating procedure (SOP). • Set parameters to monitor and control the successful process. • Ensure appropriate training on the new methods. ACTION Dr Ibrahim Nor 11
- 12. QUALITY TOOLS Dr Ibrahim Nor 12
- 13. CHECKSHEET • A Checksheet is used to gather relevant data in a simple, standardised format. The use of checksheet makes the collection of data easier and more systematic. Dr Ibrahim Nor 13
- 14. CHECKSHEET Description 1 2 3 4 5 6 7 8 9 T 1 Poor coordination and communication 25 2 No review on test run 22 3 Insufficient information 28 4 Errors in report 39 5 Inaccuracy of report 37 Total 34 17 19 10 12 29 7 13 10 151 Observed by: Frequency: Daily Checksheet on: Generation of Final Report Date: 13TH JAN 09 Location: IKRAM PAVES (Case study) Dr Ibrahim Nor 14
- 15. CAUSE AND EFFECT DIAGRAM • The Cause and Effect diagram helps us to find out all possible causes, to sort them out, and to organise their inter-relationship. Typically, the causes are brainstormed in a free-flowing session. Dr Ibrahim Nor 15
- 16. CAUSE AND EFFECT DIAGRAM MAN METHODMACHINE MATERIAL Lack Manpower Error in Report Generation Lack of training Tight Dateline Time wasted Data entry error Insufficient skill Paper wastage (Cost) Graphics module not on line No reference materials (Guide book) Unable to cater complex reports No proper evaluation of equipment (vendor) Take over of vendor No user requirement No proper planning No review during testing Redundant work Two individuals processing same data Manual to computer format Shipping delay Holiday season No inter dept communication Dr Ibrahim Nor 16
- 17. PARETO DIAGRAM • A Pareto diagram is constructed to show the relative importance of different categories in a process. The vital few is separated from the trivial many. This will ensure that focus is made on the what is important. Dr Ibrahim Nor 17
- 18. PARETO DIAGRAM Observed by: Frequency: Daily 100151Total 100.014.622No review on test run5 85.416.625Poor communication & coordination 4 68.818.528Insufficient information3 50.324.537Inaccuracy in report2 25.825.839Errors in report1 Cum % %TOTDescription Dr Ibrahim Nor 18
- 19. PARETO DIAGRAMPARETO DIAGRAM 0 20 40 60 80 100 120 140 160 Errors in report Inaccuracyin report Insufficient Information Poor communoation No Review 100% 0 50% 25.8% 75% Dr Ibrahim Nor 19
- 20. HISTOGRAM • A histogram is used to summarise the frequency of occurrence of something from a sample of data. The shape of distribution is displayed and descriptive statistics can be calculated. Dr Ibrahim Nor 20
- 22. CONTROL CHART • A control chart has a center line corresponding to the average quality at which the process should perform. An out of control situation and other unnatural patterns can be detected from the chart Dr Ibrahim Nor 22
- 24. SCATTER DIAGRAM • A scatter diagram is a special type of graph which shows the relationship between two variables. If there is an empirical relationship between the two variables it can be easily seen. Dr Ibrahim Nor 24
- 25. SCATTER DIAGRAM 0 5 10 15 20 25 0 2 4 6 8 10 12 14 16 Dr Ibrahim Nor 25
- 26. STRATIFICATION • Stratification is the technique of analysing data by separating it into several groups with similar characteristics. The data must be stratified by group to better understand the differences Dr Ibrahim Nor 26
Editor's Notes
- #5: Note: What Is the Objective of This Step? To clearly identify the problem to be resolved or the area to be improved. How Do We Identify Problem Areas or Areas to Be Improved? We do this by selecting issues from your departmental objectives, customer complaints, big Q [quality, service,delivery and cost], or small q [housekeeping, morale, training and safety]. Once you have identified an issue, select a project theme. For example if the issue involves:- Quality, then we can have a theme that reads, “Providing quality products is first amongst equals.” Service, then we can have a theme that reads, “Fulfilling requirements efficiently and effectively.” Delivery, then we can have a theme that reads, “Ensuring prompt and expeditious delivery at all times.” Cost, then we can have a theme that reads, “Reducing failure cost by doing things right the first time.”
- #7: Note: What is the Objective of this Step? To gather as much relevant information as possible relating to the issue so as to have a general understanding of the problem situation. How Do We Gather Information? We do this by collecting and collating as much data and information as possible and record them in the checksheet. Data and information gathering should be done over a period of time depending on the nature of the problem. It is imperative that sufficient data and information are gathered to ensure that nothing is amiss. What Do We Do with the Information? To make sense of the data and information gathered, we have to transfer them to a Pareto chart to make them more meaningful for us to study the problem situation. The Pareto chart will show the ‘vital few and the trivial many’ nature of the problems in the form of vertical bars. Usually the first two bars contribute to the greatest percentage of the whole problem. Note: According to Pareto 80/20 rule, 20 percent of the defects contribute to 80 percent of the whole problem.
