Abstract

We review dynamic quality models both in single-agent setup and in a competitive framework. Our objectives are: (1) to give the reader a vantage point on the state of the art in this area, (2) to identify the boundaries between the different concepts of quality to help build a bridge between the various communities interested in the management of quality, and (3) to sketch out a research agenda for the area. The paper should not only be of interest to active researchers in the field, but also to a broader community of scholars and practitioners working in operations management, marketing, industrial engineering and operations research, who are interested in quality dynamics.

Introduction

During the three last decades or so, a large literature has developed on product quality management, with one stream using a dynamic model to reflect the idea that firms can improve the quality of their products over time. This paper surveys this dynamic literature both in a one-firm framework and in a competitive setup. There are several reasons for adopting a dynamic model when dealing with product’s quality. First, consumers’ tastes change over time and if the firm wants to remain competitive (not to say survive) in the market, it must upgrade continuously the quality of its products. Second, changing product’s quality requires constant investments over time to, e.g., improve production processes and use better materials. Third, environmental and safety regulations change over time and so must the product’s quality about may impose some. Finally, establishing a high-quality reputation needs sustained marketing as well as operations efforts.

We all have probably witnessed two friends discussing a product quality or brand’s quality and ending in a deadlock. One likely reason for this outcome is that each party had her own personal definition of quality. Such differences in opinion also occasionally involve consumers and firms. For instance, a new product can be a true marvel from an engineering and production point of view, but be snubbed by consumers because it does not suit their tastes, and therefore be labeled a poor-quality product. Then, one can ask the following simple question: can quality be assessed through a reliable and valid measurement scale? In a seminal paper, Garvin (1988) suggested the following eight dimensions to measure this construct:

Conclusion
This paper surveyed the contributions on quality dynamics both in single-agent and in competitive situations. In this section, we propose some ideas for future investigations, with some of them being relatively straightforward extensions to what has been done, and others being more conceptual (and sometime speculative) ones.

Stochastic models: Looking at the “Model” column in the Appendix, we clearly see that the literature has largely adopted a deterministic model. Given that one can hardly argue that e.g., future demands and costs are known with certainty, the choice of a deterministic setup is, at least partially, motivated by mathematical tractability. Still, this choice has allowed researchers to obtain a series of insights into the management of quality over time. Extending most of the proposed deterministic models to a stochastic environment is conceptually easy, but it will come with the cost of losing the opportunity of having closed-form solutions. Also, depending on how uncertainty is introduced, obtaining a numerical solution may be computationally challenging, especially for multi-agent models with strategic interactions. One good place to start the analysis is to consider a two-stage model where quality decision is made once, while other decisions, e.g., pricing and advertising, are made in both periods. This would allow one to gain intuition on how uncertainties in, e.g., demand or cost, affect the quality decision in current and future periods. Next, one can move to a multistage model where quality decision is made in each period.