Abstract:When tackling high-dimensional, continuous simulation-based optimization (SO) problems, it is important to balance exploration and exploitation. Most past SO research focuses on the enhancement of exploitation techniques. The exploration technique of an SO algorithm is often defined as a general-purpose sampling distribution, such as the uniform distribution, which is inefficient at searching high-dimensional spaces. This work is motivated by the formulation of exploration techniques that are suitable for large-scale transportation network problems and high-dimensional optimization problems. We formulate a sampling mechanism that combines inverse cumulative distribution function sampling with problem-specific structural information of the underlying transportation problem. The proposed sampling distribution assigns greater sampling probability to points with better expected performance as defined by an analytical network model. Validation experiments on a toy network illustrate that the proposed sampling distribution has important commonalities with the underlying and typically unknown true sampling distribution of the simulator. We study a high-dimensional traffic signal control case study of Midtown Manhattan in New York City. The results show that the use of the proposed sampling mechanism as part of an SO framework can help to efficiently identify solutions with good performance. Using the analytical information for exploration, regardless of whether it is used for exploitation, outperforms benchmarks that do not use it, including standard Bayesian optimization. Using the analytical information for exploration only yields solutions with similar performance than when the information is used for exploitation only, reducing the total compute times by 65%. This paper sheds light on the importance of developing suitable exploration techniques to enhance both the scalability and the compute efficiency of general-purpose SO algorithms. Funding: T. Tay thanks the Agency for Science, Technology and Research (A*STAR) Singapore for funding his work. Supplemental Material: The online appendix is available at https://doi.org/10.1287/trsc.2023.0110 .

A comparison of sample-path-based simulation-optimization and stochastic decomposition for multi-location transshipment problems

Surrogate-Based Optimization for Solving a Mixed Integer Network Design Problem

A Simulation-Based Approach for Sustainable Transportation Systems Evaluation and Optimization: Theory, Systematic Framework and Applications

Parallel Stochastic Optimization Framework for Large-Scale Non-Convex Stochastic Problems

Simulation-based Optimization of Mixed Road Pricing Policies in a Large Real-world Network

A simulation based optimization method for multi-stage production-distribution systems subject to non-analytical constraints

Stochastic Systems Simulation Optimization

Simulation Optimization: A New Approach

A Sampling Strategy for High-Dimensional, Simulation-Based Transportation Optimization Problems

Simulation-Based Discrete Optimization of Stochastic Discrete Event Systems Subject to Non Closed-Form Constraints.

Simultaneous Locomotive Assignment and Train Scheduling on a Single-Track Railway Line: A Simulation-Based Optimization Approach

Multi-objective optimization for multimodal transportation routing problem with stochastic transportation time based on data-driven approaches

Stochastic Optimization of Distributed Sampling System Placement for Passive Measurement

Numerical simulation of optimal transport paths

A Simulation Optimization Framework for Ambulance Deployment and Relocation Problems

Stochastic Multi-Objective Production-Distribution Network Design Using Simulation-Based Optimization

Efficient Optimization Procedures for Stochastic Simulation Systems

A Simulation-Based Optimization Method for Production-Distribution Network Design.

Comprehensive Optimization of Traffic Strategy Based on Simulation

Blackbox Simulation Optimization

A survey on simulation optimization for the manufacturing system operation