Essays on the allocation of scarce capacity among multiple market segments.

Abstract

We consider settings in which: (1) A firm's manufacturing facilities are shared by multiple market segments, and these segments have different characteristics. (2) Production capacity is constrained and inflexible. (3) Demand is random. We provide mathematical models that are formulated to represent a particular applied setting---the automotive glass industry. Here, glass from the same manufacturer is installed into new vehicles and also into existing vehicles under repair. These two segments have different characteristics (e.g. very different prices) but share the same production facilities. Relevant managerial issues are the allocation (primarily through long-term contracts) of capacity between the segments, product pricing, and production sequencing. Our first model applies a Markov decision process formulation to a mixed make-to-stock/make-to-order production system. We provide structural results about optimal production sequencing and customer admission policies. The second model is a single period model that captures the profit implications of different allocation decisions. Lastly, we place a version of our second model into a price-competitive setting. We find that several of our previous results are now violated; we analyze how, why, and when this occurs.