Read the full-text of this article for free

Title: IoT-based construction site safety management: real-time monitoring and early warning system construction

Authors: Zhengkun Li; Fei Ye; Qiaozhen Liang

Addresses: Department of Food Engineering, Anhui Vocational College of Grain Engineering (AVCGE), Hefei, 230011, China ' Department of Food Engineering, Anhui Vocational College of Grain Engineering (AVCGE), Hefei, 230011, China ' Department of Food Engineering, Anhui Vocational College of Grain Engineering (AVCGE), Hefei, 230011, China

Abstract: For the goal of ensuring the smooth progress of construction, it is urgent to design a real-time construction safety management method. First, the overall architecture of internet of things (IoT) real-time monitoring is constructed, which includes sensing layer, network layer, platform layer, and application layer; second, according to the accident causation theory, the construction risk monitoring index system is determined, and the key risk features are extracted. Subsequently, the improved ReliefF algorithm is used to select important construction risk features, and the hyperparameters of the support vector machine (SVM) are optimised by the particle swarm optimisation (PSO) algorithm, and important risk features are inputted into the PSO-SVM model to obtain final risk warning results. Application results of a construction project show that the data transmission delay of the system is less than 0.2 s, and the monitoring accuracy can reach 91.31%, showing excellent real-time and accuracy.

Keywords: construction site safety management; monitoring and early warning; internet of things; IoT; feature selection; support vector machine; SVM; particle swarm optimisation; PSO.

DOI: 10.1504/IJICT.2025.146835

International Journal of Information and Communication Technology, 2025 Vol.26 No.21, pp.55 - 69

Received: 15 Apr 2025
Accepted: 29 Apr 2025
Published online: 20 Jun 2025 *