We propose to use Deep Neural Networks to solve data-driven stochastic optimization problems. Given the historical data of the observed covariate, taken decision, and the realized cost in past periods, we train a neural network to predict the objective value as a function of the decision and the covariate. Once trained, for a given covariate, … Read more
We study a class of robust assortment optimization problems that was proposed by Farias, Jagabathula, and Shah (2013). The goal in these problems is to find an assortment that maximizes a firm’s worst-case expected revenue under all ranking-based choice models that are consistent with the historical sales data generated by the firm’s past assortments. We … Read more
In this paper, we study a retailer price optimization problem which includes the practical constraints: maximum number of price changes and minimum amount of price change (if a change is recommended). We provide a closed-form formula for the Euclidean projection onto the feasible set defined by these two constraints, based on which a simple gradient … Read more
In this paper, we solve the multiple product price optimization problem under interval uncertainties of the price sensitivity parameters in the demand function. The objective of the price optimization problem is to maximize the overall revenue of the firm where the decision variables are the prices of the products supplied by the firm. We propose … Read more
When a firm selects an assortment of products to offer to customers, it uses a choice model to anticipate their probability of purchasing each product. In practice, the estimation of these models is subject to statistical errors, which may lead to significantly suboptimal assortment decisions. Recent work has addressed this issue using robust optimization, where … Read more
The fraction of customers who choose a particular item from among a set of available items can be increased significantly by the inclusion of a related inferior (and apparently irrelevant) item in the choice set. This violation of the independence from irrelevant alternatives and the regularity properties is called the decoy effect, dominance effect, or … Read more
This paper describes a revenue management project with a major airline that operates in a fiercely competitive market involving two major hubs and having more than 30 parallel daily flights. The market has a number of unusual characteristics including (1) almost half of customers choose not to purchase the tickets after booking; (2) about half … Read more
We propose an approach to the timing of markdowns over a finite time horizon that does not require the precise knowledge of the underlying probabilities, instead relying on range forecasts for the arrival rates of the demand processes, and that captures the degree of the manager’s risk aversion through intuitive budget-of-uncertainty functions. These budget functions … Read more
This paper studies a multi-stage stochastic programming model for large-scale network revenue management. We solve the model by means of the so-called Expected Future Value (EFV) decomposition via scenario analysis, estimating the impact of the decisions made at a given stage on the objective function value related to the future stages. The EFV curves are … Read more
Revenue management (RM) enhances the revenues of a company by means of demand-management decisions. An RM system must take into account the possibility that a booking may be canceled, or that a booked customer may fail to show up at the time of service (no-show). We review the Passenger Name Record data mining based cancellation … Read more