Subcontractor Selection using Genetic Algorithm

Abstract

In the construction industry, subcontracting is a very common practice. Nowadays, most of the general contractors tend to sublet the large portions of construction works to subcontractors and they only act as construction management agencies. In other words, while subcontractors carry out the actual production work, general contractors organize and coordinate the subcontractors and control their works in terms of time, cost and quality. In the construction industry, since the general contractors are responsible to the owners for the works carried out by the subcontractors, the general contractors should select the most appropriate subcontractors for the work packages that constitute the entire project. In the literature, there are a great number of studies that focus on subcontractor selection practices in the construction industry. These studies can mainly be categorized into two groups, which are; 1) the studies that aim to identify the subcontractor selection criteria and their importance levels, and 2) the studies that aim to propose tools, techniques and/or methodologies for subcontractor selection. The common point of the studies in the second category is that they all focus on selecting the most appropriate subcontractor for one work package in the project. For instance, if the case is the airport construction project, the construction of car park may be one work package. In this study, selection of subcontractors for the all work packages in a construction project is made using genetic algorithm technique considering time, cost and quality performances. A real-life construction project is selected as a case study and the actual data regarding alternative subcontractors are collected. The subcontractors for all work packages were selected successfully using the genetic algorithm technique.