Novel and Effective Integer Optimization Approach for the NSF Panel-Assignment Problem: A Multiresource and Preference-Constrained Generalized Assignment Problem
Stacy L. Janak, Martin S. Taylor, Christodoulos A. Floudas, Maria K. Burka, Triantafillos J. Mountziaris · Industrial & Engineering Chemistry Research · 2005
The NSF panel-assignment problem studied in this work involves selecting an assignment of three or four reviewers to each proposal in a panel so as to optimize the sum of a set of preference criteria for each reviewer on each proposal while ensuring that each reviewer is assigned to approximately the same number of proposals. In addition, each proposal has three or four distinct positions that are assigned to reviewers based upon the preference criteria so that each reviewer holds each position approximately the same number of times. This multiresource and preference-constrained generalized assignment problem can be formulated as an integer linear programming problem and can be solved to optimality. In this work, a mathematical model is developed to address the NSF panel-assignment problem, and some representative example problems are solved to demonstrate the effectiveness of the proposed approach.