Title :
A system for controlling software inspections
Author :
Gupta, V. ; Patnaik, A.R. ; Goel, Nishith
Author_Institution :
Cistel Technol. Inc., Ottawa, Ont., Canada
Abstract :
Software inspections are a powerful tool for detecting faults in software during the early phases of the life cycle. Deciding when to stop inspections is an important determinant of inspection effectiveness. Capture-recapture (CR) models can be used to estimate defect content, and hence help to make a reinspection decision. We present Monte Carlo simulations of six CR models. The objective is to find the best CR model. This builds on previous work by simulating the context of high-reliability systems. The results indicate that model MtCh, which underestimates median relative error, gives zero failures and has the best decision accuracy.
Keywords :
Monte Carlo methods; program testing; program verification; Monte Carlo simulation; MtCh model; capture-recapture model; defect content estimation; median relative error; software fault detection; software inspection controlling; software reinspection decision; Animals; Chromium; Context modeling; Control systems; Inspection; Object detection; Power system modeling; Sampling methods; Software systems; Wildlife;
Conference_Titel :
Electrical and Computer Engineering, 2003. IEEE CCECE 2003. Canadian Conference on
Print_ISBN :
0-7803-7781-8
DOI :
10.1109/CCECE.2003.1226148
