Your browser version is outdated. We recommend that you update your browser to the latest version.

Manuscript Title: Sieve Algorithm - A New Method for Optimization Problems

Author : Hemmak Allaoua, Bouderah Brahim

Email :hem_all@yahoo.fr,bouderah_b@yahoo.fr

Abstract: 

In this paper, we propose a new approach for solving combinatorial optimization problems as scheduling problems,  traveling salesman problem, transport problems, and images segmentation which have exponential complexity and known as NP-hard problems. This approach, known as Sieve approach is based on the sieving operation idea used to sift grains by translating it on an algorithmic tool. This approach generates randomly and iteratively a great number of feasible solutions by batches. The bad items are removed, while the good items are assembled in a smaller central set according to an appropriated fitness function. The best solution is computed from this small set according to the problem objective. The fitness function may be easy to compute but it may simulate the objective. We provide a mathematical formulation for representing the problem environment. The implementation is done on a single machine scheduling problem and the results are compared to show the approach efficiency.  The findings show that our proposed method can give better solutions than existing meta heuristics like genetic algorithms, ant colonies, and simulated annealing.

Keywords: Meta heuristics, Optimization, Combinatorial problems, Sieve method, Scheduling.

 Vol 5 (2)