A Study for Real-Time Identification of Unsafe Behavior of Taking off Safety Helmet Based on VSM Model

Abstract

The unsafe behavior of workers is one of the main causes of injury accidents. However, there are two main management methods to control unsafe behavior: on the one hand, through safety education and training to improve the safety awareness of personnel, so that they reduce the occurrence of unsafe behavior; on the other hand, e.g. Behavior-Based Safety (BBS), safety managers supervise and manage workers' behavior to end the persistence of unsafe behavior. However, the former not only needs to invest a lot of money and time, but also has a slow effect and a long profit period. The latter also consumes large cost, and safety managers can not supervise all workers in real time. In order to solve the problems existing in the management of unsafe behavior and improve the safety protection effect of workers, taking the removal of safety helmet as an example, a vector space model based on computer modeling technology was proposed to identify workers' unsafe behavior in real time. With the low cost wearable experimental device, the key parameters and value ranges of the vector space model are obtained. The experimental results show that the model is feasible, real-time and effective, and can identify unsafe behaviors. This study is conducive to the real-time management of workers' unsafe behavior, and with the help of future related research, the wearable experimental device and the scope of this method can be improved, and the cost can be further reduced. Therefore, this study is conducive to enhancing the effectiveness of personnel safety management.