Flowcharts and Diagrams
Flowcharts can be helpful in understanding the cause and effect relationships between the process of performing work and the results that are inspected through measurement or attribute inspection. A flowchart is simply an organized way to look at the steps that have to be carried out to perform some goal. There are many techniques and styles of flowcharting.
The cause and effect diagram, also known as the fishbone diagram because of its appearance, was developed by Kaoru Ishikawa. This is a way of diagramming the flow of work that is useful in determining the cause and effect of problems that are encountered.
As can be seen in Figure 4-4, the process is separated logically into branches. Each of the branches can be dealt with separately. If the work on one branch is excessive, a separate meeting may be used to investigate it or any other branches requiring input from more people or more time to consider the branch.
Like the work breakdown structure, the cause and effect diagram allows for an orderly consideration of each of the possible causes of a problem and then allows for the consideration of each effect and the solution that will reduce the problem.
Vilfredo Pareto is given credit for developing the concept of 80—20 rule. He was an economist who found that typically 80 percent of the wealth of a region was concentrated in 20 percent of the population. This concept describes a number of phenomena that
Figure 4-3. Seller's risk versus buyer's risk.
Probability of accepting
• Reject a lot
that is good
Percent of defects
Figure 4-4. Cause and effect diagram, or fishbone diagram.
Figure 4-4. Cause and effect diagram, or fishbone diagram.
occur in the real world. In terms of quality, it can be said that 80 percent of the cost of defects is caused by 20 percent of the problems. In other words, if there were one hundred possible things that could be considered to be defects in a process, 20 percent, or twenty of the problems, will account for 80 percent of the cost.
By identifying these twenty items it is possible to expend the energy of the organization where it will do the most good. In quality control, as well as in many areas of project management, it is important that the always limited effort available in the organization be concentrated on the problems where the most benefits will result.
The Pareto chart is a simple way of determining the places where this effort might be concentrated. The problems in a process are arranged in the order of importance and are generally arranged by ranking according to the most important factors, such as cost, time delay, or some other parameter (Table 4—1; Figure 4-5).
It can be plainly seen that problem ''a'' is the most serious problem and will have the greatest effect on the process if it is solved. If problem ''a'' is solved, we can redraw the Pareto chart to show the improvement (Table 4—2; Figure 4-6).
Control charts are used to determine whether the observed variations in a process are due to normal process variations or whether they are due to the process getting out of control. Control charts allow the observations to be interpreted in such a way as to allow corrections to the process prior to the process producing bad output.
In order to accomplish this goal, the known variations must be determined first. This is done by measuring the dimensions in question on a group of known parts. A lot
Figure 4-5. Pareto chart.
3SG 3GG ^2SG S 2GG
b c d e f g h i j is selected that is known to be acceptable. The dimension is measured on each of the parts, and the mean and standard deviation of the dimension is determined.
By determining the mean and standard deviation of the group of parts we can define the probability distribution of this dimension. In the area of quality we are normally interested in maintaining process control to plus or minus 3 standard deviations. This means that if we consider the dimension of the part in question and the process is under control, then 99.7 percent of the parts coming out of the process should fall in the dimensional range of plus or minus 3 standard deviations from the mean value dimension. If a part is measured and found to be outside this range of values, we have cause for concern. This is a concern even though the part dimensions could still be well within the engineering design tolerance of the part's dimensions.
The control chart is constructed by marking the middle line as the mean value. The upper and lower control limits are determined by adding and subtracting three times the standard deviation of the measurement of the part (Figure 4-7).
If parts are later measured and found to be greater or less than the upper and lower control limits specified, then the process is considered to be out of control and to have an assignable cause. This assignable cause should be investigated to determine what the problem is and appropriate corrective action should be taken. Normally, the upper and lower control limits are less than the engineering dimensional part tolerance for this dimension.
Frequently, a guideline to the use of the control chart is the ''rule of seven.'' If there
Figure 4-7. Control chart.
are seven or more points in succession that are either above or below the mean value there is cause for concern about the process. This is because the probability of there being seven measurements in a row that are all on the same side of the mean value is very small, and therefore, it can be concluded that the process is no longer functioning properly.
For example, suppose a part is designed to have a dimension of 1.000 inches, with a tolerance of plus or minus .005 inches. A group of parts are made after the process has been developed and stabilized and been in operation for several hundred parts. These parts are individually measured for this dimension. The mean and standard deviation for the group of parts is determined.
As seen in Figure 4-7, the mean value is found to be 1.0100 inches and the standard deviation is found to be .0001 inches. The upper control limit can be set at 1.0103 inches, and the lower control limit can be set at 1.0097.
Notice that the mean value of the process is not necessarily the nominal dimension on the engineering specification. This is because the process engineer in this situation has determined to run the process deliberately on the high side of the dimension. By taking this approach, if the process goes out of control and bad parts are made, there is a greater chance that the bad parts can be reworked by re-machining them. This is a better alternative than scrapping parts that are too small.
Run charts are simple diagrams that are used to plot an attribute of a product. These are similar to control charts except that upper and lower control limits are not used. Periodically an observation is made of a characteristic attribute. The observation is recorded. Over time, trends may be seen and a history of the observed characteristic is made. Run charts are frequently done prior to the calculation of the control limits of a control chart.
Checklists are a sample tool that is used to keep from overlooking items of importance. A checklist is really just an instruction sheet for an inspector to use. The items in the checklist should be significant items. If a checklist is seen as a superfluous document, it will not be used.
Kaizen is one of the many quality techniques that come to us from the work of the Japanese. The Japanese word for continuous improvement is kaizen. Using this method, the managers as well as the workers and everyone else are continuously on the lookout for opportunities to improve quality. Thus the quality of a process improves in small increments on a continuous basis.
In this kaizen way of doing things, even the processes that are operating without problems are continuously under scrutiny. A process is observed to be making acceptable parts, but is seen to be slow, or there is an opportunity for even greater quality than is required.
Benchmarking is the process of comparing the performance of a current process to that of another similar process to determine the differences between them. If a machine can manufacture two hundred parts an hour and a new machine is compared to the old machine, the benchmark for the existing process on the old machine is two hundred parts per hour.
Was this article helpful?
What you need to know about… Project Management Made Easy! Project management consists of more than just a large building project and can encompass small projects as well. No matter what the size of your project, you need to have some sort of project management. How you manage your project has everything to do with its outcome.