Towards Implementing a Collaborative Manufacturing Cloud Platform: Experimenting Testbeds Aiming Asset Efficiency

Digitalization of business processes within the industrial sector has led to an increase of complexity in technological solutions. As the industry is becoming increasingly more pervasive, the Internet of Things (IoT) in the industrial sector, known as Industrial IoT (IIoT), is now a reality in the context of industrial transformation. IIoT system designers and providers face new challenges of different natures along the entire product development life cycle, from requirements elicitation and quality to development and deployment. In projects from this nature, it is important to uncover assumptions, challenges and approaches faced in the analysis, design, implementation and deployment of IIoT systems [1].

IIoT is the enabler to achieve interoperability between ground floor assets (Edge tier), and the Cloud and IT infrastructure Tiers (if taking an architecture pattern from reference models like IIRA). IIoT promotes the supervision of manufacturing and maintenance operations, as well as the automation of Enterprise Resource Planning (ERP) production orders and logistics activities. Industry is also moving from verticalized organizations by functions and departments into product-oriented organizations. These reasons raise the need for heterogeneous systems to communicate with each other. In addition, due to the technological advances, the systems are located in different places from edge to cloud.

Using the standards’ layers allows to understand the involved systems that compose the smart manufacturing ecosystem in these projects. Industrial reference models like Industrial Internet Reference Architecture (IIRA) [2], Industrie 4.0 Reference Architecture Model (RAMI 4.0) [3], and ANSI/ISA-95.00-2000 Enterprise-Control System Integration (or ISA-95, in short) [4] provide a standardized organization of concepts within the modernized industrial environment.

Adopting these reference architectures however is not sufficient for providing technical and architectural requirements required for developing IIoT systems. The closest to real implementation in literature relates to specific use cases where industrial internet technologies are being applied, i.e., testbeds based in IIRA and RAMI 4.0. Even though the conducted Industrial Digital Thread (IDT) and Asset Efficiency (AE) testbeds allow architecture implementations, using these reference models does not provide technical specifications in an abstraction level required for conducting organizational projects, as there is still a lack in properly align the technology with the business requirements.

This paper describes and discusses aspects regarding different challenges in implementing a collaborative manufacturing cloud platform. Challenges start by properly adopting a suitable architecture strategy, followed by eliciting software requirements, addressing interoperability, and implementing an end-to-end solution (i.e., encompassing the shopfloor, edge, cloud and business layers).

The reference models and separation of concerns were used as basis for defining the entities that interact for a set of scenarios. Process modeling of such scenarios is being increasingly adopted in IIoT projects, since it allows to takes in consideration the various aspects of the manufacturing system, not only the real-world business approaches, the technical parameters and the human interaction, but also the life cycle of the system, the product and the services. This paper uses a funded Portuguese research and development (R&D) projects for the IIoT domain called PRODUTECH-SIF (Solutions for the Industry of the Future).