Discrete Gravitational Search Algorithm (DGSA) applied for the Close-Enough Travelling Salesman Problem (TSP / CETSP)

Abstract

TSP (travelling salesman problem) is a NPhard problem, and several exact and heuristics solutions exist. Exact solutions consume too many resources (computation and time) and heuristic solutions do not provide global optimum path. We propose another heuristic variation of the TSP and CETSP solver (DGSA-TSP and DGSA-CETSP), based on discrete gravitational search algorithm and two novel rubber-band algorithm (RBA) implementations; we benchmark it against the GSOA (growing self-organizing array) variant and obtain similaraccuracy results.