Neural network embedded multiobjective genetic algorithm to solve non-linear time-cost tradeoff problems of project scheduling

Abstract

This paper presents a novel method to solve non-linear time-cost tradeof f (TCT) problem of real world engineering projects. Multiobjective genetic algorithm (MOGA) is employed to search for optimal TCT profile. Applicability of ANN based model for rapid estimation of time-cost relationship by invoking its function approximation capability is investigated. ANN models are then integrated with MOGA so as to develop a comprehensive approach to solve non-linear TCT problems of project scheduling. The study has implications in real time monitoring and control of project scheduling process.