Managing product variety in inventory and service operations

The recent explosion in product variety has created a need for better managerial understanding and management tools. Empirical research suggests that higher product variety results in higher costs, with a large part of these costs resulting from higher relative demand uncertainty. The dissertation addresses the problem of managing product variety when products and services are substitutes. Researchers have recognized that in many situations products and services offered to customers may be substitutes for one another. In cases of physical goods, a customer may choose to substitute an out-of-stock product with a similar one (customer-directed substitution). In cases of services, a company may decide to offer upgrade customers who can not be served due to insufficient capacity (company-directed substitution).;First, we consider a static single-period model where customers can select an alternative product (substitute) from a given product-line in case their first-choice product is out of stock. Substitution may occur within a product line offered by one retailer, as well as between products offered by different retailers. We consider both situations within one framework.;We extend the customer-directed substitution problem by analyzing dynamic multi-period competition between two retailers offering substitute products. In a multi-period setting, a new problem arises: if competitors offer substitutable products, the customer may decide to backorder the missing product with the first company she visits, switch to a competitor, or leave. Hence, customer backordering behavior greatly affects inventory decisions and the resulting profitability. We analyze several plausible backordering scenarios and also address situations in which the company can influence customers' backordering behavior by offering economic incentives.;Finally, we analyze the problem of managing product variety under company-directed substitution in services. We consider situations where the company may choose to upgrade the customer (as car rental companies and airlines often do). We look at the problem of optimal capacity investment when single-level upgrades are allowed. We find the optimal capacity decisions, and provide an in-depth analysis of the impact that demand correlation has on the optimal capacity decision.;Throughout the dissertation we demonstrate the advantages of accounting for substitution when allocating limited resources.