Multi-job lot streaming to minimize the mean completion time in m-1 hybrid flowshops

In this paper we study the multi-job lot-streaming problem in two-stage hybrid flowshops with m identical machines at the first stage and a single machine at the second stage. The objective is to minimize the mean completion time of the jobs. We propose two heuristic algorithms to solve the problem. Both algorithms share the same approach that first sequences the jobs and then schedules (with splitting) the jobs in the sequence one at a time. But they use different methods to sequence the jobs. A mixed integer linear programming (MILP) formulation is also given and used to calculate a lower bound. This lower bound is used jointly with two other lower bounds obtained from the analysis of dominant stages. Extensive computational experiments are carried out comparing the heuristic solutions with the lower bounds.