Please use this identifier to cite or link to this item:
Title: Product line selection, inventory and contracting for inter-dependent products
Authors: TAO HUA
Keywords: Utility Correlation, Semi-definite Programming, Multinomial Logit Model, Cross Moment Model, Inventory, Contracting
Issue Date: 14-Aug-2009
Citation: TAO HUA (2009-08-14). Product line selection, inventory and contracting for inter-dependent products. ScholarBank@NUS Repository.
Abstract: Product proliferation has become so common that most companies now offer hundreds, if not thousands, of stock keeping units (SKUs) in order to compete in the market place. High correlations may exist among the customers' utilities of these products due to the common attributes among them. These correlations may affect the demand for each product, which makes demand forecasts and production/inventory decisions even harder. Therefore, such correlations should be properly incorporated into the supply chain management to improve the profitability. This thesis developed a product line selection model in conjunction with a utility maximization model to describe the choice behavior of customers. Semi-definite Programming (SDP) is used to approximate the expected utility and the customer choice probabilities. The product line selection problem is then solved by incorporating the SDP approach with popular product swapping and greedy heuristics. With the ability to incorporate the correlation between products arising from common attributes in the choice behavioral model, this model successfully address the issue of Independence of Irrelevant Attributes (I.I.A.) property, which is an inherent limitation of the popular Multinomial Logit (MNL) model. We compare the performance of the new SDP model with the classic MNL based product line selection model in a simulated example. Our experimental results indicate that for both the buyer's welfare problem and seller's profit problem, our model can lead to better design of the product line, and can perform significantly better than MNL model, especially when the products share many common attributes. Furthermore, we also extend the above work to include the inventory decisions. We embed our Cross Moment Model into the assortment and inventory joint decision problem for retailers, and focus on comparing the resulting offer set and inventory levels decision with those decision under classic MNL choice models. We also quantify theimprovement of the total expected profits through Monte Carlo simulation. We found that under static substitution, less correlated products set can bring more profit. We also show that the total varieties of products can be reduced under dynamic substitution. And through simulation, considerable improvement in expected profits results from taking account of utilities' correlations.
Appears in Collections:Ph.D Theses (Open)

Show full item record
Files in This Item:
File Description SizeFormatAccess SettingsVersion 
TaoH.pdf1.98 MBAdobe PDF



Page view(s)

checked on Apr 20, 2019


checked on Apr 20, 2019

Google ScholarTM


Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.