Modelling and executing IoT-enhanced business processes through BPMN and microservices

Abstract

[EN] The Internet of Things enables to connect the physical world to digital business processes (BP) and allows a BP to (1) consider real-world data to take more informed business decisions, (2) automate and/or improve BP tasks, and (3) adapt itself to the physical execution environment. We refer to these processes as IoT-enhanced BPs. Although numerous researchers have studied this subject, there are still some challenges to be faced. For instance, the need of a modelling solution that does not increase the notation complexity to facilitate further analysis and engineering decision making, or an execution approach that provides a high degree of independence between the process and the underlying IoT device technology. The objective of this work is defining an approach that (1) considers important intrinsic characteristics of IoT-enhanced BPs at modelling level without growing the complexity of the modelling language, and (2) facilitates the execution of the IoT-enhanced BPs represented in models independently from IoT devices' technology. To do so, we present a modelling approach that uses standard BPMN concepts to model IoT-enhanced BPs without modifying its metamodel. It applies the Separation of Concern (SoC) design principle: BPMN is used to describe IoT-enhanced BPs while low-level real-world data is captured in an ontology. Finally, a microservice architecture is proposed to execute BPMN models and facilitate its integration with the physical world. This architecture provides high flexibility to support multiples IoT device technologies as well as their evolution and maintenance. The evaluation done allows us to conclude that the application of the SoC principle using BPMN and ontologies facilitates the definition of intrinsic characteristics of IoT-enhanced BPs without increasing the complexity of the BPMN metamodel. Besides, the proposed microservice architecture provides a high degree of decoupling between the created models and the underlying IoT technology. (C) 2021 Published by Elsevier Inc.