Lab scale use case deployment with lessons learnt I


This deliverable outlines the architecture defined for integration activities, a brief overview of thecomponents developed so far as well as any initial findings and lessons learnt during the initial integration activitiesin a lab scale environment. It also includes an integration plan for the inter-factoryand intra-factory use casesand a set of recommendations, conclusions and next steps. An ISMB docker server has been set up that acts as our lab scale environment where each COMPOSITION partnerdockerizesand deploys theircomponent. Areference architecture has been successfully developed that is nowbeing used in individual use cases.To date the following components have been developed based on this architecture, where possible leveraging from commercially available softwaremodules and open source widely used protocols. A detailed description of the functionality of each of these components can be found in section 5and a tabulated summary found in section 10.

  • Authentication
  • Big Data Analytics
  • BlockChain Connector
  • Commissioning System
  • Deep Learning Toolkit
  • Intrafactory Interoperability Layer
  • Manufacturing Big Data Storage
  • Manufacturing Decision Support System
  • Market Event Broker
  • MatchMaker
  • Real Time Multi Protocol Event Broker
  • Requester Agent
  • SIEM (Security Information and Event Management)
  • Simulation and forecasting tool
  • Supplier Agent

To development and dockerisation state of the components have been outlined, along with action plans for use in inter and intra-factory use cases.At a high level there are two communication mechanisms in the COMPOSITION system: message-based communication over MQTT or AMQP using the Event Broker1, and request-response REST HTTP interfaces (D2.3 The COMPOSITION Architecture Specification I, 2017). With a few exceptions2, the components are loosely coupled and integrated in the system by conforming to a common communication infrastructure and data schemas. There are dependencies on data generated by other components, but for most of the data, the dependency is not direct as this data is distributed through the event broker, so the direct connection is to this component.

Ecosystem for Collaborative Manufacturing Processes _ Intra- and Interfactory Integration and Automation
Not specified (see website if available)
  • Not specified