1、1. Cause Analysis ToolsUse these cause analysis tools when you want to discover the cause of a problem or situation:1） Fishbone DiagramAlso Called: Cause-and-Effect Diagram, Ishikawa DiagramVariations: cause enumeration diagram, process fishbone, time-delay fishbone, CEDAC (cause-and-effect diagram
2、with the addition of cards), desired-result fishbone, reverse fishbone diagramDescriptionThe fishbone diagram identifies many possible causes for an effect or problem. It can be used to structure a brainstorming session. It immediately sorts ideas into useful categories.When to Use When identifying
3、possible causes for a problem. Especially when a team 抯 thinking tends to fall into ruts. Procedure Materials needed: flipchart or whiteboard, marking pens.1. Agree on a problem statement (effect). Write it at the center right of the flipchart or whiteboard. Draw a box around it and draw a horizonta
4、l arrow running to it. 2. Brainstorm the major categories of causes of the problem. If this is difficult use generic headings: o Methods o Machines (equipment) o People (manpower) o Materials o Measurement o Environment 3. Write the categories of causes as branches from the main arrow. 4. Brainstorm
5、 all the possible causes of the problem. Ask: 揥 hy does this happen?As each idea is given, the facilitator writes it as a branch from the appropriate category. Causes can be written in several places if they relate to several categories. 5. Again ask 搘 hy does this happen?about each cause. Write sub
6、-causes branching off the causes. Continue to ask 揥 hy?and generate deeper levels of causes. Layers of branches indicate causal relationships. 6. When the group runs out of ideas, focus attention to places on the chart where ideas are few. ExampleThis fishbone diagram was drawn by a manufacturing te
7、am to try to understand the source of periodic iron contamination. The team used the six generic headings to prompt ideas. Layers of branches show thorough thinking about the causes of the problem. For example, under the heading 揗 achines,?the idea 搈 aterials of construction?shows four kinds of equi
8、pment and then several specific machine numbers. Note that some ideas appear in two different places. 揅 alibration?shows up under 揗 ethods?as a factor in the analytical procedure, and also under 揗 easurement?as a cause of lab error. 揑ron tools?can be considered a 揗 ethods?problem when taking samples
9、 or a 揗 anpower?problem with maintenance personnel.2） Pareto ChartAlso called: Pareto diagram, Pareto analysisVariations: weighted Pareto chart, comparative Pareto chartsDescriptionA Pareto chart is a bar graph. The lengths of the bars represent frequency or cost (time or money), and are arranged wi
10、th longest bars on the left and the shortest to the right. In this way the chart visually depicts which situations are more significant.When to Use When analyzing data about the frequency of problems or causes in a process. When there are many problems or causes and you want to focus on the most sig
11、nificant. When analyzing broad causes by looking at their specific components. When communicating with others about your data. Procedure1. Decide what categories you will use to group items. 2. Decide what measurement is appropriate. Common measurements are frequency, quantity, cost and time. 3. Dec
12、ide what period of time the chart will cover: One work cycle? One full day? A week? 4. Collect the data, recording the category each time. (Or assemble data that already exist.) 5. Subtotal the measurements for each category. 6. Determine the appropriate scale for the measurements you have collected
13、. The maximum value will be the largest subtotal from step 5. (If you will do optional steps 8 and 9 below, the maximum value will be the sum of all subtotals from step 5.) Mark the scale on the left side of the chart. 7. Construct and label bars for each category. Place the tallest at the far left,
14、 then the next tallest to its right and so on. If there are many categories with small measurements, they can be grouped as 搊 ther.?/li Steps 8 and 9 are optional but are useful for analysis and communication.8. Calculate the percentage for each category: the subtotal for that category divided by th
15、e total for all categories. Draw a right vertical axis and label it with percentages. Be sure the two scales match: For example, the left measurement that corresponds to one-half should be exactly opposite 50% on the right scale. 9. Calculate and draw cumulative sums: Add the subtotals for the first
16、 and second categories, and place a dot above the second bar indicating that sum. To that sum add the subtotal for the third category, and place a dot above the third bar for that new sum. Continue the process for all the bars. Connect the dots, starting at the top of the first bar. The last dot sho
17、uld reach 100 percent on the right scale. ExamplesFigure 1 shows how many customer complaints were received in each of five categories. Figure 2 takes the largest category, 揹 ocuments,?from Figure 1, breaks it down into six categories of document-related complaints, and shows cumulative values.If al
18、l complaints cause equal distress to the customer, working on eliminating document-related complaints would have the most impact, and of those, working on quality certificates should be most fruitful.Figure 1Figure 23） Scatter DiagramAlso called: scatter plot, X 朰 graphDescriptionThe scatter diagram
19、 graphs pairs of numerical data, with one variable on each axis, to look for a relationship between them. If the variables are correlated, the points will fall along a line or curve. The better the correlation, the tighter the points will hug the line.When to Use When you have paired numerical data.
