Object-Oriented Evolutionary Fuzzy Neural Inference System for Construction Management

Abstract

Problems in construction management are complex, full of uncertainty, and vary with environment. Fuzzy logic, neural networks, and genetic algorithms (GAs) have been successfully applied in construction management to solve various kinds of problems. Considering the characteristics and merits of each method, this paper combines the above three techniques to develop an Evolutionary Fuzzy Neural Inference Model (EFNIM). Integrating these three methods, the EFNIM uses GAs to simultaneously search for the fittest membership functions with the minimum fuzzy neural network (FNN) structure and optimum parameters of FNN. Thus, the best adaptation mode is automatically identified. Furthermore, this research work integrates the EFNIM with an object-oriented (OO) computer technique to develop an OO Evolutionary Fuzzy Neural Inference System for solving construction management problems. Simulations are conducted to demonstrate the application potential of the EFNIS. This system could be used as a multifarious intelligent decision support system for decision-making to solve manifold construction management problems.