G. Anandalingam

Optimal pricing for multiple services in telecommunications networks offering quality-of-service guarantees

We consider pricing for multiple services offered over a single telecommunications network. Each service has quality-of-service (QoS) requirements that are guaranteed to users. Service classes may be defined by the type of service, such as voice, video, or data, as well as the origin and destination of the connection provided to the user. We formulate the optimal pricing problem as a nonlinear integer expected revenue optimization problem. We simultaneously solve for prices and the resource allocations necessary to provide connections with guaranteed QoS. We derive optimality conditions and a solution method for this class of problems, and apply to a realistic model of a multiservice communications network.

The Landscape of Electronic Market Design

This paper presents an introductory survey for this special issue of Management Science on electronic markets. We acquaint the reader with some fundamental concepts in the study of electronic market mechanisms, while simultaneously presenting a survey and summary of the essential literature in this area. Along the way, we position each of the papers presented in this special issue within the existing literature, demonstrating the deep impact of these 14 articles on an already broad body of knowledge.

Multi-level programming and conflict resolution

Abstract We examine conflict resolution problems by postulating the existence of an arbitrator who acts as the leader in a Stackelberg game. We present models for different configurations of the resulting multi-level linear programs, and provide new algorithms for solving them. We also examine the problems of coordinating such systems under informational decentralization. The models are illustrated with an application to a water conflict problem between India and Bangladesh which shows that both parties can gain by the arbitration of an international agency such as the United Nations.

A multi-stage multi-attribute decision model for project selection

Abstract In most multi-attribute decision models, much effort is directed at obtaining utility functions of the decision makers for the individual attributes that encode behaviour towards risk. A three-stage methodology is provided to simplify the process and to reduce cognitive stress for the decisionmaker. At the first stage, alternatives which are dominated and/or have significant technological uncertainties are screened out. At the second stage, concepts are modified from some psychological theories of decisionmaking to eliminate alternatives in which some or all attributes do not posses certain aspects. At the end of the first two stages, alternatives that remain have acceptable risk. At the final stage, value functions are used to obtain the preferred alternative under imprecise preference information. The methodology is applied to the real world problem of choosing a project for providing fresh water to the city of Newport News, Virginia, USA.

Iterative Combinatorial Auctions with Bidder-Determined Combinations

In combinatorial auctions, multiple distinct items are sold simultaneously and a bidder may place a single bid on a set (package) of distinct items. The determination of packages for bidding is a nontrivial task, and existing efficient formats require that bidders know the set of packages and/or their valuations. In this paper, we extend an efficient ascending combinatorial auction mechanism to use approximate single-item pricing. The single-item prices in each round are derived from a linear program that is constructed to reflect the current allocation of packages. Introduction of approximate single-item prices allows for endogenous bid determination where bidders can discover packages that were not included in the original bid set. Due to nonconvexities, single-item prices may not exist that are exact marginal values. We show that the use of approximate single-item prices with endogenous bidding always produces allocations that are at least as efficient as those from bidding with a fixed set of packages based on package pricing. A network resource allocation example is given that illustrates the benefits of our endogenous bidding mechanism.

A genetic algorithm approach to policy design for consequence minimization

Abstract We characterize the policy design problem as an event network where a series of interrelated decisions are made in a sequential fashion. As such, each node represents a decision point and determines the next arc to be connected in the network. A key objective in designing new policies is the minimization of negative outputs (consequence minimization). To address this type of problem we have applied a genetic algorithm (GA) to generate multiple network configurations (policy alternatives) for evaluation by human decision-makers. Our approach differs from typical genetic algorithms because decision-maker participation has been intimately linked to the genetic search so that policies designed will simultaneously meet the objectives for which they have been designed and remain acceptable and implementable in practice. In this paper we construct a solid waste management example to illustrate the usefulness of our approach to policy design.

Pricing strategies for information goods

Digital or information goods are becoming the norm across a wide variety of industries including books, music, entertainment, gaming and education. Due to the fact that the marginal cost of producing or reproducing information goods is very low, it is much easier to customise and personalise them for individual users. Furthermore, sellers of these information goods are increasingly using bundling and versioning strategies to appropriate a greater share of the surplus. This paper examines recent research on pricing of information goods with particular focus on customisation, bundling and versioning strategies adopted by information goods providers. The paper highlights both game-theoretic as well as optimisation models that not only provide different perspectives, but also examine issues of information goods pricing at different levels of abstraction and complexity.

Formulating and solving production planning problems

Production planning problems frequently involve the assignment of jobs or operations to machines. The simplest model of this problem is the well known assignment problem (AP). However, due to simplifying assumptions this model does not provide implementable solutions for many actual production planning problems. Extensions of the simple assignment model known as the generalized assignment problem (GAP) and the multi-resource generalized assignment problem (MRGAP) have been developed to overcome this difficulty. This paper presents an extension of the (MRGAP) to allow splitting individual batches across multiple machines, while considering the effect of setup times and setup costs. The extension is important for many actual production planning problems, including ones in the injection molding industry and in the metal cutting industry. We formulate models which are logical extensions of previous models which ignored batch splitting for the problem we address. We then give different formulations and suggest adaptations of a genetic algorithm (GA) and simulated annealing (SA). A systematic evaluation of these algorithms, as well as a Lagrangian relaxation (LR) approach, is presented.