Chung Piaw Teo: Data Driven Approach to Pricing and Resource Allocation Problems
Increased computing power and the explosion of data have created opportunities for the OM profession to analyse data to identify new models and approaches to drive decisions and actions. We look at two canonical case studies in this talk. The first case develops an estimation and optimization framework for the multi-product pricing problem, by exploiting properties of the marginal distributions in a class of discrete choice models. We establish a set of closed-form relationships between prices and market shares, and use aggregate sales information from a set of pricing experiments to guide us to the optimal pricing solution. This approach partially addresses the problem of model misspeciﬁcation for pricing problems, since we do not explicitly assume the structural form of the marginal distributions in the consumer’s utility model. The second case develops a real time resource deployment approach to match supply/capacity with demands, incorporating multiple and possibly conflicting objectives in the system. We show that a data driven approach can be used to guide the system to allocate resources so that the performance attained has the smallest deviation away from an utopia point for the system.