Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

TESO Tabu Enhanced Simulation Optimization for Noisy Black Box Problems

Bulent Soykan, Sean Mondesire, Ghaith Rabadi

TL;DR

TESO addresses the core challenge of noisy, expensive black-box simulation optimization by marrying Tabu Search principles with Elite Memory in a direct-search framework. The algorithm uses a short-term Tabu List to diversify searches and an Elite Memory to intensify around high-potential regions, guided by an aspiration criterion and an adaptive noise schedule to cope with stochastic evaluations. Empirical results on a stochastic M/M/k queue optimization demonstrate that full TESO achieves the best convergence speed and reliability compared with ablated variants and random sampling, highlighting the complementary roles of diversification and intensification. The work advances practical SO by providing a robust, memory-enhanced methodology that does not rely on surrogate modeling, offering a scalable approach to hard noisy optimization problems with real-world applicability in operations research and systems design.

Abstract

Simulation optimization (SO) is frequently challenged by noisy evaluations, high computational costs, and complex, multimodal search landscapes. This paper introduces Tabu-Enhanced Simulation Optimization (TESO), a novel metaheuristic framework integrating adaptive search with memory-based strategies. TESO leverages a short-term Tabu List to prevent cycling and encourage diversification, and a long-term Elite Memory to guide intensification by perturbing high-performing solutions. An aspiration criterion allows overriding tabu restrictions for exceptional candidates. This combination facilitates a dynamic balance between exploration and exploitation in stochastic environments. We demonstrate TESO's effectiveness and reliability using an queue optimization problem, showing improved performance compared to benchmarks and validating the contribution of its memory components. Source code and data are available at: https://github.com/bulentsoykan/TESO.

TESO Tabu Enhanced Simulation Optimization for Noisy Black Box Problems

TL;DR

TESO addresses the core challenge of noisy, expensive black-box simulation optimization by marrying Tabu Search principles with Elite Memory in a direct-search framework. The algorithm uses a short-term Tabu List to diversify searches and an Elite Memory to intensify around high-potential regions, guided by an aspiration criterion and an adaptive noise schedule to cope with stochastic evaluations. Empirical results on a stochastic M/M/k queue optimization demonstrate that full TESO achieves the best convergence speed and reliability compared with ablated variants and random sampling, highlighting the complementary roles of diversification and intensification. The work advances practical SO by providing a robust, memory-enhanced methodology that does not rely on surrogate modeling, offering a scalable approach to hard noisy optimization problems with real-world applicability in operations research and systems design.

Abstract

Simulation optimization (SO) is frequently challenged by noisy evaluations, high computational costs, and complex, multimodal search landscapes. This paper introduces Tabu-Enhanced Simulation Optimization (TESO), a novel metaheuristic framework integrating adaptive search with memory-based strategies. TESO leverages a short-term Tabu List to prevent cycling and encourage diversification, and a long-term Elite Memory to guide intensification by perturbing high-performing solutions. An aspiration criterion allows overriding tabu restrictions for exceptional candidates. This combination facilitates a dynamic balance between exploration and exploitation in stochastic environments. We demonstrate TESO's effectiveness and reliability using an queue optimization problem, showing improved performance compared to benchmarks and validating the contribution of its memory components. Source code and data are available at: https://github.com/bulentsoykan/TESO.
Paper Structure (18 sections, 2 equations, 2 figures, 1 table, 1 algorithm)

This paper contains 18 sections, 2 equations, 2 figures, 1 table, 1 algorithm.

Table of Contents

  1. Introduction
  2. Background and Related Work
  3. Simulation Optimization (SO)
  4. Metaheuristics in Simulation Optimization
  5. Memory-Based Heuristics: Tabu Search and Elite Memory
  6. TESO: Algorithm Methodology
  7. Framework Overview
  8. Algorithm Methodology
  9. Numerical Experiments
  10. Experimental Objectives
  11. Test Problem: M/M/k Queue Optimization
  12. Benchmark Algorithms for Comparison
  13. Experimental Design
  14. Results and Discussion
  15. Comparison against Benchmarks
  16. ...and 3 more sections

Figures (2)

  • Figure 1: Flow Diagram of the TESO Algorithm, illustrating the iterative cycle including candidate generation, tabu/aspiration filtering, evaluation, and memory updates.
  • Figure 2: Convergence Plot for Queue Optimization