Perform Quality Control
Where quality assurance looks at whether standards and procedures are being followed, quality control looks at specific measurements to see if the project and its processes are in control. It is during quality control that the height of tables in a manufacturing process will be measured, where the number of bugs per module will be measured.
Quality control helps answer the questions; “Is everything all right on the project?” “Do I have to spend any additional time or change my project management activities?” “Will the project succeed?” Quality control also involves taking action to eliminate the root causes of unsatisfactory project performance.
Quality control is done during the monitoring and controlling process group of the project and its focus is on the correctness of work. A major feature of quality control is inspection; checking the quality of work to see if it conforms to standards. Inspection can be done while the work is being done or after it is completed.
One of the key traits of the best project managers is making sure the plans are working. This shows up in the quality area as the need to evaluate the effectiveness of the quality control system.
You might ask yourself whether you are measuring the right things. Are your measurements providing real, valuable information? Are you comfortable that your quality control efforts are telling you how the project is really doing? Stop here and read this again. A great project manager makes time to evaluate the effectiveness of his plans and processes. Do you?
What is Perform Quality Control?
Quality control results in recommend changes, corrective and preventive actions and defect repair to integrated change control. Once approved, such recommendations could lead to a changed quality baseline. It is also during quality control that defects are actually repaired after approval in integrated change control.
Let’s think about the manufacturing industry again. Remember, many questions on the exam describe situations in one industry or another. Do not let that confuse you. A future PMP should easily understand the situation described in each question on the exam, no matter what ‘industry’ is described.
Population/Sample Let’s say you make tables. Would each table be the same exact height? No, there would be some allowable variation. Even so, the tables must be checked to see if they meet quality standards on the project.
What if inspecting each table would cause damage or take too much time? Then taking a statistically valid sample would be best. It is best to take a sample of a population if we believe there are not many defects, or if studying the entire population would:
- Take too long
- Cost too much
- Be too destructive
There can sometimes be statistical references on the exam. One that often confuses people is mutual exclusivity. Two events are said to be mutually exclusive if they cannot both occur in a single trial. For example, flipping a coin once cannot result in both a head and a tail.
The likelihood that something will occur, usually expressed as a decimal or a fraction, on a scale of zero to one.
A normal distribution is the most common probability density distribution chart. It is in the shape of a bell curve and is used to measure variations.
Another confusing statistical term often showing up on the exam is statistical independence—the probability of one event occurring does not affect the probability of another event occurring (e.g., the probability of rolling a six on a die is statistically independent from the probability of rolling a five on the next roll).
Standard Deviation (or sigma)
A measure of a range is its standard deviation. Also sometimes stated as a measure of how far you are from the mean (not the median). (Remember (P – O)/6 is the three-point estimate formula for standard deviation using optimistic, pessimistic and most likely estimates described in the Time lesson.)
3 or 6 Sigma Sigma is another name for standard deviation. 3 or 6 sigma represents the level of quality that a company has decided to try to achieve. At 6 sigma, less than 1.5 out of 1 million doors produced will have a problem. At 3 sigma, approximately 2,700 will have a problem. Therefore, 6 sigma represents a higher quality standard than 3 sigma. 3 or 6 sigma are also used to calculate the upper and lower control limits in a control chart, described later.
You should MEMORIZE:
- Sigma is taken on both sides of the mean. Half the curve is to the right of the mean, and half the curve is to the left of the mean
- +/- 1 sigma is equal to 68.26 percent which is the percentage of occurrences to fall between the two control limits
- +/- 2 sigma equals 95.46 percent
- +/- 3 sigma equals 99.73 percent
- +/- 6 sigma equals 99.99985 percent