Analyze the underlying causes of selected incidents.


The purpose of conducting causal analysis on incidents is to determine the best course of action to address incidents in the future so that their impact will be minimized most effectively. While completely preventing incidents is usually desirable, other business objectives can limit the extent to which incident prevention is effective. In some cases, it can be more effective to respond to certain incidents after they occur via reusable solutions than it is to try to reduce or prevent their occurrence in the first place. Therefore, a possible course of action includes not addressing an underlying cause at all and continuing to deal with selected incidents after they occur by using newly established or revised workarounds and other reusable solutions.

Often, analyzing incidents involves work products that are under configuration management.

It is essential to have a single repository established that contains all known incidents, their underlying causes, and approaches to addressing these underlying causes. This repository can be used to quickly determine the causes of related incidents.

Refer to the Configuration Management (CM) (CMMI-SVC) process area for more information about tracking and controlling changes.

Example Work Products

  1. Report of underlying causes of incidents
  2. Documented causal analysis activities


1. Identify underlying causes of incidents.


Examples of approaches to identifying underlying causes of incidents include the following:
  • Analyze incidents reported by customers to a help desk
  • Monitor the service system to identify potential incidents
  • Analyze trends in the use of resources
  • Analyze strengths and weaknesses of the service system
  • Analyze mean times between service system failures and availability
  • Analyze external sources of information such as alerts, news feeds, and websites

Refer to the Risk Management (RSKM) (CMMI-DEV) process area for more information about identifying and analyzing risks and mitigating risks.

2. Record information about the underlying causes of an incident or group of incidents.

When recording information about the underlying causes of an incident, record sufficient information to properly support causal analysis and resolution.


Examples of information to record include the following:
  • Incidents affected or potentially affected by the underlying cause
  • Configuration items involved
  • Relevant characteristics of the situation in which the incidents did or could occur

3. Conduct causal analysis with the people who are responsible for performing related tasks.

For underlying causes of major incidents, the analysis can involve assembling a separate team to analyze the underlying cause.

Refer to the Causal Analysis and Resolution (CAR) (CMMI-DEV) process area for more information about determining causes of selected outcomes.

4. Determine the best overall approach for dealing with selected incidents in the future.

This approach can include service system changes that reduce or prevent the occurrence of similar incidents, that limit the impact of similar incidents through reusable solutions, or that combine some of these approaches.