The scientific method of choosing model fidelity

Simulation modeling currently enjoys great popularity as a tool for solving problems within Department of Defense activities. In this work we consider the process of upgrading an existing simulation model by increasing the fidelity of the model. The submodels to be upgraded and the degree to which they are upgraded should be chosen in a coherent, scientific manner. This is currently not the norm. In this work we describe a method which a simulation analyst can use to choose from a set of proposed model upgrades that accounts for both the costs of the upgrade as well as the benefits.

[1]  Jack P. C. Kleijnen,et al.  Statistical Techniques in Simulation , 1977, IEEE Transactions on Systems, Man, and Cybernetics.

[2]  Charles A. Pratt.  The society for computer simulation , 1984, SIML.

[3]  Jeff Rothenberg,et al.  Knowledge-Based Simulation: An Interim Report , 1989 .

[4]  Osman Balci,et al.  Guidelines for successful simluation studies (tutorial session) , 1990, WSC' 90.

[5]  Osman Balci.  Guidelines for successful simulation studies , 1990, 1990 Winter Simulation Conference Proceedings.

[6]  O. J. Dahl,et al.  Simula Begin , 1979 .

[7]  Lee W. Schruben,et al.  An experimental procedure for simulation response surface model identification , 1987, CACM.

[8]  Jack P. C. Kleijnen,et al.  Statistical Techniques in Simulation , 1977, IEEE Transactions on Systems, Man and Cybernetics.

[9]  Paul Glasserman,et al.  Gradient Estimation Via Perturbation Analysis , 1990 .