Software Engineering and Simulation Credibility
Abstract
Most people think of 'validation' as the hallmark of simulation credibility. But some simulations, by their very nature (e.g., mission level models, highly complex physics-based simulations, etc.) are notoriously difficult to validate. There are also situations in which the process of validation, even if feasible, cannot keep pace with the dynamic nature of simulation evolution, or where the cost of validation data is prohibitively high. Are there any other 'hallmarks' of simulation credibility that can help us in cases like these? This paper demonstrates that significant insights into simulation credibility can be obtained via detailed examination of the software engineering processes and practices by which the simulation was developed and is maintained. The nature and value of good software engineering processes is often unknown, or under-appreciated, by the end-users of simulation results and therefore is often overlooked as a barometer of simulation credibility.
Document Details
- Document Type
- Technical Report
- Publication Date
- Aug 01, 2001
- Accession Number
- ADA405033
Entities
People
- Paul R. Muessig
- Ronald L. Ketcham
Organizations
- Naval Air Warfare Center Weapons Division