Genetic algorithm based optimization for resource leveling problem with precedence constrained scheduling

Abstract

Resource Leveling Problem (RLP) in construction projects mainly deals with reducing the peaks and valleys of daily resource requirements. RLP is an interesting field of research and several researchers have tried to solve it over the past few decades. Numerical methods and heuristics do not provide good results for solving Non-deterministic polynomial-hard (NP-hard) problems. So that, a metaheuristic approach is highly required to predict the solution with greater efficiency. This research work aims to solve RLP using Genetic Algorithm as the meta-heuristic approach, developed in a programming platform. Modified critical path scheduling method, was adopted to develop a schedule that determines the partial critical nature of activity with a start to start precedence relationship. Activities from a real-time construction project were considered to evaluate the proposed idea. The result obtained was significantly better than the results provided by existing commercial project management software and helps to solve RLP with the advantage of considering real-time floats.