Perception data-driven optimization of manufacturing equipment service scheduling in sustainable manufacturing

• A joint model of energy consumption and production efficiency is established.
• We proposed a data-driven optimization scheduling method which is dynamic.
• A novel Pareto-based bees algorithm is proposed for scheduling optimization.

Both sustainable manufacturing and manufacturing service are the trends in industry because they are regarded as ways to reduce the resource cost and energy consumption in manufacturing process, to improve the flexibility and responding speed to customers’ demand, and to improve the production efficiency. In order to improve the sustainability of manufacturing equipment services in job shop, this paper presents a multi-objective joint model of energy consumption and production efficiency. The model is related to multi-conditions of manufacturing equipment services. The conditions are monitored in real-time to drive a multi-objective dynamic optimized scheduling of manufacturing services. In order to solve the multi-objective problem, an enhanced Pareto-based bees algorithm (EPBA) is proposed. In order to ensure the variety of population, to prevent the premature convergence, and to improve the searching speed, several key technologies are utilized such as variable neighborhood searching, mutation and crossover operation, fast non-dominated ranking, critical path local search, archive Pareto set, critical path taboo set, etc. Finally, the proposed method is evaluated and shows better performance in static and dynamic scenarios compared with the existing optimization algorithms.