软件试验缺陷预测模型研究

Research on Software Test Defect Prediction Model

装备软件缺陷分析、预测是实现软件可靠性评估重要依据,是软件试验急需解决的难点问题。本文依据软件测试及试验经验数据对Rayleigh模型进行改进,使Rayleigh曲线更好地拟合真实缺陷数据。新模型可在软件试验阶段基于软件测试数据预测软件试验缺陷及软件整体缺陷数,进而更好地指导软件试验开展,更加科学的对软件可靠性进行评估。

Analysis and prediction of equipment software defects is an important basis for software reliability evaluation,and is a difficult problem to be solved urgently in software test.In this paper,the Rayleigh model is improved according to the software testing and experimental data,so that the Rayleigh curve can better fit the real defect data.The new model can predict the number of software test defects and overall software defects based on the software test data in the software test stage,so as to better guide the development of software test and evaluate the software reliability more scientifically.