IIoT System Canvas — From architecture patterns towards an IIoT development framework
Marvin Carl May,Daniel Glatter,Dominik Arnold,Daniel Pfeffer,Gisela Lanza
DOI: https://doi.org/10.1016/j.jmsy.2023.12.001
IF: 12.1
2024-01-04
Journal of Manufacturing Systems
Abstract:Industrial Internet of Things (IIoT) solutions are poised to enable digital transformation across industry verticals. However, the diversity of IIoT applications has resulted in a multitude of constantly evolving technologies, standards, and reference architectures with differing terminologies and frequent incompatibilities. A unified approach to describe and develop IIoT systems is needed not only to ease development, but also identify cross-cutting research gaps and exploit synergies. Building on a substantial literature survey, this work proposes the IIoT Development Framework, taking a systems engineering perspective to derive an IIoT Systems Development Process along with several key tools: After defining business objectives, an IIoT system is broken down into system components using the IIoT Architecture Pattern. Second, functionalities for each component are identified leveraging morphological boxes. Third, suitable technologies capable of providing these functionalities need to be selected. For this task, technology selection criteria and processes as well as technology comparison sheets are detailed, covering key technologies. Throughout the entire process of technology selection, the IIoT System Canvas is the central tool summarizing components, required functionalities and selected technologies to provide a comprehensive end-to-end view of the entire IIoT system. It may be understood as a modeling template akin to the Business Model Canvas. The IIoT Development Framework aims to achieve three objectives: Completeness (covering all aspects of IIoT systems), actionability (low barriers for researchers and practitioners) and generalizability (covering not just one product family, but cutting across industry verticals). The framework is validated by comparison to the state of the art and application to a set of use cases covering multiple industry verticals with differing requirements, showing superior results especially in terms of completeness and actionability, while maintaining generalizability.
engineering, manufacturing, industrial,operations research & management science