Scheduling of a hub reentrant job shop to minimize makespan.

This paper focuses on a hub reentrant shop scheduling problem which is motivated by actual packing production line in the iron and steel industry. There are five operations for processing a job where three operations must be processed on a hub machine, and between any two consecutive operations on the hub machine, two operations must be processed on other two secondary machines, respectively. We mainly consider two types of the problem. The first is basic hub reentrant shop (BHRS) problem where only one machine in each secondary machine center. The second is hybrid hub reentrant shop (HHRS) problem where multiple machines in each secondary machine center. The objective is to minimize the makespan. For BHRS problem, we show that the problem is NP-hard in the strong sense. Some properties are derived for identifying an optimal schedule. Furthermore, a heuristic algorithm is proposed with the worst performance ratio 6:5, and this ratio is demonstrated tight. For HHRS problem, two well-solvable cases are proposed, respectively. Under a split condition, HHRS is equivalent to a two-stage hybrid flow shop problem. The worst case of a class of proposed algorithms is analyzed. The performance ratio is demonstrated relatively with the secondary machine numbers. [ABSTRACT FROM AUTHOR]