Indonesian Journal of Electrical Engineering and Computer Science Vol. 23, No. 2, August 2021, pp. 1011~1017 ISSN: 2502-4752, DOI: 10.11591/ijeecs.v23.i2.pp1011-1017  1011 Journal homepage: http://ijeecs.iaescore.com A self adaptive new crossover operator to improve the efficiency of the genetic algorithm to find the shortest path Mrinmoyee Chattoraj, Udaya Rani Vinayakamurthy School of Computing & IT, REVA University, Bengaluru, India Article Info ABSTRACT Article history: Received Feb 15, 2021 Revised May 29, 2021 Accepted Jun 1, 2021 Route planning is an important part of road network. To select an optimized route several factors such as flow of traffic, speed limits of road. are concerned. Total cost of such a network depends on the number of junctions between the source and the destination. Due to the growth of the nodes in the network it becomes a tough job to determine the exact path using deterministic algorithms so in such cases genetic algorithms (GA) plays a vital role to find the optimized route. Crossover is an important operator in genetic algorithm. The efficiency of the genetic algorithm is directly influenced by the time of a crossover operation. In this paper a new crossover operator closest-node pairing crossover (CNPC) is recommended which is explicitly designed to improve the performance of the genetic algorithm compared to other well-known crossover operators such as point based crossover and order crossover. The distance aspect of the network problem has been exploited in this crossover operator. This proposed technique gives a better result compared to the other crossover operator with the fitness value of 0.0048. The CNPC operator gives better rate of convergence compared to the other crossover operators. Keywords: Chromosome representation Convergence Genetic algorithm Order crossover Point based crossover This is an open access article under the CC BY-SA license. Corresponding Author: Mrinmoyee Chattoraj School of Computing & IT REVA University Kattigenahalli, Yelahanka, Bengaluru, India Email: mrinmoyee2005@gmail.com 1. INTRODUCTION In the recent era, route optimization is gaining a lot of importance. There are various techniques to find the correct path. A lot of significance is given to genetic algorithms since it helps us to give an end-to- end optimized solution. In case of the current road network as the rate of traffic increases, the service quality also decreases. In case of genetic algorithms from individual search space is generated where a respective individual gives a specific solution. genetic algorithms (GAs) which was developed by Holland in 1992, simulated Darwin's evolution theory through natural selection by a particular type of bio-inspired approach. According to this theory there is maximum chances for the survival of the fittest organism. In the search space, genetic algorithm will explore all the solutions and the optimal solution will be retained. All individuals of a particular solution are encoded in the form of chromosomes. The important genetic operators such as crossover and mutation are applied to the parent chromosome to achieve better solutions with more potential. Crossover operator recombines the offspring’s and produces new chromosomes which are more enhanced than the parent chromosomes. To discover new states, mutation is often always needed, and it helps the genetic algorithm to escape local optima. These practises typically result in finding an optimal or near- optimal global solution to a given problem [1], [2]. There are various types of crossover operators which are