Simulation of error detection in computer programs

The relationship between computer program complexity and error detection capability is investigated by representing a program as a directed graph and simulating the detection and correction of errors. Variables of interest are test coverage, number of inputs, residual errors, execution time, correction time and node-arc-loop relationships. One application is in software design where the information provided by the model would be used to select program structures which are easy to test. A second application is in software testing where test strategies and allocation of test effort would be based on error detection and complexity considerations.