Access Type

Open Access Dissertation

Date of Award

January 2012

Degree Type


Degree Name



Industrial and Manufacturing Engineering

First Advisor

Ratna B. Chinnam


A manufacturer's assortment is the set of products or product configurations that the company builds and offers to its customers. While the literature on assortment planning is growing in recent years, it is primarily aimed at non-durable retail and grocery products. In this study, we develop an integrated framework for strategic assortment planning of configurable products, with a focus on the highly complex automotive industry. The facts that automobiles are highly configurable (with the number of buildable configurations running into thousands, tens of thousands, and even millions) with relatively low sales volumes and the stock-out rates at individual dealerships (even with transshipments) are extremely high, pose significant challenges to traditional assortment planning models. This is particularly the case for markets such as the U.S. that mostly operate in a make-to-stock (MTS) environment. First, we study assortment planning models that account for exogenous demand models and stock-out based substitution while considering production and manufacturing complexity costs and economies-of-scale. We build a mathematical model that maximizes the expected profit for an Original Equipment Manufacturer (OEM) and is a mixed-integer nonlinear problem. We suggest using linear lower/upper bounds that will be solved through a Modified-Branch and Bound procedure and compare the results with commercial mixed-integer nonlinear solvers and show superiority of the proposed method in terms of solution quality as well as computational speed.

We then build a modeling framework that identifies the optimal assortment for a manufacturer of automotive products under environmental considerations, in particular, Corporate Average Fuel Economy (CAFE) requirements as well as life-cycle Greenhouse Gas (GHG) emission constraints. We present a numerical experiment consisting of different vehicle propulsion technologies (conventional, Diesel, and hybrid) and study the optimal shares of different technologies for maximizing profitability under different target levels of CAFE requirements. Finally, we develop assortment planning formulations that can jointly identify optimal packages and stand-alone options over different series of the product model. Our numerical experiment reveals that product option packaging has a considerable effect on managing product complexity and profitability.