Machine learning predictions for lost time injuries in power transmission and distribution projects

• DOI: 10.1016/j.mlwa.2021.100158
• Publication Date: 2021-09-11T23:46:33Z
• Authors: Ahmed O. Oyedele; Anuoluwapo O. Ajayi; Lukumon O. Oyedele
• Keywords: Predictive Analytics

Although advanced machine learning algorithms are predominantly used for predicting outcomes in many fields, their utilisation incident outcome construction safety is still relatively new. This study harnesses Big Data with Deep Learning to develop a robust management system by analysing unstructured datasets consisting of 168,574 data points from power transmission and distribution projects delivered across the UK 2004 2016. compared performance popular (support vector machine, random forests, multivariate adaptive regression splines, generalised linear model, ensembles) concerning lost time injury risk assessment utility projects. gave best prediction high skills (AUC = 0.95, R2 0.88, multi-class ROC 0.93), thus outperforming other algorithms. The results this also highlight significance quantitative analysis empirical science contribute an enhanced understanding patterns using predictive analytics conjunction experts' perspectives. Additionally, will enhance managers domain advance intervention efforts.