An efficient parallel genetic algorithm solution for vehicle routing problem in cloud implementation of the intelligent transportation systems

Mahdi Abbasi,Varun G. Menon,Alireza Jolfaei,Milad Rafiee,Javad Mokhtari Koushyar,Mohammad R. Khosravi

doi:10.1186/s13677-020-0157-4

Abstract

A novel parallelization method of genetic algorithm (GA) solution of the Traveling Salesman Problem (TSP) is presented. The proposed method can considerably accelerate the solution of the equivalent TSP of many complex vehicle routing problems (VRPs) in the cloud implementation of intelligent transportation systems. The solution provides routing information besides all the services required by the autonomous vehicles in vehicular clouds. GA is considered as an important class of evolutionary algorithms that can solve optimization problems in growing intelligent transport systems. But, to meet time criteria in time-constrained problems of intelligent transportation systems like routing and controlling the autonomous vehicles, a highly parallelizable GA is needed. The proposed method parallelizes the GA by designing three concurrent kernels, each of which running some dependent effective operators of GA. It can be straightforwardly adapted to run on many-core and multi-core processors. To best use the valuable resources of such processors in parallel execution of the GA, threads that run any of the triple kernels are synchronized by a low-cost switching mechanism. The proposed method was experimented for parallelizing a GA-based solution of TSP over multi-core and many-core systems. The results confirm the efficiency of the proposed method for parallelizing GAs on many-core as well as on multi-core systems.

Highlights

Vehicle routing problems have been the focus of extensive research over the past 60 years, driven by their economic importance and their theoretical interest
Using efficiently parallelizable optimization algorithms for solving equivalent Traveling Salesman Problem (TSP) of vehicle routing problems is a key in minimizing the costs of any intelligent transportation system with limited profitability margins
In this paper we presented an enhanced genetic algorithm (GA) solution to TSP problem in Vehicle Routing Problem (VRP) which could be and highly parallelized on multi-core and many-core machines of suitable vehicular cloud computing (VCC) platforms

Summary

Introduction

Vehicle routing problems have been the focus of extensive research over the past 60 years, driven by their economic importance and their theoretical interest. By increasing the problem size, the size and number of genes increases and, the necessary computation tasks increase They propose a parallel implementation for their proposed GA which can be executed on GPU-like many-core machines, experimental results show the inefficiency of this method in achieving a high level of parallelism on GPUs. Other researches like [21,22,23,24] have obscurely implemented parallel kernels for GA-based solutions of TSP. We investigate the effect of different parameters of GA-based solutions of TSP on the performance of parallel kernel on both multi-core systems as well as many-core systems For this purpose, we provide parallel kernels for multi-core systems using TBB and for many-core GPUs using CUDA. The results of running this kernels with identical GA settings are used to provide an in-depth comparison of the efficiency of the kernels as well as to assess the effect of each GA parameter on the overall efficiency of each kernel

Background

Findings

Conclusion