Risk Management In Software Testing

Risk management evaluates how significant a defect is (or could be). These evaluations permit prioritizing software testing and repairing cycles. The true is that there is usually not enough time or resources for complete testing, so it is requisite to make prioritization. 

Risk Driven Testing

The failure of each component is rated by Likelihood and Impact.

Likelihood is an evaluation of how possible it is that the component would fail.

Impact is a severity rating, based on what would be if the component defective.

Together, Impact and Likelihood determine the Risk for the piece.

The significance of Risk for the piece will differ from project to project and organization to organization.

A system level dimension for risk management is the Mean Time To Failure (MTTF).

Test data sampled from realistic beta testing is used find the average time until system failure.  This data is extrapolated to forecast overall uptime and the expected time the system will be operational.

It happens so measured with MTTF is Mean Time To Repair (MTTR). This represents the expected time until the system will be repaired and back in use after a failure is observed.

Availability, received by calculating MTTF / (MTTF + MTTR), is the probability that a system is available when needed.