To search, Click below search items.


All Published Papers Search Service


Reducing Premature Convergence Problem through Numbers Structuring in Genetic Algorithm


Abu Bakar Md Sultan, Ramlan Mahmud, Muhammad Nasir Sulaiman


Vol. 7  No. 4  pp. 215-217


Genetic Algorithm (GA) has been widely used in many types of optimization problem. Premature convergence was the main problem for GA performance cause by lower diversity of the population. Obviously maintaining higher diversity is important to obtained better result. Classic GA representing the solution using binary system that is 0 and 1. From the origin idea of binary system, we present odd and even number representation for GA’s population. The odd and even number structure simulate the classic binary representation 0 and 1. The intention of proposed representation is to increase the diversity as well as preventing premature convergence. This short paper evaluates the performance of GA responding to new representation scheme.It was evaluated to solving timetabling problem and the result obtained is promising.


Genetic Algorithm, Premature Convergence, Diversity, Timetabling, Optimization