20、 When your dependent variable may have multiple values for each value of your independent variable. When trying to determine whether the two variables are related, such as when trying to identify potential root causes of problems. After brainstorming causes and effects using a fishbone diagram, to d
21、etermine objectively whether a particular cause and effect are related. When determining whether two effects that appear to be related both occur with the same cause. When testing for autocorrelation before constructing a control chart. Procedure1. Collect pairs of data where a relationship is suspe
22、cted. 2. Draw a graph with the independent variable on the horizontal axis and the dependent variable on the vertical axis. For each pair of data, put a dot or a symbol where the x-axis value intersects the y-axis value. (If two dots fall together, put them side by side, touching, so that you can se
23、e both.) 3. Look at the pattern of points to see if a relationship is obvious. If the data clearly form a line or a curve, you may stop. The variables are correlated. You may wish to use regression or correlation analysis now. Otherwise, complete steps 4 through 7. 4. Divide points on the graph into
24、 four quadrants. If there are X points on the graph, 5. Count X/2 points from top to bottom and draw a horizontal line. 6. Count X/2 points from left to right and draw a vertical line. 7. If number of points is odd, draw the line through the middle point. 8. Count the points in each quadrant. Do not
25、 count points on a line. 9. Add the diagonally opposite quadrants. Find the smaller sum and the total of points in all quadrants. 10. A = points in upper left + points in lower right 11. B = points in upper right + points in lower left 12. Q = the smaller of A and B 13. N = A + B 14. Look up the lim
26、it for N on the trend test table. 15. If Q is less than the limit, the two variables are related. 16. If Q is greater than or equal to the limit, the pattern could have occurred from random chance. ExampleThe ZZ-400 manufacturing team suspects a relationship between product purity (percent purity) a
27、nd the amount of iron (measured in parts per million or ppm). Purity and iron are plotted against each other as a scatter diagram, as shown in the figure below.There are 24 data points. Median lines are drawn so that 12 points fall on each side for both percent purity and ppm iron.To test for a rela
28、tionship, they calculate:A = points in upper left + points in lower right = 9 + 9 = 18B = points in upper right + points in lower left = 3 + 3 = 6Q = the smaller of A and B = the smaller of 18 and 6 = 6N = A + B = 18 + 6 = 24Then they look up the limit for N on the trend test table. For N = 24, the
29、limit is 6.Q is equal to the limit. Therefore, the pattern could have occurred from random chance, and no relationship is demonstrated.Scatter Diagram ExampleConsiderationsHere are some examples of situations in which might you use a scatter diagram: Variable A is the temperature of a reaction after
30、 15 minutes. Variable B measures the color of the product. You suspect higher temperature makes the product darker. Plot temperature and color on a scatter diagram. Variable A is the number of employees trained on new software, and variable B is the number of calls to the computer help line. You sus
31、pect that more training reduces the number of calls. Plot number of people trained versus number of calls. To test for autocorrelation of a measurement being monitored on a control chart, plot this pair of variables: Variable A is the measurement at a given time. Variable B is the same measurement,
32、but at the previous time. If the scatter diagram shows correlation, do another diagram where variable B is the measurement two times previously. Keep increasing the separation between the two times until the scatter diagram shows no correlation. Even if the scatter diagram shows a relationship, do n
33、ot assume that one variable caused the other. Both may be influenced by a third variable. When the data are plotted, the more the diagram resembles a straight line, the stronger the relationship. If a line is not clear, statistics (N and Q) determine whether there is reasonable certainty that a rela
34、tionship exists. If the statistics say that no relationship exists, the pattern could have occurred by random chance. If the scatter diagram shows no relationship between the variables, consider whether the data might be stratified. If the diagram shows no relationship, consider whether the independ
