GAMP- Management of Test Incidents Exceptions
This article was written by Ilan Shaya a world specialist in validation, automation & control
Development life cycle of a system of product containing SW often involves unexpected incidents, predominantly in the testing phase. Whilst defects and failures are unlikely to ever be totally eliminated, occurring incidents should be promptly and correctly dealt with by the test team to minimize the likelihood of passage of avoidable defects into production environment
Incident notifications may be received via different media (e.g., paper, e-mail, or a dedicated incident reporting system), and may, therefore, require variations in the way they are handled. Thus, an Incident Management Plan (or Procedure), which defines the approach for dealing with individual input types, should be prepared and communicated across the test team before starting any testing. In small projects, this plan may form a sub-section of the overall Quality Plan/
:The aim of any test incident management process should be
To establish and operate an effective service to manage the recording and subsequent handling of test incident, and any other observations noted by the testers
To act as a single point of contact between end users and problem solvers, service suppliers, system developers and testers
To handle all incidents in a consistent manner
To demonstrate that all incidents have been properly resolved
To report performance against the service levels set out in the pertinent Service Level Agreements -SLA
Details of all incidents should be logged centrally for each project, and each incident should be allocated a unique reference number, which should be communicated to the relevant parties for future reference and tracking