Capacity and flexibility investment decisions under pricing, product substitution and supply chain performance considerations

This dissertation studies capacity investment decisions of a manufacturing firm facing high demand uncertainty in a make-to-order environment. Two important strategies are discussed to better respond to uncertainty in product demands; manufacturing/capacity flexibility and price flexibility. The first part considers high level models of pricing, production and capacity investment decisions of a firm offering substitutable products, where the price and production decisions are made when demand is realized. The study of the impact of product substitution on the firm\u27s optimal investment strategy and the design of the production network leads to following conclusions: (1) Under realistic assumptions on the distribution of the demand functions, the optimal investment in flexible capacity tends to decrease as the products become closer substitutes. (2) By not taking into account product substitution at the investment stage, the firm significantly overestimates the optimal capacity investment levels. (3) Assigning substitutable products to different production plants increases the firm\u27s expected profits with lower investment costs; this is because the system becomes more flexible since production can be transferred from one plant to the other by diverting the demand for one product to its substitute through pricing. The second part considers a more detailed capacity investment model where prices are fixed at the beginning of the season and the company faces short-term (period-to-period) demand variability associated with a make-to-order environment. An optimization-based simulation model is used to understand the impact of pricing and increased manufacturing flexibility on the performance of the supply chain. The results show that both manufacturing and pricing flexibilities significantly reduce the optimal investment levels while increasing the expected profits, but lead to higher production variability, system inventory and variability observed by suppliers upstream. As a second step, the model is modified to include outbound transportation costs and service constraints, which help to allocate the production to satisfy at least a certain percentage of demand in different demand regions. We show that considering service constraints significantly increases the minimum service level among different demand regions at the expense of higher system inventory and outbound transportation costs, especially when capacity is scarce