Bruce L. Golden and Douglas R. Shier, Editors-in-Chief of the journal Networks, are proud to announce the winners of the Glover-Klingman Prizes for the best paper published in Networks in 2019 and the best paper published in Networks in 2020. A list of past winners of the award can be found here.
The prize pays tribute to the high-quality work at the interface of operations research and computer science pioneered by Fred Glover and Darwin Klingman. However, the award recognizes outstanding work in the general area of network modeling, analysis, and implementation, and is not limited to papers at the OR/CS interface. There were 50 papers published in 2019 and 54 papers published in 2020. Both winning papers detail significant developments on the topic of constrained shortest paths and are, therefore, especially suitable in honoring the foundational work of Glover and Klingman.
The 2019 winners are Barrett W. Thomas, Tobia Calogiuri, and Mike Hewitt, authors of “An Exact Bidirectional A* Approach for Solving Resource-Constrained Shortest Path Problems.” Each author will receive a cash award and a certificate bearing the following citation:
This paper develops a new exact bidirectional dynamic programming algorithm for the solution of resource-constrained shortest path problems. Such problems arise in a wide variety of applications arising in transportation, telecommunications, location, and scheduling. The algorithm cleverly incorporates bidirectional dynamic programming and heuristic cost estimates for path extension. The reported computational experiments show the effectiveness and competitiveness of the proposed algorithm in solving large-sized instances.
The 2020 winners are Nicolás Cabrera, Andrés L. Medaglia, Leonardo Lozano, and Daniel Duque, authors of “An Exact Bidirectional Pulse Algorithm for the Constrained Shortest Path.” Each author will receive a cash award and a certificate bearing the following citation:
This paper presents an exact algorithm for resource-constrained shortest paths on directed networks that satisfy knapsack-type constraints on resource consumption. Such problems arise in a wide variety of settings, namely as subproblems in column generation algorithms. The algorithm performs a bidirectional search leveraged by parallelism. A so-called pulse-based heuristic also computes near-optimal solutions quickly. The algorithm was successfully tested on very large real-road networks with millions of nodes and arcs. The potential of this work for the network optimization community is significant.
The Editors-in-Chief express their genuine appreciation to Dr. James Kelly at OptTek Systems, Inc. (Boulder, Colorado) and Dr. Larry Levy at RouteSmart Technologies, Inc. (Columbia, Maryland) for their generous financial support. John Wiley & Sons has also contributed financial and administrative support to this project.