Steelmaking-continuous casting scheduling problem with multi-position refining furnaces under time-of-use tariffs
Ruilin Pan,Qiong Wang,Zhenghong Li,Jianhua Cao,Yongjin Zhang
DOI: https://doi.org/10.1007/s10479-021-04217-7
IF: 4.82
2021-08-12
Annals of Operations Research
Abstract:Multi-position refining furnaces are a critical strategy for energy-intensive industries to meet its demands of fast-paced production. In most literature, however, they serve only as a buffer, holding up to at most two ladles to maintain the proper temperature of ladles. These studies do not take full advantage of them, nor do they study the production scheduling of energy-intensive enterprises with multi-position refining furnaces under time-of-use (TOU) tariffs. Therefore, this paper presents a steelmaking-continuous casting (SCC) scheduling problem with multi-position refining furnaces under TOU tariffs. We firstly develop a mixed integer nonlinear programming (MINLP) model with the goals of minimizing jobs completion time, machines idle time, and total electricity costs, which subjects to the double-position characteristics and other process constraints. Owing to the complexity between the time-slots of TOU tariffs and the processing cycles of jobs, we design an intermediate function to calculate objectives efficiently. Furthermore, a Lagrangian relaxation (LR) algorithm based on a subgradient algorithm is utilized to solve the proposed model, and an interior point algorithm is adopted to solve sub-problems obtained by job-level and batch-level decomposition, whose solution approximates optimality comparing to GUROBI solver. The computational results demonstrate that the solution of job-level decomposition algorithm approximates the optimal scheduling scheme in an acceptable time and is superior to that of GUROBI solver. In addition, double-position instance can find a better scheduling scheme than nondouble-position one.
operations research & management science