N. Shahsavari Pour; M.H. Abolhasani Ashkezari; H. Mohammadi Andargoli
Volume 2, Issue 1 , February 2013, , Pages 20-29
Abstract
Considering flow shop scheduling problem with more objectives, will help to make it more practical. For this purpose, we have intended both the makespan and total due date cost simultaneously. Total due date cost is included the sum of earliness and tardiness cost. In order to solve this problem, a genetic ...
Read More
Considering flow shop scheduling problem with more objectives, will help to make it more practical. For this purpose, we have intended both the makespan and total due date cost simultaneously. Total due date cost is included the sum of earliness and tardiness cost. In order to solve this problem, a genetic algorithm is developed. In this GA algorithm, to further explore in solution space a Tabu Search algorithm is used. Also in selecting the new population, is used the concept of elitism to increase the chance of choosing the best sequence. To evaluate the performance of this algorithm and performing the experiments, it is coded in VBA. Experiments results and comparison with GA is indicated the high potential of this algorithm in solving the multi-objective problems.
H. Mohammadi-Andargoli; R. Tavakkoli-Moghaddam; N. Shahsavari Pour; M.H. Abolhasani-Ashkezari
Volume 1, Issue 2 , July 2012, , Pages 10-26
Abstract
This paper addresses the permutation of a flexible job shop problem that minimizes the makespan and total idleness as a bi-objective problem. This optimization problem is an NP-hard one because a large solution space allocated to it. We use a duplicate genetic algorithm (DGA) to solve the problem, which ...
Read More
This paper addresses the permutation of a flexible job shop problem that minimizes the makespan and total idleness as a bi-objective problem. This optimization problem is an NP-hard one because a large solution space allocated to it. We use a duplicate genetic algorithm (DGA) to solve the problem, which is developed a genetic algorithm procedure. Since the proposed DGA is working based on the GA, it often offers a better solution than the standard GA because it includes the rational and appropriate justification. The proposed DGA is used the useful features and concepts of elitism and local search, simultaneously. It provides local search for the best solution in every generation with the neighborhood structure in several stages and stores them in an external list for reuse as a secondary population of the GA. The performance of the proposed GA is evaluated by a number of numerical experiments. By comparing the results of the DGA other algorithms, we realize that our proposed DGA is efficient and appropriate for solving the given problem.