Ecosystem for Collaborative Manufacturing Processes _ Intra- and Interfactory Integration and Automation

logo_COMPOSITION
Summary

Data and services have become the key factor in optimising manufacturing processes. The need to react to dynamically changing market demands is dramatically rising. One of the most imperative problems so far is to connect supply chain data and services between enterprises and to connect value chain data within a factory so that it can meaningfully support decision-making. COMPOSITION will create a digital automation framework, the COMPOSITION Integrated Information Management System (IIMS) that enhances the manufacturing processes by exploiting existing data, knowledge and tools to increase productivity and dynamically adapt to changing market requirements. This technology acts as the technical operating system for business connections between factories and their suppliers. Furthermore, it opens a new space for third party entities to actively interact along the supply chain, e.g., by providing services to optimise cycle time, cost, flexibility or resource usage. In addition to the supply chain improvements, the processes inside the company can be addressed and optimised. Data across the (multi-sided) company internal value chain is integrated by the IIMS with optimisation and modelling tools for resource management including innovative, multi-level, real-time cross-domain analytics and a Decision Support System. The technology will extend existing FI-WARE, FITMAN and LinkSmart catalogues, in addition to a Building Management System for environmental data gathering. Inter-operability, ease of retrofit and scalability are critical parameters that are taken carefully into account in devising such IIMS. 

More information
Website: http://www.composition-project.eu/
Duration: 36 months
Start date: 01-09-2016
End date: 31-08-2019
Number of participants: 12
Total budget - EC contribution: 7634254 Euro - 7634254 Euro
Call topic: Digital automation - Collaborative manufacturing and logistics (FoF.2016.11-i)
Instrument: Collaborative project (generic)
picture_COMPOSITION
Location
Characteristics

Use Case Requirements and Lessons Learned - (3)  

View Kanban
    1. Comment:

      1. The COMPOSITION Marketplace Management System shall enable stakeholder to gain access to the COMPOSITION open marketplace.

      2. COMPOSITION Marketplace(s) should have possibility of restricted access. 

      3. The line visualization shall compare the actual processed units to the target ones. 

      4. Alarms/Notifications are forwarded to subscribers depending on their impact level. 

      5. It must be possible to reset an alert when the necessary measures have been taken. 

      6. Ecosystem components should be deployed as Docker images. 

      7. Agents shall be writable in any programming language. 

      8. The Decision Support System shall import data coming from the simulation and prediction engines. 

      9. Supplying companies register their products/services in specific topic(s) within the ecosystem. 

      10. The needs and requirements of companies shall be registered/published within the ecosystem. 

    2. Comment:

      1. Early design decisions on deployment and communication protocols were made. (Docker, MQTT, AMQP). Deciding on the deployment and communication platforms has made test deployment and integration work easier to manage.

      2. Inception design (from the DoA) did not specify some components, e.g., for operational management or configuration. The architecture needed additional components to cover system configuration and monitoring. 

      3. Blockchain is still not a plug-and-play technology and requires a substantial amount of low-level configuration. 

      4. The Matchmaker should match agents (requester and suppliers). Moreover, the Matchmaker should match a request with the best available offer.

      5. Use cases need to be solidly anchored in the real world of the actors and end users. They must not solely represent what is feasible from a technical point of view, but also reflect non-functional requirements such as regulations and business practices. Otherwise, the business cases would become unsustainable for further exploitation. 

Challenges - (9)  

View Kanban
      1. Comment: Using the COMPOSITION platform, companies will be able to enter a supply chain and broaden their clientele, thus improving their economic performance. The COMPOSITION Marketplace will facilitate matching of demand and supply across the supply chain. Buyers and sellers will have close to perfect knowledge of offers, with searchable, low latency access to information. Agent technology and automated matching of requests offers lowers transaction costs. By providing a common interface for entering the marketplace, the platform also lowers the barriers of entry. In economics terms, these are all prerequisites for competitive market (from which everyone benefits).
      2. Comment: Improvements of the manufacturing process by using Process Modelling and Monitoring Framework combined with Integrated Digital Factories Models will lead to better resource management and will reduce the use of resources. Optimising the process involving factories with recycling companies will generate economic savings. Monitoring the health of machinery and using predictive maintenance and reducing energy consumption will reduce not only manufacturing efficiency (less downtime & material waste) but also the environmental impact of manufacturing.
      1. Comment: The project will focus on the reduction of energy consumption via the detection of excessive consumption and anomalous equipment behavior. The health of equipment will be improved, using machine learning/deep learning algorithms for predicting failures & arranging preventative maintenance. It will be possible to monetise the results into suggestions for actions at specific processes. The possibility of letting the Marketplace matchmaker take externalities into account when rating offers and partners by providing it with information on e.g. environmental performance of marketplace actors, is being examined.
        1. Comment: The reduction of scraps will be addressed in the sense of attempting to reduce unused material, rather than reducing the number of defective workpieces or rework operations. Furthermore, the project will reduce the amount of scraps by optimising the processes in a way that less material has to be scrapped because certain maximal idle times are not reached.
      2. Comment:

