Modern combine harvesters are complex systems that can be configured in various ways. Their performance is determined by a wide range of operating parameters. In AGATA, a collaborative project funded by the Federal Ministry of Education and Research, scientists and partners from the industry are investigating the optimal configuration of these parameters aiming to increase the efficiency of harvesting processes.
To this end, up to 250 messages per second with several hundred parameters - from the identification number to the tank level - are collected and stored. The data is gathered centrally and processed with big data analysis to find an optimized configuration of combine harvesters and improve their performance.
In order to prevent data misuse, early pseudonymization of the processed data is required. The position data of the machines is essential not only for the analysis, but it can also be used to create a precise action profile of the employees. This affects aspects of personal data protection and is therefore of decisive importance. Furthermore, data protection is significant from an operational point of view because some sensitive data, for example data concerning yield levels, are closely linked to the data used in the analysis.
The Fraunhofer IOSB has developed an innovative approach to ensure the protection of the employees´ privacy, while maintaining the operationality of the processed data. The data is pseudonymised in a multi-level process that obfuscates the real position. Data protected in this way shows how fast the harvesting maschine was on certain sections of a given field and, combined with the other data collected, allow a robust evaluation of the harvesting process. Due to the data pseudonymization, it is not possible to misuse the data, for example for monitoring of the employees or to draw conclusions about the harvest quantities of competitors. The goal of Big Data evaluation is achieved without employees being concerned about data misuse.