Validating metrics for ensuring Space Shuttle flight software quality
- 1 August 1994
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in Computer
- Vol. 27 (8) , 50-57
- https://doi.org/10.1109/2.303613
Abstract
In this article, we cover the validation of software quality metrics for the Space Shuttle. Experiments with Space Shuttle flight software show that the Boolean OR discriminator function can successfully validate metrics for controlling and predicting quality. Further, we found that statement count and node count are the metrics most closely associated with the discrepancy reports count, and that with data smoothing their critical values can be used as predictors of software quality. We are continuing our research and comparing validated metrics with actual results in an attempt to determine whether the use of additional metrics provides sufficiently greater discriminative power to justify the increased inspection costs.Keywords
This publication has 5 references indexed in Scilit:
- Minimizing risk in applying metrics on multiple projectsPublished by Institute of Electrical and Electronics Engineers (IEEE) ,2003
- Software reliability model with optimal selection of failure dataIEEE Transactions on Software Engineering, 1993
- Methodology for validating software metricsIEEE Transactions on Software Engineering, 1992
- Applied Multivariate Data AnalysisPublished by Springer Nature ,1992
- Predictive modeling techniques of software quality from software measuresIEEE Transactions on Software Engineering, 1992