Proceedings of the 2002 Winter Simulation Conference
نویسندگان
چکیده
Response Surface Methodology (RSM) searches for the input combination that optimizes the simulation output. RSM treats the simulation model as a black box. Moreover, this paper assumes that simulation requires much computer time. In the first stages of its search, RSM locally fits first-order polynomials. Next, classic RSM uses steepest descent (SD); unfortunately, SD is scale dependent. Therefore, Part 1 of this paper derives scale independent ‘adapted’ SD (ASD) accounting for covariances between components of the local gradient. Monte Carlo experiments show that ASD indeed gives a better search direction than SD. Part 2 considers multiple outputs, optimizing a stochastic objective function under stochastic and deterministic constraints. This part uses interior point methods and binary search, to derive a scale independent search direction and several step sizes in that direction. Monte Carlo examples demonstrate that a neighborhood of the true optimum can indeed be reached, in a few simulation runs.
منابع مشابه
Proceedings of the 2002 Winter Simulation Conference
We discuss methods for statistically analyzing the output from stochastic simulations. Both terminating and steadystate simulations are considered.
متن کاملProceedings of the First MEFOMP International Conference of Medical Physics: November 2-4, 2011, Shiraz, Iran
متن کامل
Proceedings of the 6th International Conference on Science and Social Research (CSSR)(Malaysia)
متن کامل
Proceedings of the 2002 Winter Simulation Conference
Simulation optimization has received considerable attention from both simulation researchers and practitioners. In this tutorial we present a broad introduction to simulation optimization and the many techniques that have been suggested to solve simulation optimization problems. Both continuous and discrete problems are discussed, but an emphasis is placed on discrete problems and practical met...
متن کاملProceedings of the 2002 Winter Simulation Conference
In this paper, we discuss issues concerning the simulation of transportation systems. In particular, we demonstrate a number of implementation tricks that are designed to make the modeling and coding processes more efficient and transparent. We present examples involving the simulation of commercial airline and military sealift operations.
متن کاملProceedings of the 2002 Winter Simulation Conference
This paper demonstrates how to use Simkit to create Discrete Event Simulation models using a component framework. The component framework is based on a listener design pattern especially useful for simulation models. The objects created are called Listener Event Graph Objects, so the component framework is called LEGO. Simkit is an Open Source package written in Java.
متن کامل