Just-in-time production of large-scale assemblies: An integrative perspective

Since its advent, the 'just-in-time'-paradigm has considerable influence on production planning and control. While originally employed for high-volume production, 'just-in-time'-concepts are meanwhile documented for low-volume, make-to-order production as well. An example is the assembly of multiple customer-specific products as recently considered by Agrawal et al. The problem has been inspired by the production planning of Westinghouse ESG, Maryland. Agrawal et al. modelled the problem as MILP and proposed a suited heuristic with acronym LETSA which gives near-optimal solutions. We show that the problem can be modelled as simple single resource-constrained project scheduling problem (RCPSP). This has several advantages: First, it allows an integrated perspective on 'just-in-time'-steered low-volume, make-to-order assemblies. Second, it makes available the rich supply of models, insights, and methods of resource-constrained project scheduling. The latter includes the well-known serial scheduling algorithm which brings forth the same schedule as LETSA with a computational effort which is only O(n2) instead of O(n4), where n is the number of assembly operations.