DEVELOPING GENETIC LOCAL SEARCH ALGORITHM (GLS) SOFTWARE PROTOTYPE FOR QUADRATIC MINIMUM SPANNING TREE (q-MST) PROBLEM

• Main Author: Violina (NIM 235 04 040), Sriyani
• Format: Theses
• Language: Indonesia
• Online Access: https://digilib.itb.ac.id/gdl/view/9326
• Institution: Institut Teknologi Bandung

Abstract: <br /> <br /> <br /> <br /> <br /> Minimum Spanning Tree (MST) is the most important spanning tree and have a broad application in practice. The problem of finding the minimum spanning tree has known for over 100 years. Compared to conventional minimum spanning tree which only considered weight, the quadratic minimum spanning tree (q-MST) is a more realistic representation of the practical problem in the real-world by adding cost to be considered as an additional constraint. This research use the Genetic Local Search (GLS) method. GLS itself is the hybrid heuristic method combining Genetic Algorithm (GA) advanced in great scale search space and the Local Search that feed GA with local optima solution in order to hopely speed up the search. This GLS for q-MST software prototype is developed using object oriented software engineering and Delphie 6.0. as Tool.