A Graph Pattern Based Approach for Automatic Decomposition of IoT Aware Business Processes

摘要

The context information that business process can get from the Internet of Things (IoT) can be used as a competitive advantage in terms of optimisation and agility. However, the exchange of messages between central systems and IoT devices come with a price, battery consumption, a scarcely resource of such devices. Despite the literature offers many technical proposals to tackle this problem, we take an approach driven by the process definition perspective. We propose to reduce the number of exchanged messages by decentralising process execution, moving parts of the business processes to IoT devices, and taking advantage of their computational capabilities. The first step for decentralisation is decomposition, i.e., the division of processes into parts and identify those that IoT devices can execute. In this paper, we present an automatic decomposition solution for IoT aware business processes, described using the Business Process Model and Notation (BPMN). We start from a BPMN definition that follows a centralised approach and apply our decomposition method to transfer to the IoT devices the operations that can be performed there. We use a graph based approach and transform a BPMN definition into a directed graph. Thereafter, we identify cuts that define the parts to be transferred to the IoT devices. This decomposition preserves the control and the data dependencies of the original process, reduces the number of exchanged messages as well as the central processing. The code that IoT devices execute is automatically generated from the BPMN process being decentralised.