Monitor the performance of selected subprocesses using statistical and other quantitative techniques.


The intent of this specific practice is to use statistical and other quantitative techniques to analyze variation in subprocess performance and to determine actions necessary to achieve each subprocess’s quality and process performance objectives.

Example Work Products

  1. Natural bounds of process performance for each selected subprocess attribute
  2. The actions needed to address deficiencies in the process stability or capability of each selected subprocess


1. Collect data, as defined by the selected measures, on the subprocesses as they execute.

2. Monitor the variation and stability of the selected subprocesses and address deficiencies.

This analysis involves evaluating measurements in relation to the natural bounds calculated for each selected measure and identifying outliers or other signals of potential non-random behavior, determining their causes and preventing or mitigating the effects of their recurrence (i.e., addressing special causes of variation). During such analysis, be sensitive to the sufficiency of the data and to shifts in process performance that can affect the ability to achieve or maintain process stability.

Analytic techniques for identifying outliers or signals include statistical process control charts, prediction intervals, and analysis of variance. Some of these techniques involve graphical displays. Other deficiencies in process performance to consider include when variation is too large to have confidence that the subprocess is stable, or too great to assess its capability (next subpractice) of achieving the objectives established for each selected attribute.

3. Monitor the capability and performance of the selected subprocesses and address deficiencies.

The intent of this subpractice is to identify what actions to take to help the subprocess achieve its quality and process performance objectives. Be sure that the subprocess performance is stable relative to the selected measures (previous subpractice) before comparing its capability to its quality and process performance objectives.


Examples of actions that can be taken when the performance of a selected subprocess fails to satisfy its objectives include the following:
  • Improving the implementation of the existing subprocess to reduce its variation or improve its performance (i.e., addressing common causes of variation)
  • Identifying and implementing an alternative subprocess through identifying and adopting new process elements, subprocesses, and technologies that may help better align with objectives
  • Identifying risks and risk mitigation strategies for each deficiency in subprocess capability
  • Renegotiating or re-deriving objectives for each selected attribute of a subprocess so that they can be met by the subprocess

Some actions can involve the use of root cause analysis, which is further described in SP 2.3.

Refer to the Project Monitoring and Control (PMC) (CMMI-DEV) process area for more information about managing corrective action to closure.