Abstract
The practicality in recent process synthesis and process optimization of highly reliable systems leads to redundancy allocation problems and reliability assignment/redundant allocation problems. These optimization problems will be formulated as nonlinear integer programming (nIP) and nonlinear mixed integer programming (nMIP) models. In such reliability optimization, numerous exact methods and approximate methods have been proposed for the past two decades. However, they involve more computation efforts and usually require larger computer memory so that many researchers have placed more emphasis on the heuristic methods. From this viewpoint, genetic algorithm (GA) as one of the meta-heuristic methods, has been receiving much attention in this field. Unfortunately, GA cannot guarantee the optimal solution because of the fundamental requirements of not using a priori knowledge and not exploiting local search information. In order to overcome this kind of problem, in this paper, we propose a hybrid genetic algorithm combined with a fuzzy logic controller (FLC) and local search method. When applying the proposed method for the nMIP model, we also combine a neural network (NN) technique for devising of initial values for the GA. The efficacy and efficiency of the proposed algorithm is demonstrated by applying it to reliability optimization design problems of large-scale systems and by comparing its results with those of other traditional methods.
