Simultaneous production planning and pricing: Theory, empirical estimation, and supply chain contracts

Simultaneous pricing and quantity setting are further explored in a one-shot, single-product, produce to stock, expected profit maximization model. I assume that, for each given price, demand is a discrete random variable, with a finite number of outcomes. I also assume that each of these outcomes is a piecewise linear function of the price. I show how to approximate a general problem by my model. My approach decomposes the problem into two subproblems: first, finding the best price and quantity when the service level is given as one of the finite number possible, and then choosing the service level to offer. I find that both the service level constrained problem and the unconstrained problem with specialized demand (additive or multiplicative or a mixture of the two) are well behaved, while the unconstrained general problem produces some counterintuitive results.; In the second chapter, I analyze the supply-chain contracts for the case where the retailer can decide not only on his order quantity but also on the retail price. I focus on cases where the power relationship is such that both the retailer and the manufacturer have sufficient power that neither can capture the entire chain profits. Four possible contracts are considered: a price only contract, a buyback contract, a revenue sharing contract where the retailer chooses the portion of revenue to share with the manufacturer and a quantity discount contract with asymmetric information.; In the third chapter of the dissertation, I utilize one of the most commonly used methods for modeling consumer preferences, conjoint analysis, in order to estimate the distribution of demand for the model described above. While the conjoint analysis approach is used mostly in the product design stage, in this chapter, I use conjoint analysis for forecasting purposes of existing products and specifically, I use the product-based conjoint analysis approach. Through a survey that includes 89 respondents, I derive the parameters of the demand distribution as a function of the individual-level utility functions data. I then show how the estimated demand parameters can be used in the optimization problem of the firm by applying the model developed above.