Research
Our lab investigates the intersection of AI, optimization, and systems engineering to address real-world challenges in manufacturing, energy systems, and other engineering applications. Some examples of our research are shown below.
Advanced Manufacturing System
Explainable multi-agent deep reinforcement learning for manufacturing system scheduling
We develop deep reinforcement learning frameworks that integrate domain knowledge from manufacturing system modeling for dynamic scheduling and process control in manufacturing environments. Our research focuses on real-time decision-making under uncertainty, enabling adaptive scheduling and control strategies that not only respond to disruptions and changing system conditions, but also ensures critical production constraints such as production order fulfillment.
Deep reinforcement learning for adaptive disassembly scheduling
In this study, a constrained multi-agent deep reinforcement learning approach is proposed to maximize the disassembly profit by dynamically changing the batch mixing ratios of different-sized components in self- disassembly workstations and adapting real-time scheduling to stochastic product quality, changes in operational sequences, and self-disassembly failures.
Energy Management
Demand response for fast-charging battery powered material handling equipment
In this study, an analytical manufacturing and material handling system model is proposed to obtain cost-effective production schedules under demand response. The interactions between machines and MHE, production throughput requirement, battery charging characteristic, and time- varying electricity pricing are jointly considered in the integrated model to help manufacturers reap substantial benefits of demand response.