This thesis handles fundamental problems faced by retailers everyday: how do consumers make choices from an enormous variety of products? How to design a product portfolio to maximize the expected profit given consumers’ choice behavior? How to frame products if consumers’ choices are influenced by the display location? We solve those problems by first, constructing mathematical models to describe consumers’ choice behavior from a given offer set, i.e., consumer choice models; second, by designing efficient algorithms to optimally select the product portfolio to maximize the expected profit, i.e., assortment optimization. This thesis consists of three main parts: the first part solves assortment optimization problem under a consideration set based choice model proposed by Manzini and Mariotti (2014) [Manzini, Paola, Marco Mariotti. 2014. Stochastic choice and consideration sets. Econometrica 82(3) 1153-1176.]; the second part proposes an approximation algorithm to jointly optimize products’ selection and display; the third part works on optimally designing a product line under the Logit family choice models when a product’s utility depends on attribute-level configurations.
| Identifer | oai:union.ndltd.org:columbia.edu/oai:academiccommons.columbia.edu:10.7916/D8B86M2H |
| Date | January 2018 |
| Creators | Li, Anran |
| Source Sets | Columbia University |
| Language | English |
| Detected Language | English |
| Type | Theses |
Page generated in 0.002 seconds