Artificial Intelligent Systems and Machine Learning

Nitesh M. Sureja ¹, Ved Vyas Dwivedi ², Sanjeev Kumar ³

Affiliations
1 NIMS Institute of Engineering and Technology, NIMS University, Jaipur, Rajasthan, IN
2 Noble Group of Institutions, Junagadh, Gujarat, IN
3 Department of Information Technology, NIMS Institute of Engineering and Technology, NIMS University, Jaipur, Rajasthan, IN

Abstract

The Memetic algorithms (MA) are computational models inspired by the Cultural Revolution. Robustness, scalability and simplicity are the advantages of Memetic algorithms. By looking at this strength of MA, they are very much suitable for solving some combinatorial optimization problems which falls in the category of NP-hard problems. In this paper, we propose MA based optimization model for Dynamic Traveling Salesman Problem (DTSP), to achieve overall higher system performance and better throughput.

During implementation, results are obtained with number of randomly generated Traveling Salesman Problem instances, measured in terms of required time, quality and path length. It has been observed that the proposed model works well in terms of quality and path length. As it uses Simulated Annealing, it is found little bit less effective in terms of time.

Keywords

Memetic Algorithms, Simulated Annealing, Dynamic Traveling Salesman Problem (DTSP), Optimization Problems.

Full Text

     

Nitesh M. Sureja ¹, Ved Vyas Dwivedi ², Sanjeev Kumar ³

Affiliations
1 NIMS Institute of Engineering and Technology, NIMS University, Jaipur, Rajasthan, IN
2 Noble Group of Institutions, Junagadh, Gujarat, IN
3 Department of Information Technology, NIMS Institute of Engineering and Technology, NIMS University, Jaipur, Rajasthan, IN

Abstract

Ant Colony Optimization (ACO) is a novel and competitive optimization method for numerous combinatorial optimization problems. It is already applied to various optimization problems. Normally it proved best in terms of solution quality, accuracy and other parameters. This paper presents an Elitist Ant System (EAS) which uses tuning in pheromone update process to improve the performance of the basic Ant System (AS) approach. Dynamic Traveling Salesman Problem is solved in this research with the various pheromone update strategy for finding the improvements in the results. The results obtained are empirically compared with the results obtained with the basic pheromone update strategy of Ant System.

Keywords

Ant Colony Optimization, ElitistAnt System, Dynamic Traveling Salesman Problem (DTSP), Optimization.

Full Text

     

Nitesh M. Sureja ¹, Ved Vyas Dwivedi ², Sanjeev Kumar ³

Affiliations
1 NIMS Institute of Engineering and Technology, Nims University, Jaipur, Rajasthan, IN
2 Noble Group of Institutions, Junagadh, Gujarat, IN
3 Department of Information Technology, NIMS Institute of Engineering and Technology, Nims University, Jaipur, Rajasthan, IN

Abstract

Cuckoo Search is a recently developed, novel and competitive Nature Inspired Optimization algorithm used for solving various linear and discrete combinatorial optimization problems. It is already applied to the problems like PCB Drilling, finding optimal features, optimizing the parameters of various classifiers including Neural Network, RBF, SVM parameters, finding optimizing cluster centers, job scheduling, find optimal path and many more in the different domains including industry, health sector, wireless sensor network, image processing and others. Due to its proven capacity in solution quality, accuracy and other parameters, in this paper we present a cuckoo search based model to solve Dynamic Traveling Salesman Problem. The results obtained are compared with the results obtained with the other models proposed for the same problem. Results obtained are very good in terms of all related parameters.

Keywords

Cuckoo Search Algorithm, Combinatorial Optimization, Dynamic Traveling Salesman Problem, Nature Inspired Algorithms, Random Walk.

Full Text