Evaluation of hybrid monte-carlo and genetic algorithm for tropical timber joint strength

• Published in: Journal of Engineering and Applied Sciences
• Main Author: Yusoff M.; Othman D.F.; Latiman A.T.; Hassan R.
• Format: Article
• Language: English
• Published: Medwell Journals 2016
• Online Access: https://www.scopus.com/inward/record.uri?eid=2-s2.0-85006993424&doi=10.3923%2fjeasci.2016.1682.1686&partnerID=40&md5=6cdfd3aff9989d46b39b071c0005684c

The timber strength is one of the prime important aspects of timber structure design. A lot of laboratory experiments have been conducted to determine an appropriate load for timber strength. This paper addresses a new design of a hybrid genetic algorithm-Monte Carlo for load prediction in timber joint. A hybrid of Genetic Algorithm Monte-Carlo is employed to determine the best load value for the prediction of timber joint strength. This study discusses the initial solution to overcome the time consuming and costly incurred of the laboratory experiments. A new solution representation of Genetic Algorithm was addressed with the introduction of Monte-Carlo calculation. Two types of tropical timbers which are Keruing and Sesenduk are used. The results demonstrate faster solution due to fast convergence of obtaining a feasible solution. At the same time the hybrid solution also gives a sub-optimal solution. However, more computational experiments are expected to be done for various types of timbers. The comparison with other computational methods and different parameters should be considered to find better solutions. © Medwell Journals, 2016.