Catálogo de publicaciones - libros

This is an abridged and edited transcription of the author’s opening remarks delivered on February 25, 2006.

Saul Gass has practiced operations research for 55 years and continues to be a vigorous presence in the field. His multifaceted contributions as a scholar, practitioner, involved citizen, world ambassador, and chronicler of operations research are reviewed in this article.

This paper is an abridged and edited transcription of the author’s lecture given on February 25, 2006, in honor of Saul Gass.

This paper is an edited transcription of the author’s lecture given on February 25, 2006, in honor of Saul Gass.

This paper is an edited transcription of the author’s lecture given on February 25, 2006, in honor of Saul Gass.

This paper is an edited transcription of the author’s lecture given on February 25, 2006, in honor of Saul Gass.

Introductory operations research and management science textbooks survey a standard set of modeling techniques: linear programming, queueing theory, inventory control, decision trees, etc. The discussion throughout focuses on decision making and often explores the possibility of purchasing additional resources to improve performance. All of the modern texts provide a rich array of real-world examples of successful OR projects. By referring to recognized best practices, we argue that there is another factor: management’s multiple roles that are missing from the problem context. We believe this gap is a factor in why other approaches to problem solving are often more visible and more marketable than OR. We provide examples of discussions of managerial oversight, leadership, and effort that could easily be added to chapters on mathematical programming and queueing, so as to place OR models in a broader context and increase their value.

As the worthy profession of operations research enters the twenty-first century, the outlook is bright in many respects. Yet various impediments to progress deserve attention. In this essay I offer a personal survey of the situation, focused on the delivery of operations-research services in practice. I begin by reviewing the nature and scope of modern operations research. Then I examine the current status of the profession — what’s going well and what isn’t. Finally, I propose actions that I believe promise to improve the opportunities for operations-research professionals and their clients to undertake beneficial projects.

This paper summarizes a talk given in honor of Saul Gass’ 80 Birthday celebration. The paper is modeled after Saul’s well-known book, , and presents some of the illustrations provided during that talk. In this paper, we explain why specific rules might be chosen within a general combinatorial auction framework. The purpose of such rules is to assure that the market mechanism is fair to both buyers and sellers, and so that the auction will end in an efficient outcome, i.e., the goods are won by those that value them the most. The paper describes some of the issues, both computational and economic, that one faces when designing such auctions.

In this paper we study label-correcting algorithms, which are routinely used to solve single-source shortest path problems. Several variants of label-correcting algorithms have been proposed in the literature, differing primarily in the strategy implemented to handle the candidate list of nodes. In particular, we study both one-list and two-list versions of the basic label-correcting algorithm; these variants implement either one or two lists to manage the set of candidate nodes. We examine the theoretical complexity and empirical behavior of these algorithms. In contrast to previous studies of shortest path algorithms, our focus is on explaining observed empirical performance in terms of certain intrinisic properties of the algorithms (namely, “sharpness” and “maturity”). In addition, a new variant of the label-correcting algorithm is proposed (PRED), which insures a type of “local sharpness” relative to the candidate list. Computational evidence suggests that this new algorithm, in both one-list and two-list versions, performs quite well in practice and deserves further study.