A JIT part supply scheduling strategy with electric transport device between central receiving store and supermarkets in the automobile industry considering energy

Abstract

Sustainable scheduling has been attracting arousing attention of modern manufacturers, and energy efficiency becomes a critical issue related to sustainability. This paper aims at providing an effective solution method for a real-world energy-efficient part supply scheduling problem (EPSSP) between the central receiving store and supermarkets with electric transport device on purpose of coordinating production and transportation in the automobile industry. A bi-objective mixed integer linear programming (MILP) model is formulated to jointly optimize the total energy consumption and a just-in-time (JIT) metric. An epsilon–constraint approach is presented to obtain the optimal solutions in small-scale problems. Owing to EPSSP's N P-hard nature, an indicator-based bi-objective teaching-learning-based optimization (I-BTLBO) algorithm is proposed by combining the binary quality indicator and teaching-learning-based optimization (TLBO) metaheuristic. I-BTLBO adopts a novel solution codification embedded with an adjustment technique to accommodate EPSSP's characteristic. For infeasible solutions, penalty function is applied to deal with the solution that violate vehicle capacities. A variable neighborhood search-based local optimizer is introduced to enhance I-BTLBO's convergence ability. In addition, an external population (EP) is employed to preserve non-dominated solutions' diversity and to guide the evolution process based on the Pareto dominance and preference indicators. Experimental studies demonstrate that the developed novel approach can effectively and efficiently solve EPSSP problem at different scales, and a case study is carried out to give instructive significance on real-world production. • A bi-objective MILP is formulated to minimize energy consumption and stocks. • Introduce an epsilon–constraint approach to obtain small-scale optimal solutions. • Propose an I-BTLBO algorithm to solve EPSSP. • Develop a novel solution codification with adjustment technique. • Develop a variable neighborhood search-based local optimizer.