Investigating the effectiveness of Variance Reduction Techniques in Manufacturing, Call Center and Cross-docking Discrete Event Simulation Models

TL;DR

This study evaluates the robustness of various variance reduction techniques across manufacturing, call center, and cross-docking simulation models, highlighting Control Variates as universally effective and revealing new insights about Antithetic Variates in cross-docking.

Contribution

It provides a comparative analysis of variance reduction techniques across multiple domains, offering practical guidelines for their application in discrete event simulation.

Findings

Control Variates reduces variance in all tested domains

Antithetic Variates performs well in cross-docking scenarios

Guidelines for applying variance reduction techniques in diverse simulation models

Abstract

Variance reduction techniques have been shown by others in the past to be a useful tool to reduce variance in Simulation studies. However, their application and success in the past has been mainly domain specific, with relatively little guidelines as to their general applicability, in particular for novices in this area. To facilitate their use, this study aims to investigate the robustness of individual techniques across a set of scenarios from different domains. Experimental results show that Control Variates is the only technique which achieves a reduction in variance across all domains. Furthermore, applied individually, Antithetic Variates and Control Variates perform particularly well in the Cross-docking scenarios, which was previously unknown.