Data mining for forecasting labor resource requirements: a case study of project management staffing requirements

Abstract

Accurate labor resource allocation ensures that tasks are assigned to the most suitable individual(s) and the optimal number of staff are available to complete certain tasks, making it essential for the success of construction projects. This study proposes a methodology for forecasting labor resource requirements for upcoming projects using data mining techniques. The framework consists of two components, a data acquisition model and a forecasting model. The data acquisition model provides a structured approach for tracking and storing project data, while the forecasting model uses the stored project data and applies machine learning algorithms to predict labor requirements. A case study is used to illustrate the application of the framework in actual projects. The results indicate the usefulness of machine learning in providing low error estimates compared to methods currently adopted in the industry. The results also revealed that forecasting accuracy considerably increases as the number of historical projects increases which signifies the importance of the data acquisition model. The models presented in this study are expected to help guide and promote the application of more accurate workforce forecasting techniques.