PMBOK Tools and Techniques to Plan Quality – Part Two
Control charts are used to determine whether a process is stable and predictable. The planned value of a process is the center line. For many processes, the upper and lower control limits are plus and -3 standard deviations from the plan, or the mean, depending upon the circumstances.
The upper and lower specification limits are the limits specified in the quality requirements. If the measurement is getting close to the control limits, you should take action to get it back towards the midline.
Here are the PMBOK Guide definitions:
Control chart. A graphic display of process data over time and against established control limits, and that has a center line that assists in detecting the trend of plotted values toward either control limit
Control limits. The area composed of three standard deviations on either side of the center line, or mean of a normal distribution of data plotted on a control chart that reflects the expected variation in the data
Specification limit. The area on either side of the center line, or mean, of data are plotted on a control chart that meets the customer’s requirements for a product or service. This area may be greater than or less than the area defined by the control limits
Although control charts were developed to track manufacturing and repetitive processes, they can be used to track defects, cost and schedule variance, or any other predictable event on a project.
For example, it is predictable that you will have cost and schedule variance is on your project.
However, you want to make sure they are within an allowable limits. You can track the variance is using a control chart to spot trends and to monitor the quality of cost and schedule performance.
When planning for quality, establish the items that you will measure via a control chart and set the upper and lower control limits and upper and lower specification limits.
When planning your quality parameters, you might want to look at best in class or best outcomes from other projects, and use those as a target. This is benchmarking, and as an example you could benchmark customer satisfaction ratings for your organization.
Design of experiments
This technique is used to optimize the product or process. It entails working with multiple variables to find the best possible combination. As an example suppose you are a fast food outlet. The kitchen crew might try to optimize a menu with a healthy treat for snack time by adjusting ingredient ratios, cooking time, and cooking temperature to find the best interaction of these variables.
To determine whether a group of deliverables meets the quality requirements, you don’t have to test every single one. In many circumstances, you can test a sample of them and determine the viability of the entire group based on the outcome of the sample.
Suppose that the project management office director wanted to determine how well all of the project managers are following the change control policies. If 15 project managers were managing 42 projects, the director could mandible a select five change control and from different project managers for different projects and infer the performance of the rest of the projects based on the results of those five change control logs.
When sampling, you can employ the following approaches:
Attribute sampling. This indicates whether the result conforms for example that a light is on or off
Variable sampling. This indicates the degree to which the result complies as rated on a continuous scale. For example, if desired measurement is 200°, a sample at 196°is 98 per cent of the desired measurement
During the planning process, you identify the items to be sampled, the number sampled, and the criteria for sampling
For all projects, but particularly four process improvement projects, you can flowcharts any processes you will use, including the change control processes, a product development process, or a testing process.
A flow charts can help you see the relationship between the process steps. You can use this information to optimize the process and to see where problems and defects can occur.
When identifying the quality requirements, you can use any project requirements collecting technique such as the following:
- Affinity diagrams
- Nominal group technique
In addition, you can use a number of proprietary techniques particularly if they used by your organization and these could include CMMI, Six Sigma,TQM, and so on