Date of Award

Spring 2010

Document Type


Degree Name

Doctor of Business Administration (DBA)



First Advisor

Hani Mesak


This dissertation is composed of three journals examining the interfaces between the marketing variable of advertising and various aspects of the operations function of the enterprise, namely, (1) production cost [Chapter 2], (2) inventory control [Chapter 3], and (3) service cost learning [Chapter 4]. The first journal identified the optimum advertising allocation policy over time in the presence of a quadratic convex/concave production cost function when the advertising response function is concave using a modified Vidale-Wolfe model. Through analytical proofs and numerical simulations, the results indicated the potential superiority of a pulsation policy in the presence of concavity in the advertising response function only if the production cost function is convex; otherwise, the uniform policy would be optimal. The study is seen as applicable to frequently purchased products in the maturity stage of their life cycles of dominant firms in their industries practicing a zero-inventory policy in a just-in-time environment.

The research objective pertaining to the second journal was to study how a firm would adapt optimum ordered quantity/production lot size and optimum advertising expenditure in response to changes in its own parameters, rival's parameters, or parameters that are common to all firms in a symmetric duopoly/oligopoly market. This was accomplished by developing comparative statics (sensitivity analysis) of a symmetric competitive inventory model with advertising-dependent demand based on a market share attraction model. Both optimum advertising expenditure and ordered quantity were found to be sensitive to changes in marketing and operations parameters. The robustness of the symmetric comparative statics was assessed by using data from the brewing industry in the US that represents an asymmetric oligopoly. The empirical analysis indicated that the theoretical results obtained for a symmetric oligopoly remained valid for an oligopoly where each firm had a market share less than 50% and the market shares were further apart from one another. The study is thought to be applicable to low-priced frequently purchased consumer items in competitive mature markets.

In the third journal, the original Bass model for new products was modified to incorporate advertising and customers' disadoption to characterize the optimum advertising policy over time for subscriber service innovations where service cost follows a learning curve. After characterizing the optimal policy for a general diffusion model, the results pertaining to a specific diffusion model for which advertising affects the coefficient of innovation were reported. On the empirical side, four alternative diffusion models were estimated and their predictive powers, using a one-step-ahead forecasting procedure, were compared. Empirical research findings suggest that the specific diffusion model considered in this study is not only of theoretical appeal, but also of notable empirical relevance. Taken together, the analytical and empirical findings argue in favor of advertising more heavily during the early stage of the diffusion process of the new subscriber service innovation and including a related message that would predominantly target innovators.

Furthermore, it might be inappropriate to model the diffusion of subscriber services as if they were durable goods. The study is thought to be applicable to service innovations that are made available to customers periodically at a subscription fee. Typical examples include, but are not limited to, cable TV, health clubs, pest control, and the internet.