ИСПОЛЬЗОВАНИЕ ИСКУССТВЕННЫХ НЕЙРОННЫХ СЕТЕЙ ДЛЯ МОНИТОРИНГА РАБОТЫ ПРОИЗВОДСТВЕННОЙ ЗОНЫ АВТОСЕРВИСА

Abstract

The article deals with the issues of monitoring the operation of the zone for the maintenance and repair of vehicles at a service station using information technology. The production facilities of many stations are equipped with video surveillance cameras that could be used to monitor the operation of the service area, in particular, to determine the degree of loading of the post, its condition at a particular point in time, the fact that a mechanic and a car are at the post, as well as determining the routes of movement of production personnel. The information obtained can be integrated into the enterprise database and used to make management decisions and generate zone operation statistics. The implementation of the system is possible using a pre-trained ResNet50 convolutional neural network, which is used to detect objects «car» and «person». The developed model can be part of a management decision support system or a digital twin of a motor transport company.