Research
Research Program
Learning interpretable decision rules for dynamic scheduling and industrial optimization
At the State Key Lab of Autonomous Intelligent Unmanned Systems (自主智能无人系统全国重点实验室), I study how evolutionary computation, genetic programming, reinforcement learning, and data-driven optimization can be used to learn robust scheduling heuristics for environments where jobs, resources, and objectives change over time.
Directions
Core research areas
Dynamic Flexible Job Shop Scheduling
Learning dispatching rules and adaptive policies for flexible manufacturing systems with changing job arrivals, machine states, and operational objectives.
Genetic Programming for Heuristic Learning
Developing genetic programming methods including lexicase selection, diverse partner selection, semantic search, niching, and quality diversity for interpretable heuristic discovery.
Multi-Objective and Data-Driven Optimization
Designing methods for Pareto set learning, multi-objective decision making, and expensive optimization under dynamic industrial constraints.
AI for Manufacturing and Service Systems
Applying scheduling and decision optimization methods to fog-computing workflows, inventory control, replenishment, transshipment, and intelligent manufacturing operations.
Methods