Abstract
Metaheuristics is one of the most well-known field of researches uses to find optimum solution for Non-deterministic polynomial hard problems (NP-Hard), that are difficult to find an optimal solution in a polynomial time. Over time many algorithms have been developed based on the heuristics to solve difficult real-life problems, this paper will introduce Metaheuristic-based algorithms and its classifications, Non-deterministic polynomial hard problems. It also will compare the performance two metaheuristic-based algorithms (Elephant Herding optimization algorithm and Tabu Search) to solve Traveling Salesman Problem (TSP), which is one of the most known problem belongs to Non-deterministic polynomial hard problem and widely used in the performance evaluations for different metaheuristics-based optimization algorithms. the experimental results of the paper compare the results of EHO and TS for solving 10 different problems from the TSPLIB95.
Highlights
Over the last years, complexity of problems has been increased so that it is very difficult for basic mathematical approaches to obtain an optimum solution in an optimal time [1]
This paper presents comparisons between two algorithms, namely Elephant herding optimizations (EHO) which belongs to the population-based metaheuristics algorithms and Tabu Search (TS) which belongs to the single-based metaheuristics algorithms, in solving one of the widely used non-deterministic polynomial hard (NP-hard) problems, i.e. the STSP
The results of simulations are compared with the optimal solutions from TSPLIB library
Summary
Complexity of problems has been increased so that it is very difficult for basic mathematical approaches to obtain an optimum solution in an optimal time [1]. Using approximate search algorithms is more preferable for most NP-hard problems, because it obtains a near-optimal solution in a significantly short time These algorithms are known as metaheuristic-based algorithms [3]. Metaheuristic algorithms characterize a master strategy that guides and modifies other heuristics to produce solutions beyond those that are normally generated in a quest for local optimality [10] These algorithms attempt to discover an optimal solution for the optimization of NP-hard problems in a polynomial time by constructing random modification and local-searches in the problem search space [4]. Local search metaheuristics discover optimal solutions by iteratively changing procedures from the current single solution These changes are called “Move” and could be regarded as walks through neighbourhoods or search trajectories of the search space of the problem.
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
