Social and individual learning in the construction industry

Abstract

Labor is a critical resource for the success of any construction project. Evidence from the literature suggests that the productivity of workers in the construction industry increases with their experience. Therefore, it is vital to examine the venues by which these workers develop their experience. Given the absence of formal vocational training in this industry, workers develop their experience in two ways; individual and social. At the individual level, workers learn by repeating a certain task. On the other hand,at the social level, workers learn by socially interacting with one another. Learning curves (LC) are used to mathematically model individual learning, whereas social learning is studied under the umbrella of knowledge sharing (KS). Despite the abundance of the LC studies in the construction literature, there is little research conducted on how one could integrate the LC theory with project management techniques(E.g. scheduling) to accommodate for various project types. The volume of studies addressing KS on construction is limited and generally focuses on white-collar employees and the factors that affect their knowledge sharing behavior. The factors affecting knowledge sharing at the blue-collar level are different and tend to be related to salient properties such as demographics and therefore, should be studied further. Considering that the KS construction literature is still in its infancy, KS is not yet integrated with standard construction practices such as scheduling and crew formation. As such, this work aims to full fill the following three objectives: (i) Integrating LC theory with a scheduling tool that is compatible with various project types, (ii) Understanding the factors that affect KS among construction workers and in particular the effect of demographics, (iii) Integrating KS with crew formation at the level of the front line workers. In order to achieve these objectives, this research provides three contributions. The first contribution is an \dak[Automated Learning Curve Modeling