Risk Factor Recognition for Automatic Safety Management in Construction Sites Using Fast Deep Convolutional Neural Networks

Identifying and Evaluating the Essential Factors Affecting the Incidence of Site Accidents Caused by Human Errors in Industrial Parks Construction Projects

In terms of safety management, the implementation of industrial parks construction projects (IPCPs) is incredibly challenging due to the special working conditions and the specific type of use of the buildings. On the other hand, the possibility of causing accidents in these areas based on human errors is high and important for project execution due to the risks of human errors and financial losses. Therefore, this study tries to fill this existing research gap by identifying and evaluating the effective key factors leading to the occurrence of construction accidents caused by human errors in the development of IPCPs. After a holistic review of the reported literature, four rounds of fuzzy Delphi survey were launched to capture the individual opinions and feedback from various project experts. Accordingly, 41 key factors affecting human errors in the implementation of industrial parks construction projects in Iran were identified and classified into nine main groups of wrong actions, observations/interpretations, planning/processes, equipment, organization, individual activities, environmental conditions, rescue, and technology. Then, the step-wise weight assessment ratio analysis (SWARA) method was adopted to rate and rank the identified factors of human errors in the implementation of IPCPs in Iran. The research findings indicated that among the elicited factors, time factor (0.1226), delayed interpretation (0.1080), and incorrect diagnosis/prediction (0.0990) are the three most crucial factors leading to human errors in the implementation of IPCPs in Iran. The results of this research study have provided various major project stakeholders with an effective decision-aid tool to make better-informed decisions in managing and reducing the occurrence of construction site accidents particularly caused by human errors associated with IPCPs.

Safety Risk Recognition Method Based on Abnormal Scenarios

Construction safety monitoring is a significant issue in practical engineering. Unfortunately, specific techniques in this field still heavily depend on artificial monitoring. To detect the abnormal scenarios during the construction process automatically, a method was proposed for the detection and localization of abnormal scenarios in time and space. The method consists of three components: (1) an I3D-AE video prediction model, which extracts the video features from multiple I3Ds and reconstructs the video by 3D deconvolution; (2) a spatial localization module AS-CAM, which determines the location of abnormal areas via back-propagating the I3D-AE; (3) a temporal parameter St, which can calculate the abnormal time period. The effectiveness of the method was verified with the use of a dataset, and the resulting data were plotted as ROC curves. The results indicated that the proposed method exceeded 0.9 on the frame-level test and 0.76 on the pixel-level test with the use of the AUC evaluation metric. Therefore, it can be used to assist the construction managers to improve the efficiency of construction safety management.