Real-Time Execution Systems

In the previous chapters, we developed planning models based on assumptions such as imbalances in net inventories at retailers would not occur, demand processes were independent and identically distributed at and among retailers, and allocations from the central warehouse to the retailers occurred on a first-come, first-served basis. In this chapter we make none of these assumptions. Rather than focusing on planning inventories in supply chains, we construct models that can be employed to assist in making real-time decisions for production and allocation of inventories. In Chapter 6, we explore the question of what do we do now. That is, we develop mathematical approaches for managing production and distribution in real-time when demands are time-dependent and uncertain at each of many locations that are supplied by a central facility. We begin by determining how to allocate inventory in real time in a two echelon distribution system. We then develop a sequence of increasingly complex models for managing the repair and allocation of aviation items in real time. Demand for these items is uncertain every day at every operating location. We show how these several problem environments can be represented as linear programs.