Customer forecasts and supplier commitment and production decisions in supply contract relationships

Manufacturing firms in capital-intensive industries face inherent demand volatility for their products and the inability to change their capacity quickly. To cope with these challenges, manufacturers often enter into contracts with their customers that offer greater certainty of supply in return for more predictable orders. This dissertation focuses on two such contractual agreements found in practice. We develop analytical models of the supplier's decisions under these contracts and characterize his optimal strategies. One contract also involves a decision on the part of the customer, and we provide a complete analysis of his optimal strategy. The first contract specifies a supply and purchase commitment range for each customer, and penalties for deviating from these ranges. We characterize the supplier's optimal policy for allocating capacity to customers in the current and future periods when the customers face uncertain demands. The optimal allocation may lead to poor service for large customers paying low prices unless penalties are set judiciously. We also show that giving a customer more flexibility in his purchase quantity may increase the supplier's profits. The second contract is a Forecast-Commitment contract in which the customer provides a forecast for a future order and a guarantee to purchase a fraction (perhaps all) of it, and the supplier chooses a supply commitment quantity. There are penalties for the supplier if he fails to commit to the entire forecast and/or to deliver the committed quantity. We characterize the supplier's optimal commitment and production policy and show that the contract serves to limit both the customer's incentive to over-forecast and the supplier's motivation to under-produce. We model the customer's decision of choosing a forecast as a sequential game, with the customer as the first mover. When the penalties are tacit (loss of goodwill), the customer's optimal forecast depends only on the supplier's production strategy. On the other hand, when the supplier makes penalty payments to the customer, the set of candidate equilibria is more complex and depends on both the supplier's commitment and production decisions. We show that the penalty payments serve as a subsidy and thus, sometimes increase the customer's optimal forecast.