35、ent (x-axis) variable has been varied widely. Sometimes a relationship is not apparent because the data don 抰 cover a wide enough range. Think creatively about how to use scatter diagrams to discover a root cause. Drawing a scatter diagram is the first step in looking for a relationship between vari
36、ables. 2. Evaluation and Decision-Making ToolsUse evaluation and decision-making tools when you want to narrow a group of choices to the best one, or when you want to evaluate how well youve done something. This includes evaluating project results.1) Decision MatrixAlso called: Pugh matrix, decision
37、 grid, selection matrix or grid, problem matrix, problem selection matrix, opportunity analysis, solution matrix, criteria rating form, criteria-based matrix.Description A decision matrix evaluates and prioritizes a list of options. The team first establishes a list of weighted criteria and then eva
38、luates each option against those criteria. This is a variation of the L-shaped matrix.When to Use When a list of options must be narrowed to one choice. When the decision must be made on the basis of several criteria. After the list of options has been reduced to a manageable number by list reductio
39、n. Typical situations are: When one improvement opportunity or problem must be selected to work on. When only one solution or problem-solving approach can be implemented. When only one new product can be developed. Procedure 1. Brainstorm the evaluation criteria appropriate to the situation. If poss
40、ible, involve customers in this process. 2. Discuss and refine the list of criteria. Identify any criteria that must be included and any that must not be included. Reduce the list of criteria to those that the team believes are most important. Tools such as list reduction and multivoting may be usef
41、ul here. 3. Assign a relative weight to each criterion, based on how important that criterion is to the situation. Do this by distributing 10 points among the criteria. The assignment can be done by discussion and consensus. Or each member can assign weights, then the numbers for each criterion are
42、added for a composite team weighting. 4. Draw an L-shaped matrix. Write the criteria and their weights as labels along one edge and the list of options along the other edge. Usually, whichever group has fewer items occupies the vertical edge. 5. Evaluate each choice against the criteria. There are t
43、hree ways to do this: Method 1: Establish a rating scale for each criterion. Some options are:1, 2, 3: 1 = slight extent, 2 = some extent, 3 = great extent1, 2, 3: 1 = low, 2 = medium, 3 = high1, 2, 3, 4, 5: 1 = little to 5 = great1, 4, 9: 1 = low, 4 = moderate, 9 = highMake sure that your rating sc
44、ales are consistent. Word your criteria and set the scales so that the high end of the scale (5 or 3) is always the rating that would tend to make you select that option: most impact on customers, greatest importance, least difficulty, greatest likelihood of success.Method 2: For each criterion, ran
45、k-order all options according to how well each meets the criterion. Number them with 1 being the option that is least desirable according to that criterion.Method 3, Pugh matrix: Establish a baseline, which may be one of the alternatives or the current product or service. For each criterion, rate ea
46、ch other alternative in comparison to the baseline, using scores of worse (1), same (0), or better (+1). Finer rating scales can be used, such as 2, 1, 0, 1, 2 for a five-point scale or 3, 2, 1, 0, 1, 2, 3 for a seven-point scale. Again, be sure that positive numbers reflect desirable ratings.1. Mul
47、tiply each options rating by the weight. Add the points for each option. The option with the highest score will not necessarily be the one to choose, but the relative scores can generate meaningful discussion and lead the team toward consensus ExampleFigure 1 shows a decision matrix used by the cust
48、omer service team at the Parisian Experience restaurant to decide which aspect of the overall problem of “long wait time” to tackle first. The problems they identified are customers waiting for the host, the waiter, the food, and the check.The criteria they identified are “Customer pain” (how much d
49、oes this negatively affect the customer?), “Ease to solve,” “Effect on other systems,” and “Speed to solve.” Originally, the criteria “Ease to solve” was written as “Difficulty to solve,” but that wording reversed the rating scale. With the current wording, a high rating on each criterion defines a state that would encourage selecting the problem: high customer pain, very easy to solve, high effect on other systems, and quick solution.Figure 1 Decision matrix example “Customer pain” has been weighted with 5 points, showing that the
