Monitoring land-centric business processes using remote sensing and satellite data

Process mining has been intensively used for business processes that are extensively supported by information systems. The tight integration of information processing and process execution, as leveraged in the service sector, is however often absent in land-centric processes such as farming. Land-centric processes exhibit some challenging characteristics that make it difficult to monitor them in real-time: they unfold continuously over time, yet with clearly identifiable states. In this paper, we address the challenge of monitoring land-centric processes. We introduce a framework to generate event logs of land-centric processes by utilizing remote sensing systems such as satellites. We demonstrate the feasibility of our approach using publicly available data on agricultural processes in the United States. • We introduce the class of land-centric processes and outline several examples from multiple domains. • We present a framework to retrieve process mining data from remote sensing data such as satellite images. • Our approach enables us to infer the timing of state changes of land-centric processes. • We evaluate our approach through a case study in agriculture using available data from the United States. • We show how process mining can be used to derive insights into land-centric processes.