软件质量保证-12-Quality Analysis-Defect Analysis.pdfVIP

Defect Classification and Analysis p  General Types of Defect Analyses. p  ODC: Orthogonal Defect Classification. p  Analysis of ODC Data. Defect Analysis p  Goal: n  (actual/potential) defect ↓ n  quality ↑ in current and future products. Defect in Quality Data/Models p  Defect data is quality measurement data: n  Extracted from defect tracking tools. n  Additional (defect classification) data may be available. n  post-release data issues p  Defect data in quality models: n  As results in generalized models. n  As response/independent variables in product specific models. General Defect Analysis p  General defect analyses: Questions n  What? identification (and classification). p  type, severity, etc., n  Where? distribution across location. n  When? discovery/observation p  what about when injection occurs? harder p  pre-release: more data p  post-release: less data, but more meaningful/sensitive n  How/why? related to injection = use in future defect prevention. General Defect Analysis p  General defect analyses: Types n  Distribution by type or area. n  Trend over time. n  Causal analysis. n  Other analysis for classified data. Defect Analysis: Data Treatment p  Variations of defect data: n  Error/fault/failure perspective. n  Pre-/post-release. n  Unique defect? n  Focus here: defect fixes – usually contain more data p  Why defect fixes (DF)? n  Propagation information (system structure, component interconnection, product evolution) n  Close ties to effort (defect fixing). n  Pre-release: more meaningful (post release: each failure occurrence.) Defect Distribution Analysis p  What: Distribution over defect types. n  Types/sub-types. n  Defect types related to products “domain. n