Lessons identified from data collection for model validation

ABSTRACT

Computer simulation models will ideally be developed within an environment where all the necessary input data is readily available and all the relevant stakeholders are supportive and co-operative. In practice, many models are developed for the purpose of evaluating organisational or process change. As a result, system performance data may often be limited or potentially biased by key parties associated with the results of the simulation model. Ensuring valid input data is therefore a key aspect of model validation and an important component in a successful simulation study. This paper addresses some examples of data collection problems that have been encountered by the author and presents a number of lessons identified.

REFERENCES

• 1: Collins, Reference English Dictionary. HarperCollins 1988.
• 2: Alan C. Cowdale, Traffic and road planning simulation: simulation modelling in support of emergency fire-fighting in Norfolk, Proceedings of the 35th conference on Winter simulation: driving innovation, December 07-10, 2003, New Orleans, Louisiana
• 3: Emily K. Lada , Mary Ann Wagner , Natalie M. Steiger , James R. Wilson, Introduction to modeling and generating probabilistic input processes for simulation, Proceedings of the 37th conference on Winter simulation, December 04-07, 2005, Orlando, Florida
• 4: Law, A. M., and D. W. Kelton. 2000. Modelling and analysis. 3rd Edition. New York: McGraw-Hill.
• 5: Robert G. Sargent, Verification and validation: verification and validation of simulation models, Proceedings of the 35th conference on Winter simulation: driving innovation, December 07-10, 2003, New Orleans, Louisiana
• 6: Deborah A. Sadowski, Tips for successful practice of simulation, Proceedings of the 37th conference on Winter simulation, December 04-07, 2005, Orlando, Florida