Abstract With the rising concern of environmental pollution caused by fossil fuels, the energy efficiency becomes a key issue for the energy sensitive company in order to save the electricity cost and take responsibility for the sustainable development. Thus, an effective energy-aware operations management for the manufacturing plant is crucial to improve its competitiveness in the global market nowadays. We propose a mathematical model integrating three interrelated operational aspects including the production, maintenance and energy for the flow shops under Time-of-Use electricity tariff. A two-layer math-heuristic is devised to solve the model efficiently based on the decomposition of decision variables. In the outer layer, the jobs’ sequences and the buffer times are optimized using the metaheuristic based on genetic algorithm. In the inner layer, the maintenances’ positions and machines’ on/off are optimized using the exact method based on dynamic programming algorithm. Compared with CPLEX and traditional GA, the math-heuristic can get the near optimal solution with a small gap for the small-sized problems and perform well for the large-sized problems. The tradeoff between energy cost and makespan shows that more profit can be achieved using our model for the instance with a later production deadline. Finally, numerical experiments are also conducted to analyze the structure of the optimal solution in order to provide the managerial insights.

中文翻译：

---

分时电价下流水车间的能源感知运营管理

摘要 随着化石燃料对环境污染的日益关注，能源效率成为能源敏感型企业节约用电成本、实现可持续发展的关键问题。因此，对制造工厂进行有效的能源感知运营管理对于提高其在当今全球市场上的竞争力至关重要。我们提出了一个数学模型，整合了三个相互关联的运营方面，包括分时电费下流水车间的生产、维护和能源。设计了两层数学启发式算法以基于决策变量的分解有效地求解模型。在外层，使用基于遗传算法的元启发式优化作业的序列和缓冲时间。内层采用基于动态规划算法的精确方法优化维护位置和机器的开/关。与 CPLEX 和传统 GA 相比，数学启发式算法对于小规模问题可以得到接近最优解，差距很小，对于大问题表现良好。能源成本和完工时间之间的权衡表明，使用我们的模型可以为生产期限较晚的实例获得更多利润。最后，还进行了数值实验来分析最优解的结构，以提供管理见解。math-heuristic对于小问题可以得到接近最优解，差距很小，对于大问题表现良好。能源成本和完工时间之间的权衡表明，使用我们的模型可以为生产期限较晚的实例获得更多利润。最后，还进行了数值实验来分析最优解的结构，以提供管理见解。math-heuristic对于小问题可以得到接近最优解，差距很小，对于大问题表现良好。能源成本和完工时间之间的权衡表明，使用我们的模型可以为生产期限较晚的实例获得更多利润。最后，还进行了数值实验来分析最优解的结构，以提供管理见解。