With the availability of connectivity solutions and a plethora of sensors that were previously unavailable on mobile machines, there exists a wealth of data that has not yet been harnessed to its full potential. Besides use cases like the automation of machines, condition monitoring, and fault detection, knowledge of process data can help in improving the efficiency in the use of machinery. Excavators have a wide range of applications and therefore are of special interest. The research presented in this paper contributes to a deeper understanding of the utilization of a hydraulic excavator in road construction which can be used for optimizing the duty cycle efficiency. For that purpose, a digital twin of an excavator is created and used to classify its various tasks and operation modes. The integration of digital twin technology in construction processes offers a promising pathway for enhancing productivity and decision-making within the industry. A concept for classifying duty cycles of hydraulic excavators in the context of road construction, with a particular emphasis on leveraging the concept of a digital twin for the excavator, is introduced in this paper. During a conducted study, machine data from several sensors is gathered. The data used in this study is obtained from a real-world construction site with various applications of the examined machine. The collected sensor data not only facilitates the classification of these tasks but also offers opportunities for enhancing automation and implementing smart metering functions within the machines. In addition, conclusions will be drawn about the impact of different concepts of classifying duty cycles on the construction site.