Dynamic Energy Control for Energy Efficiency Improvement of Sustainable Manufacturing Systems Using Markov Decision Process

Greenhouse gas emissions and global warming have become vital problems to human society. About 40% of the carbon dioxide is emitted from electric power generation in the United States. Due to the lack of consideration in the system design and the lack of a real time systematic management method for energy consumption, the energy efficiency of industrial manufacturing systems is extremely low. Most of the existing research work related to energy efficiency improvement only focuses on a single-machine manufacturing system while little work has been done to achieve the optimal energy efficiency for a typical system with multiple machines and buffers. In this paper, an analytical model is developed to establish a systems (or holistic) view of energy efficiency in typical manufacturing systems with multiple machines and buffers that dynamically control energy consumption considering both energy states and production constraints. The complex interaction between the adopted energy control decisions and system state evolutions are modeled by Markov Decision Process. An approximate algorithm for the real time application is introduced to find a near-optimal solution. A numerical case study on a section of an assembly line is used to illustrate the effectiveness of the proposed approach.