- #8: Note: What is the objective of this step? We need to indicate clearly the statement of result to be achieved. Such a statement contains three major elements. An accomplishment verb A single measurable key result A date or time period within which the result is to be accomplished. For example: “To reduce the submission time of monthly report by 100% from 4 days to 2 days by 31st July 2003” Use the SMART concept when setting target: S = Specific [Make sure that the target is specific] M = Measurable [Target must be quantified] A = Action-oriented [A plan of action must be prepared] R = Realistic [This must be balanced with available resources] T = Time-bound [Ensure that the target has a set deadline]
- #14: Note: Step 1: The time period for the collection of data must be determined. It could be as brief as a few hours, or it may require days or even a few months. This may partly be determined by how you would like to organise the data. The time period will also enable you to compare changes in a problem, like comparing one time period against another. Often problems are a result of start-up or shut-down procedures, vacations schedules, job changes, or just seasonal differences. In order to solve the problem, this must be taken into consideration. Step 2: Determine how you want to organise the data. The data on a specific problem could be ordered under different headings. Do you want to order it by defect, by department, by employee, or by location. This is an important step and members should be encouraged to participate. Remember, when organising the data, the idea is not to pick the easiest method but the one that is likely to portray the problem in a a manner the we can solve it. Step3: Design the checksheet that will enable you to gather the most imformation using the latest effort. Step 4: Collect and record the data on the checksheet. In this step you must determine where you go to collect data. It depends of course, on what day you decide to collect and under what heading you are placing it. But be rest assured that there are many sources of information.
- #15: Note: ________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________
- #16: Note: Step 1: State the problem. It is important that the problem should be define as precisely as possible. Step 2: The major groupings for possible causes to be identified are determined. Any number of such groupings is permissible, although the 4M’s are the most commonly used: Man, Method, Machine, Material. Step3: All members get involved by participating in a brainstorming session. Brainstorming works best when specified rules are observed. The team leader reviews this before each session. Step 4: All the ideas that are collected during the brainstorming session are critically examined to identify those that best described the causes to the problem. Doing this can be a time consuming process involving a critical analysis of the pros and cons of each. The process can however be greatly hastened by giving members the opportunity to vote for each cause they believe to be important. Step 5: The most likely causes are now ranked in order of importance. To do this, the members look only at those causes that have been voted and circled. Step 6: The most likely causes is then tested in an attempt to verify it. A good way to conduct verification is to brainstorm for supporting evidences, obtain feedback from process owners, or get documents that can support the authenticity of the likely causes in question.
- #17: Note: ________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________
- #18: Note: Step 1: Decide how the problem information should be classified. By employee? By defect? By workstation? By department? Step 2: Select a time period for your observation. The time period during which data is collected may be only a few hours, or a few days, or even a few months. Develop a checksheet to cover the specified time period. Step3: Organise the data into table by ranking them according to the frequency and calculate percentages of each item relative to all other items. Step 4: Draw a horizontal and vertical axes on a graph paper. Divide the horizontal axis into equal segments, one for each item. Scale the vertical axis in such a way that the top represents the total. Construct a bar graph with the most frequency first and the least frequency last. Step 5: Next is to plot the cumulative line that sums up each column with the previous one. On the right hand side of the vertical axis, scale it from 0 to 100% which must correspond to the numbers on the left hand axis, the numbers and percentage of which can be obtained from the checksheet. The process for plotting each item must be done in relation to the bar graph drawn earlier. The cumulative line is complete when it reaches the 100% level as per the percentage scale on the right side. A legend may be added so that anyone reading the chart can understand.
- #19: Note: ________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________
- #21: Note: Step 1: Calculate the range. Range=Largest Observed Value minus Smallest Observed Value. Step 2: Determine Class Interval Step3: Prepare the frequency table. Step 4: Calculate mid value I.e: Upper limit plus lower limit divided by 2. Step 5: Draw horizontal and vertical axis on a piece of graph paper. On the vertical axis indicate all class intervals. Step 6: Transfer the data from the frequency table to the graph. Class Interval Mid Value Frequency Tally Frequency
- #22: Trucks turn around time Tally Number of Trucks (frequency) 10 6 15 9 20 33 25 46 30 65 35 40 40 26 45 8 50 5
- #23: Note: Step 1: Take samples from lots. eg. Day 1 - 25 we have 25 different lots for purposes of analysing quality. Let the sample size be 10. Step 2: Each day, we compute the mean i.e. add up and divide by 10 Step3: Each day, we calculate the range i.e Upper limit minus lower limit. Step 4: We are now in a position to draw or construct a variable control chart.
- #24: STUDENTS APPLICATION TO ACCEPTANCE MOVING RANGE 1 29 2 27 2 3 30 3 4 29 1 5 33 4 6 35 2 7 35 0 8 36 1 9 28 8 10 32 4 11 25 7 12 29 4 13 46 17 14 32 14 15 28 4
- #25: Note: Step 1: Collect paired samples of data whose relationship you wish to investigate Step 2: Draw a horizontal and vertical axes of the graph with appropriate scale. If the relationship of the two variables are that of a cause and effect values on the vertical axis. Step3: Plot the data on the graph. If data values are repeated and fall on the same point, make a circle.
- #26: Note: ________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________
- #27: Note: ________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________