Modeling multiple stage manufacturing systems with generalized costs and capacity issues

This study introduces one modeling methodology that describes a broad range of multiple stage production planning issues, including multiple limited resources with setup times and joint fixed cost relationships. An existing production system is modeled in this fashion, creating a new set of 1350 highly generalized benchmark problems. A computational study is conducted with the 1350 benchmark problems introduced in this paper and 2100 benchmark problems, with more restrictive assumptions, from the existing literature. The relative merits of a decomposition-based algorithm and a neighborhood search technique known as NIPPA, or the Non-sequential Incremental Part Period Algorithm, are assessed. NIPPA is generally the more successful of the two techniques, although there are specific instances in which the decomposition-based algorithm displayed a distinct advantage.