Dynamic simulation models as digital twins of logistics systems driven by data from multiple sources

Abstract

Abstract In recent years, the development of manufacturing companies in the most dynamic industries, such as automotive, has been conditioned by the use of modern technologies, but also management concepts focused on quality and continuous improvement of processes and services. There was also a rapid increase in the level of production automation with a simultaneously growing level of flexibility in manufacturing systems, forced by the need to adapt the product to the customer’s needs (mass customization). These factors influenced the formation of a new concept for the development of industrial systems, referred to as Industry 4.0. It refers to production based on cyber-physical systems, integrating manufacturing, logistics, and distribution. One of the main components of an intelligent factory, based on cyber-physical systems, is the virtual equivalent of a real system in the form of the digital twin. Based on the feedback from the real system, a digital simulation supported by analytical software on the information platform is created. Dynamic simulation models should automatically adapt to changes in physical objects of the analysed system. Therefore, models of real objects reflecting their current state and logic of behaviour are required. The methods of automatic or semi-automatic generation of simulation models are the answer to these problems. They are based on approaches using data and information obtained from ERP/MES/CAD systems. In this paper, the proposed schemes for data exchange in the process of creation and simulation models and mapping of the dynamics of the actual system are shown.