Comparing the Finite-Time Performance of Simulation-Optimization Algorithms

TL;DR

This paper empirically compares various simulation-optimization algorithms to understand their finite-time performance and how problem properties influence their effectiveness, aiming to guide future algorithm development.

Contribution

It provides a systematic empirical evaluation of algorithms' finite-time performance and analyzes the impact of problem characteristics on their effectiveness.

Findings

Performance varies with problem properties

Certain algorithms outperform others in specific scenarios

Insights guide future algorithm improvements

Abstract

We empirically evaluate the finite-time performance of several simulation-optimization algorithms on a testbed of problems with the goal of motivating further development of algorithms with strong finite-time performance. We investigate if the observed performance of the algorithms can be explained by properties of the problems, e.g., the number of decision variables, the topology of the objective function, or the magnitude of the simulation error.