.
P-CHARTS.
A P-chart is an attributes control chart tool used with data collected in subgroups of varying sizes. Since subgroup size can vary, the data is charted to show a proportion of nonconforming items rather than the actual count. The process attribute is always described in a yes/no, pass/fail or a go/no go form. For example, if you are tracking the number of incorrect applications processed, they would be identified as a "no" or a "fail". The data is then charted as a proportion rather than the actual value in order to provide standardized results among sub-groups. P-Charts identify values in the process that deviate from the mean. P-charts show how a process changes over time.
P-Charts have several applications: assessing the stability of a process; analyzing the results of a process improvement; comparing how process is running now to the past; providing standardization through a continuous collection and analysis of data; and enabling users to pinpoint sources of variation in process.
P-Charts are used when all of the following conditions exist:.
• An assessment of the system stability is required.
• The data provides a count of nonconforming items per subgroup.
• The counts can be converted to proportions.
• There only two outcomes to any given check.
• The time order of subgroups is preserved.
Once it is determined that a P-Chart is to be used, several steps must be taken. The characteristics of the chart need to be determined. The subgroup size will then need to be determined. Size needs to be large enough to have sufficient defects present in the subgroups most of the time. The basis of calculating control limits and the centerline to be used should be established. The sampling period then needs to be determined. The data should be collected. The rate of non-conformities in each subgroup (i.e. # of errors/ subgroup size) needs to be calculated.