Dynamic pricing models to improve supply chain performance

Many industries are beginning to use innovative pricing techniques to improve inventory control, capacity utilization, and ultimately the profit of the firm. In manufacturing, the coordination of pricing and production decisions offers significant opportunities to improve supply chain performance by better matching supply and demand.; In this work, we focus on dynamic pricing in manufacturing, where prices and production values are determined jointly in a multi-period horizon with non-perishable inventory. We develop and analyze algorithms for various planning strategies and provide computational analysis to generate managerial insights.; We initially study a Full Planning problem where all decisions are made at the beginning of a horizon. We assume that demand is non-stationary but the relationship between price and demand is known; available production capacity is limited and unmet demand is lost. The objective is to maximize profit under conditions of concave revenue curves and periodically varying demand, capacity, inventory holding and production costs. We show that this problem can be formulated as an optimization problem where the constraint set characterizes a polymatroid and the objective function belongs to a class of functions referred to as lightly concave. We prove that a greedy procedure provides the optimal solution for problems in this class and thus an optimal strategy for the pricing problem. We discuss extensions of the Full Planning strategy to the case with multiple products sharing common production capacity.; We then analyze a similar model except that demand is a general, non-stationary stochastic function of price. We introduce Partial Planning strategies, where some decisions are made at the beginning of the horizon while other decisions are delayed until demand in previous periods is realized. For example, in Delayed Production, pricing is determined at the beginning of the horizon, and the production decision is made at the beginning of each period before customer orders are received. Alternatively, in Delayed Pricing, production is determined at the beginning of the horizon, and the pricing decision is made at the beginning of each period before customer orders are received. A special case of the Delayed Production policy we describe is the fixed price policy in which the firm determines a single price at the beginning of the horizon that is applied in every period. We describe an important feature of our partial planning models which is the ability to set aside products to satisfy future demand, even if this implies lost sales in the current period. We develop heuristics for the strategies based on a deterministic approximation of the pricing problem and analyze the worst-case performance of the Delayed Production policy.; We conduct an extensive computational study of both the Full and Partial Planning strategies, particularly focusing on the impact of pricing on supply chain performance and the performance of the various partial planning strategies relative to each other. Finally, we conclude with a case study from the automotive industry.