Performance of Artificial Bee Colony algorithm and its implementation on graph theory application course

Closed

Sapti Wahyuningsih, Darmawan Satyananda, Lucky Tri Oktoviana

2020 AIP Conference Proceedings Vol. 2215 Conference paper Cited by 1 Quartile

Abstract

Graph Theory Application is a course in the Mathematic Department which concerned with daily application. One of its application is a distribution problem, that is defining the minimum length route of one (or more) vehicle to distribute goods from one (or more) depot to some customers. In this distribution system, choosing the route is the most important in determining total length, travel time, and cost. The distribution problem is modeled into the Vehicle Routing Problem (VRP). Among variants of VRP, one of them is Capacitated Vehicle Routing Problem with Time Window (CVRPTW) with its main objective is to minimize total length of route and number of vehicles concerning vehicle capacity and time limit given. The Problem of CVRPTW can be solved with Artificial Bee Colony (ABC) algorithm, and the subsequent, namely Improved ABC (IABC) and Modified ABC (MABC) algorithm. IABC is a refinement of the standard ABC algorithm by adding Partial-Mapped Crossover (PMX) parameter in the refinement stage. Meanwhile, MABC is a modification of ABC by adding a parameter in the stage of initialization, refinement of the solution by using neighborhood structures, and optimization by keeping unused solution in tabu list. In performance analysis, MABC produced a better final solution than IABC dan ABC, with regard to total route length and its service time. In practice, the application can be used by students doing field survey in Graph Theory Application course and field practice (KPL) to optimize distribution problem. © 2020 Author(s).

Affiliations

Universitas Negeri Malang, Jl. Semarang 5, Malang, Indonesia