Preference-Agile Multi-Objective Optimization for Real-time Vehicle Dispatching

2026-04-12

AI Summary

This work addresses the challenge of real-time responsiveness to evolving user preferences in dynamic multi-objective optimization (DMOO) and the inability of existing methods to handle complex, sequential decision-making in real-world scenarios. To this end, we propose Preference-Agile Multi-Objective Optimization (PAMOO), a novel approach that integrates a unified model within a deep reinforcement learning framework. PAMOO explicitly accepts dynamic preference vectors at runtime and incorporates a calibration function to align the policy with user-specified preferences. As the first method enabling interactive preference adjustment during the decision process, PAMOO overcomes the limitations of conventional static or non-sequential DMOO approaches. Evaluated on a container terminal vehicle dispatching task, PAMOO significantly outperforms state-of-the-art multi-objective optimization algorithms, demonstrating superior adaptability and generalization capability.

Technology Category

• Search and Optimization: Sampling/Simulation-based Search
• Machine Learning: Learning Preferences or Rankings
• Reasoning under Uncertainty: Stochastic Optimization

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Fairness, privacy, and diversity in economic environments

Abstract

Multi-objective optimization (MOO) has been widely studied in literature because of its versatility in human-centered decision making in real-life applications. Recently, demand for dynamic MOO is fast-emerging due to tough market dynamics that require real-time re-adjustments of priorities for different objectives. However, most existing studies focus either on deterministic MOO problems which are not practical, or non-sequential dynamic MOO decision problems that cannot deal with some real-life complexities. To address these challenges, a preference-agile multi-objective optimization (PAMOO) is proposed in this paper to permit users to dynamically adjust and interactively assign the preferences on the fly. To achieve this, a novel uniform model within a deep reinforcement learning (DRL) framework is proposed that can take as inputs users' dynamic preference vectors explicitly. Additionally, a calibration function is fitted to ensure high quality alignment between the preference vector inputs and the output DRL decision policy. Extensive experiments on challenging real-life vehicle dispatching problems at a container terminal showed that PAMOO obtains superior performance and generalization ability when compared with two most popular MOO methods. Our method presents the first dynamic MOO method for challenging \rev{dynamic sequential MOO decision problems

Problem

• dynamic multi-objective optimization
• real-time vehicle dispatching
• preference adjustment
• sequential decision making
• human-centered decision making

Innovation

• preference-agile
• dynamic multi-objective optimization
• deep reinforcement learning
• sequential decision making
• real-time vehicle dispatching