Towards dominant flexibility configurations in strategic capacity planning under demand uncertainty

The research herein addresses the strategic capacity planning problem in a multi-product, multi-plant setting under demand uncertainty. Past research provides a number of useful insights on the relationship between the degree of flexibility and the optimal product-plant configuration, based on product demand, product price, and plant capacity. A few prior studies, however, have addressed both capacity and product-plant configuration together as decision variables. In this paper, we develop a two-stage stochastic programming model to determine the capacity and the product-plant configuration to maximize the expected profit. For a four-product, four-plant setting, a computational study is conducted using the sample based optimization procedure with the assumption that the product demand follows a multivariate normal distribution. The model is solved in order to understand the effect of the product-plant configurations on expected profit and investment in capacity. We also examined the extent to which the observations are sensitive to product price, flexibility investment cost, average demand, variance in demand, and product demand correlation. The results provide some new insights on the flexibility structure in capacity planning. For example, with capacity as the decision variable, none of our optimal solutions support a ‘closed chain’ structure. For asymmetric product demand, high-demand products with positive correlation should be allocated to dedicated plants. Some of the optimal flexibility structures required investment in three different degrees of flexibility: dedicated, two-product flexible and three-product flexible. In conclusion, we provide several managerial insights based on the results of the study.