Sorting things out? Machine learning in complex construction projects

May Shayboun, Christian Koch
Chalmers University of Technology, Sweden

DOI: 10.35490/EC3.2019.161
Abstract: This research includes answers from 324 main contractor representatives and 256 clients for a survey in Sweden, 2014. The literature review covers project management success in construction projects. A statistical correlation method is used to select the features that are strongly correlated with three performance indicators: cost variance, time variance and client- and contractor satisfaction. A linear regression prediction model is presented. The conclusion is an identification of the most correlating factors to project performance, and that human related factors in the project life cycle have higher impact on project success than the external factors and technical aspects of buildings.
Keywords: Machine Learning. Productivity, Building Projects, Sweden
Pages: 65 - 74
Paper:
http://ec-3.org/conf2019/contribution_161_final/

Presentation Video: