Big data analytics system for costing power transmission projects

Inaccurate cost estimates have significant impacts on the final cost of power transmission projects and erode profits. Methods for cost estimation have been investigated thoroughly, but they are not used widely in practice. The purpose of this study is to leverage a big data architecture, to manage the large and diverse data required for predictive analytics. This paper presents a predictive analytics and modeling system (PAMS) that facilitates the use of different data-driven cost prediction methods. A 2.75-million-point dataset of power transmission projects has been used as a case study. The proposed big data architecture fits this purpose. It can handle the diverse datasets used in the construction sector. The three most prevalent cost estimation models were implemented (linear regression, support vector regression, and artificial neural networks). All models performed better than the estimated human-level performance. The primary contribution of this study to the body of knowledge is an empirical indication that data-driven methods analysed in this study are on average 13.5% better than manual methods for cost estimation of power transmission projects. Additionally, the paper presents a big data architecture that can manage and process large varied datasets and seamless scalability.