A taxonomical review of software verification techniques: an illustration using discrete-event simulation
MetadataShow full item record
The use of simulation and modeling as a technique for solving today's complex problems is ever-increasing. Correspondingly, the demands placed on the software which serves as a computer-executable representation of the simulation model are increasing. With the increased complexity of simulation models comes a greater need for model verification, particularly programmed model verification. Unfortunately, current model verification technology is lacking in techniques which satisfy the verification needs. Specifically, there are few guidelines for performing programmed model verification. There is, however, an abundance of software verification techniques which are applicable for simulation model verification. An extensive review of techniques applicable for simulation programmed model verification is presented using the simulation and modeling terminology. A taxonomy for programmed model verification methods is developed. The usefulness of the taxonomy is twofold: (1) the taxonomy provides an approach for applying software verification techniques to the problem of programmed model verification, and (2) the breadth of the taxonomy provides a broad spectrum of perspectives from which to assess the credibility of simulation results. A simulation case study demonstrates the usefulness of the taxonomy and some of the verification techniques.
By referencing this work, one can determine what, and when, techniques can be used throughout the development life cycle. He will know how to perform each technique, how difficult each will be to apply, and how effective the technique will be. The simulation modeler - as well as the software engineer â will find the taxonomy and techniques valuable tools for guiding verification efforts.
- Masters Theses