Abstract:
Construction companies should schedule their projects in a manner that considers theefficient use of limited resources in order to complete a project within estimated budget, onschedule and in compliance with the specifications. In this context, the planning of resourcesbecomes crucial for a construction project, which can be accomplished by resource leveling.Resource leveling - also known as resource smoothing - is a method that attempts to reducethe fluctuations in resource usage in order to make the resource requirements as uniform aspossible while maintaining the original project duration. The studies dealing with resource leveling problems can be classified into two categories, which are; (1) analytical methods and(2) heuristic methods. Analytical methods may give optimal solutions on small-scaledproblems; however, they are inadequate in large-scaled problems. As a result of theweaknesses of analytical methods, many studies have been conducted in order to developmore efficient models by heuristic methods. Genetic algorithm-based resource leveling is oneof these models, which is developed to attain better solutions. The main objective of thisstudy is to handle the resource leveling problem of an industrial building using geneticalgorithms. In this context, a schedule for an industrial building is established using theCritical Path Method (CPM). The information about the logical constraints and the resourcesrequired to carry out activities were obtained through the interviews with civil engineers fromthe company, whose expertise is on industrial buildings. The proposed genetic algorithmbased resource leveling model attempts to improve the schedule. The developed modelprovided a decrease of 20% in the total resource-days required to complete the project. Thestudy is of benefit to participants of construction industry, because it makes them aware of the potential use of the combination of critical path method and genetic algorithms in order to solve the resource leveling problem.
