An Optimization Based Heuristic for Political Districting
Management Science
Greedy, Prohibition, and Reactive Heuristics for Graph Partitioning
IEEE Transactions on Computers
Tabu Search
A simulated annealing approach to police district design
Computers and Operations Research - Location analysis
Congressional Redistricting Using A TSP-based Genetic Algorithm
GECCO '02 Proceedings of the Genetic and Evolutionary Computation Conference
Solving continuous location-districting problems with Voronoi diagrams
Computers and Operations Research
A computational approach to unbiased districting
Mathematical and Computer Modelling: An International Journal
Weighted Voronoi region algorithms for political districting
Mathematical and Computer Modelling: An International Journal
| Hi-index | 0.00 |
Redistricting is a complex and challenging spatial optimization problem. It is to group a set of spatial objects (such as counties) into a given number of geographically contiguous districts while satisfying multiple criteria and constraints such as equal population, compact shape, and more. The various criteria are often difficult to optimize and the number of potential solutions is very large. Moreover, many criteria are vaguely defined and may not be measured exactly. Therefore, human judgment and domain knowledge are indispensable and critical in the optimization process. In this paper, we present an interactive and computing-assisted approach to redistricting optimization. Our approach leverages the power of user's domain knowledge, judgment, and interactive exploration to (1) flexibly define various criteria/constraints, (2) visually and interactively examine alternative plans and achieve a balance among different criteria, and (3) efficiently and iteratively construct a collection of high-quality plans that are difficult to obtain with existing methods. A computational optimization algorithm is integrated to assist optimization under user-provided criteria and constraints. With the visual analytics approach, a user can quickly derive high-quality redistricting plans that satisfy both individual preferences and mandatory requirements. We demonstrate the capability of the approach and system with two case studies, Iowa congressional redistricting and South Carolina congressional redistricting.