Influence of the Competencies of Lifting Crane Specialists on the Probability of Emergencies

Abstract

Introduction. The operation of lifting cranes is an integral part of the production processes. For the trouble-free operation of these mechanisms, certain knowledge, skills and abilities are required, which should also be possessed by specialists performing organizational and supervisory functions at facilities where such cranes are involved. Here there is an important problem – the lack of a reasonable connection between the level of development of professional competencies and possible emergency situations, as well as various incidents during the operation of lifting cranes. The authors of this study are trying to solve it. Their goal in this regard is to assess the probability of an emergency during the operation of lifting cranes, depending on the level of professional competence of specialists, through the use of neural networks.Materials and Methods. The competencies of workers in the operation of lifting cranes (knowledge, skills and work responsibilities) provided for by the professional standard «Specialist in the operation of lifting structures» were used as initial data to train neural networks. Based on them, a list of possible incidents was compiled. For the purposes of training, the results of the certification of 200 conditional employees were generated. During the generation, the Monte Carlo method was used, and the data were output to Excel tables. Neural networks were trained in Python 3.10 in the PyCharm development environment. Open libraries Keras and TensorFlow, as well as auxiliary libraries for data representation and processing (Pandas, NumPy, Scikit-learn) were used for neural networks training.Results. As a result, a tool was obtained – a neural network in the form of executable program code, which makes it possible to assess the probability of emergencies during the operation of lifting cranes by analyzing the degree of proficiency of specialists in professional competencies. It is proposed to implement artificial intelligence technologies based on neural networks in order to assess the knowledge, skills and abilities of specialists of facilities operating lifting cranes, both during the certification of employees and in the course of work.Discussion and Conclusion. The main result of using neural networks to assess the knowledge of employees of facilities operating lifting cranes is the expected reduction in accidents, which can be ensured by timely identification of incompetent personnel at the stages of primary certification and, most importantly, during periodic tests of knowledge based on an impartial analysis and evaluation of data.