Fall accidents in industrial and construction environments require an immediate reaction, to provide first aid. Shortening the time between the fall and the relevant personnel being notified can significantly improve the safety and health of workers. Therefore, in this work, an IoT system for real-time fall detection is proposed, using the ASSIST-IoT reference architecture. Empowered with a machine learning model, the system can detect fall accidents and swiftly notify the occupational health and safety manager. To train the model, a novel multimodal fall detection dataset was collected from ten human participants and an anthropomorphic dummy, covering multiple types of fall, including falls from a height. The dataset includes absolute location and acceleration measurements from several IoT devices. Furthermore, a lightweight long short-term memory model is proposed for fall detection, capable of operating in an IoT environment with limited network bandwidth and hardware resources. The accuracy and F1-score of the model on the collected dataset were shown to exceed 0.95 and 0.9, respectively. The collected multimodal dataset was published under an open license, to facilitate future research on fall detection methods in occupational health and safety.

Lanyards are still important and common components of personal systems protecting against falls from a height. Experience with dynamic strength tests of lanyards indicates that test methods based on EN and ISO standards do not make objective assessment possible. This paper presents the results of theoretical and laboratory investigations of the performance of adjustable lanyards during fall arrest. The obtained results indicate that methods of testing and assessment based on those standards demonstrate considerable shortcomings when applied to low elongation lanyards. The assumptions for improved requirements and test methods of lanyards made of e.g., steel wire and aramid ropes are also presented.

Designing fall arrest systems, which contain horizontal flexible anchor lines is an important technical problem related to the safety of people who work at a height and need horizontal freedom of movement. The article presents a numerical simulation of the dynamic performance of horizontal flexible anchor lines during fall arrest. The model of a 2-component system a horizontal flexible anchor line and a falling rigid mass, described with a second order non-linear differential equation is the main element of this method. This method of simulation is realised by a computer program, which allows obtaining the most important data characterising a fall arrest. The article shows laboratory tests used for the verification of this method, which turned out to be a valuable source of information and which can be used for designing fall arrest systems.

The use of personal protective equipment, including a safety harness, is one of the basic methods of protection against falls from a height. The presented studies, using an anthropomorphic dummy, identified the effect on the human body of the dangerous phenomena accompanying safety harness performance during fall arrest. The displacement of the dummy in the safety harness, the mutual displacement of the adjustment buckles and the webbing of the harness, the tightening of the straps on the dummy and the impacts exerted on the head of the dummy by the harness elements were considered. The correlation between the design of the safety harness and the parameters and phenomena in question has been demonstrated. It has been shown that for the purposes of assessment of novel harness designs it is necessary to carry out studies utilizing an anthropomorphic dummy in addition to resistance tests.