In this paper we present a mould shop scheduler (MSS) that was developed for the Moulding Division at a major US telecommunications equipment manufacturer. This plant consists of close to 100 injection-moulding machines, and over 1000 different moulds are used to manufacture more than 1500 different parts. The MSS procedure reads the MRP orders at the start of each week, and then generates for each machine the sequence of parts or orders to be processed, the production quantity, and the sequence of mould and material changeovers. The MRP orders are generated by a corporate system that is responsible for plant-wide scheduling. In developing the daily schedule, the objectives of the MSS procedure are to: (i) reduce the total time spent on setups; and (ii) minimize the inventory of finished parts. The MSS procedure consists of four heuristic algorithms that are sequentially solved. The use of the algorithms is illustrated by an example, and a comparison with a lot-for-lot MRP implementation is shown.
[1]
Thomas E. Morton,et al.
Myopic Heuristics for the Single Machine Weighted Tardiness Problem
,
1982
.
[2]
Esther M. Arkin,et al.
Weighted-Tardiness Scheduling on Parallel Machines with Proportional Weights
,
1991,
Oper. Res..
[3]
David L. Woodruff,et al.
CONWIP: a pull alternative to kanban
,
1990
.
[4]
D Silva.
Capacity management: get the level of detail right.
,
1994,
Hospital materiel management quarterly.
[5]
T.C.E. Cheng,et al.
Parallel-Machine Scheduling Problems with Earliness and Tardiness Penalties
,
1994
.
[6]
T.C.E. Cheng,et al.
A state-of-the-art review of parallel-machine scheduling research
,
1990
.
[7]
Kut C. So,et al.
Some heuristics for scheduling jobs on parallel machines with setups
,
1990
.