COAP 2008 best paper award: Paper of P.M. Hahn, B.-J. Kim, M. Guignard, J.M. Smith and Y.-R. Zhu

Since 2004, the Computational Optimization and Applications (COAP) editorial board has selected a paper from the preceding year’s COAP publications for the “Best Paper Award.” The award competition among papers published in 2008 culminated in a tie between two papers. This article concerns the award winning work of Peter Hahn, Bum-Jin Kim, Monique Guignard-Spielberg and Yi-Rong Zhu at the University of Pennsylvania and J. MacGregor Smith at the University of Massachusetts, Amherst, and their paper “An algorithm for the generalized quadratic assignment problem,” published in Volume 40, Issue 3, pages 351–372. The authors’ interest in the subject of the award paper began in December 2003, when they came across the website of Sourour Elloumi [4], which showed her work and that of her colleagues on various quadratic 0-1 assignment problems. What intrigued the authors was the similarity of these problems to the Quadratic Assignment Problem (QAP), on which they had been working for years. They had learned two important facts about the QAP. First, dual ascent algorithms based on making subtractions in the objective function cost matrices make very fast lower bound calculations [1, 6]. Second, researchers have over the years ignored the advantages afforded by combining complementary variables (i.e., variables that are equal to each other) in quadratic problems. So, the authors set about to work on these new problems as a challenge. Months later they came across the work by Lee and Ma [11], who coined the term Generalized Quadratic Assignment Problem (GQAP) and had been the first to provide an exact solution algorithm.