The disciplines of computer science and operations research (OR) have been linked since their origins, each contributing to the dramatic advances of the other. This work explores the connections between these key technologies: how high-performance computing methods have led to advances in OR de ployment, and how OR has contributed to the design and development of ad vanced systems. The collected writings-from researchers and practitioners in Computer Science, Operations Research, Management Science, and Artificial Intelligence-were among those delivered at the Fifth INFORMS Computer Science Technical Section Conference in Dallas, Texas, January 8-10, 1996. The articles advance both theory and practice. Presented are new approaches to complex problems based on: metaheuristics (neural networks, genetic al gorithms, and Tabu Search), optimization and mathematical programming, stochastic methods, constraint programming, and logical analysis. These ad vanced methodologies are applied to new applications in such areas as: telecom munications network design, financial engineering, manufacturing, project man agement, and forecasting, airline and machine scheduling, vehicle routing, mod eling and decision support systems. Featured is a remarkable paper by keynote speaker Fred Glover, creator of the Tabu Search family of metaheuristics. In it he develops the principles of memory-based heuristic methods, contrasts them with the popular genetic algorithms and simulated annealing, provides a sweeping survey of application vignettes, and points to promising avenues for future research.

Klappentext

The disciplines of Computer Science and Operations Research have been linked since their origins and each have contributed to the dramatic advances of the other. This volume examines some of the recent advances resulting from the confluence between these two technical communities. In the process the book brings together the original work of academic researchers and practitioners in Computer Science, Operations Research, Management Science, Artificial Intelligence, Neural Networks, and related fields. More specifically, it explores the connections between these two areas with an array of advances in metaheuristics, neural networks, optimization, stochastic analysis, constraint logic programming, and decision support modeling. The book's principal theme centers on the shared modeling technologies of Computer Science and Operations Research and applies these methodologies to a variety of applications in manufacturing, logistics, finance, and telecommunications.

Inhalt
I Metaheuristics.- 1 Tabu Search and Adaptive Memory Programming Advances, Applications, and Challenges.- II Neural Networks.- 2 Neural Networks in Practice: Survey Results.- 3 Tractable Theories for the Synthesis of Neural Networks.- 4 Neural Network Training via Quadratic Programming.- 5 A Neural Network Model for Predicting Atlantic Hurricane Activity.- III Optimization.- 6 An Efficient Dual Simplex Optimizer for Generalized Networks.- 7 Solving Large Scale Crew Scheduling Problems.- IV Constraint and Logic Programming.- 8 Houria III: A Solver for Hierarchical Systems of Functional Constraints.- 9 Some Recent Developments of Using Logical Analysis for Inferring a Boolean Function with Few Clauses.- V Stochastic Performance Analysis.- 10 Computational Analysis of A G/G/1 Queue with Vacations and Exhaustive Service.- 11 Stability and Queuing-Time Analysis of a Reader-Writer Queue with Writer Preference.- 12 Importance Sampling in Lattice Pricing Models.- VI Modeling And Decision Support.- 13 Data and Optimisation Modelling: A Tool for Elicitation And Browsing (Dome).- 14 Enhancing User Understanding via Model Analysis in a Decision Support System.- VII Applications In Manufacturing, Logistics, And Finance.- 15 Bank Failure Prediction Using Dea to Measure Management Quality.- 16 A Cooperative Multi-Agent Approach to Constrained Project Scheduling.- 17 Scheduling a Flow Shop to Minimize the Maximal Lateness Under Arbitrary Precedence Constraints.- 18 A Genetic Programming Approach for Heuristic Selection in Constrained Project Scheduling.- 19 Coupling A Greedy Route Construction Heuristic With A Genetic Algorithm For The Vehicle Routing Problem With Time Windows.