Managing mobile production-inventory systems influenced by a modulation process

The objective of this paper is to investigate the potential added value being able relocate production capacity, relative fixed in a network multiple, geographically distributed manufacturing sites. There growing number examples capacity that can be relocated with modest amount effort; e.g., 3D printers, bioreactors for cell and gene manufacturing, modular units pharmaceutical intermediates. Such capability shows promise enabling fast fulfillment reduction total inventory without sacrificing customer service levels or system resilience. Allowing also transshipment, we model production-inventory L sites Y relocatable develop efficient effective heuristic solution methods dynamic relocation multi-location control, analyze implementation challenges capacity. We describe (L, Y) problem as sequential decision making under uncertainty determine mobile relocation, replenishment decisions at each epoch. To enhance realism, use partially observed stochastic process, modulation exogenous observable forces (e.g., macro-economy) affect demand. then Markov process. Due considerable computational solving exactly, propose two efficient, high quality heuristics. show an instance set five locations mobility case, improve systems performance by much 41% on average over no-flexibility case yield 10% more savings compared when only transshipment permitted.