Integration of digital twins into transport management systems

Abstract

The study aims to review the key functions required to build predictive maintenance models using digital twins (DTs) in order to serve as a catalyst for future primary research in this field. It also aims to fill the gap regarding DD in the automotive industry by presenting the current state of digital integration for freight, passenger, and autonomous vehicles, as well as assessing their maturity level using a maturity assessment tool. The results of this study are expected to provide a theoretical basis for understanding current trends in the methods used, as predictive maintenance and digital twins are rapidly evolving.
The paper also addresses the rapid growth in the volume of digital sensor data from machines and the availability of this data through the Internet of Things (IoT), which enables companies to make data-driven decisions. The potential of digital twins in assessing the response of a physical system to an unexpected event before it occurs is explored. It analyzes the latest trends in improving efficiency based on DD for freight, public, and autonomous vehicles. It discusses different types of digital twins (prototype, unique, aggregate) depending on the stage of product development and its DD, as well as their potential in the context of vehicles. Six levels of DD maturity are considered.
In addition, emphasis is placed on the integration of digital twins into transport management systems. Attention is also drawn to the fact that digital twins provide a real-time representation of the physical machine and generate data that can be used by a predictive maintenance algorithm. The use of digital twins for various applications is highlighted, from battery condition monitoring to ensuring fully autonomous vehicle operation.
A practical approach involves the implementation of digital twins in predictive maintenance, which is a strategic process that requires the integration of real-time data, advanced modeling tools, and machine learning models. This approach includes: data collection and integration, model creation and calibration, data analysis and predictive model development, integration with existing maintenance systems, and continuous monitoring and improvement.
The conclusions of the study confirm that digital twins create real-time virtual models of physical assets for scenario modeling, failure prediction, and maintenance schedule optimization. Predictive maintenance using digital twins reduces downtime and costs. Digital twin technology is revolutionizing the automotive industry by providing high-precision modeling, continuous monitoring, and sophisticated predictive analytics that impact all phases of the product lifecycle. The implementation of digital twins shortens development cycles, reduces the need for expensive physical prototypes, and supports the industry's transition to electric and autonomous vehicles.