A Modified Artificial Bee Colony Algorithm for Solving Least-Cost Path Problem in Raster GIS

The computation of least-cost paths over a cost surface is a well-known and widely used capability of raster geographic information systems (GISs). It consists of finding the path with the lowest accumulated cost between two locations in a raster model of a cost surface. This paper presents a modified Artificial Bee Colony (ABC) algorithm for solving least-cost path problem in a raster-based GIS. This modification includes the incorporation of a distinct feature which is not present in the classical ABC. A new component, the direction guidance search method, is used to guide a bee walking toward the final destination more efficiently. In addition, this paper examines how the quality of the raster-based paths can be improved by using larger connectivity patterns. The experimental results show that the performance of the modified ABC model is quite close to Dijkstra’s algorithm while its computational complexity and solution time is much lower than Dijkstra’s algorithm. The results also, indicate that raster-based paths can be improved by using larger connectivity patterns