Comparison of evolutionary and swarm intelligence-based approaches in the improvement of peach fruit quality
2013
Kadrani , Abdeslam (INRA , Avignon (France). UR 1115 Unité de recherche Plantes et Systèmes de Culture Horticoles) | Ould Sidi , Mohamed Mahmoud (Institut National de la Recherche Agronomique, AVIGNON(France). UR1115 Plantes et systèmes de culture horticoles) | Quilot-Turion , Bénédicte (INRA , Montfavet (France). UR 1052 Génétique et Amélioration des Fruits et Légumes) | Génard , Michel (INRA , Avignon (France). UR 1115 Unité de recherche Plantes et Systèmes de Culture Horticoles) | Lescourret , Francoise (INRA , Avignon (France). UR 1115 Unité de recherche Plantes et Systèmes de Culture Horticoles)
We investigated two major families of algorithms for the multi-objective optimization: evolutionary andswarm intelligence-based optimization approaches. Non-dominated Sorting Genetic Algorithm II (NSGA-II) andmulti-objective particle swarm optimization (MOPSO) algorithms are biology inspired and are population-based asuse a set of solutions which evolve within the search space. These approaches employ different strategies andcomputational effort; therefore, a comparison of their performance is needed. This paper presents the applicationand performance comparison of NSGA-II and one variant of the MOPSO, namely MOPSO-CD which incorporatesthe crowding distance computation and the constraints handling, to design ideotypes for sustainable fruitproduction systems. The design of peach ideotypes that satisfy the requirement of high fruit quality and lowsensitivity to brown rot in a given environment was formulated as a multi-objective problem, and both NSGA-IIand MOPSO-CD are used to find the best combinations of genetic resources and cultural practices adapted to, andrespectful of specific environments. Statistically significant performance measures are employed to compare thetwo algorithms.
显示更多 [+] 显示较少 [-]