        The COMPOSITION Integrated Information Management System will be a digital automation framework that optimises the manufacturing processes by exploiting existing data, knowledge and tools, integrating them with newly installed cyber physical systems (CPS) and automation software, in order to increase productivity and to allow for dynamic adaptation to changing market requirements. CPS using IoT devices such as wireless sensors nodes (WSN) can be easily retrofitted to existing equipment and infrastructure to gather sensory data and detect anomalies as well as opportunities to improve productivity and cvcle time.

Technologies and enablers - (18)  

View Kanban
      1. Comment: Part of the scrap generated by the manufacturing industry is recyclable. The recycling companies need to stay connected with the industries and to plan for the further reselling and/or disposal of the material. COMPOSITION will connect the dots in the inter-factory pilot.
      1. Comment: Decision makers are not at their desk the whole day. In order to reduce the probability that they miss important information, the project investigates user interface solutions for the shop floor and everywhere else in the factory, in order to provide data monitoring, control and notification systems, that help humans in interaction with machines and with the decision making process.
      2. Comment: Continuous monitoring of the condition and performance of specific critical parts (equipment, infrastructure and environment) of the manufacturing system will provide feedback to the real-time brokering and message translations services allowing seamless integration of heterogeneous manufacturing components.
      1. Comment: The Digital Factory Model (DFM) will provide an integrated representation of the intra-factory domain at machine-level, end-user-level, and process-level. The DFM model will be used to exploit data coming from machinery, sensors and production lines and will offer interoperability in communication by providing all these heterogeneous data in a common format. The DFM is re-used in the Decision Support System, in Simulation and in the Deep Learning Toolkit. The integration of information will enable better use of the data available in the physical layer of the factories to be made.
      2. Comment: The COMPOSITION Integrated Information Management System will be a collaborative ecosystem where all members of a manufacturing supply chain can connect in a secure environment. The Security Framework will implement the security core mechanisms aiming to ensure the security, confidentiality, integrity and availability of the managed information for all authorised COMPOSITION stakeholders.
        1. Comment: The COMPOSITION ecosystem will include Simulation and Forecasting Tools, the main purpose of which will be to simulate processes and provide forecasts of events whose actuals outcomes have not yet been observed. The simulation engine will forecast possible malfunctions, delays in production processes and machine failures using both historical and real data and will feed the Decision Support System which will offer to users, suggestions and solutions for the upcoming failures.
      3. Comment: The COMPOSITION project will address the issue of extended enterprises allowing the interaction and collaboration of all stakeholders in the COMPOSITION Marketplace as every business entity will be represented by agents at the ecosystem. Some of information from companies’ IIMS will be reflected at the ecosystem in a Collaborative Manufacturing Services Ontology which will be used as a central knowledge base. The Matchmaker will infer knew knowledge from the knowledge base by applying both syntactic and semantic matching in terms of manufacturing capabilities, in order to find the best possible supplier to fulfil a request. Different criteria for selection according to several qualitative and quantitative factors will be considered.
      1. Comment: COMPOSITION defines innovations at factory performance level, but also at business model level. Collaboration among multi-disciplinary companies will be possible in the ecosystem.
      1. Comment: The COMPOSITION project contains a Digital Factory Model Component which offers representation of factory processes and resources in a common format based on well-known standards. Moreover a Simulation and Forecasting Toolkit analyses the production processes and required resources in an integrated way and extracts forecasts for possible failures. Also forecasting is provided in supply chain and logistics, especially in fill level monitoring of bins and boosts the waste management and recycling processes. Sustainable manufacturing will be assisted by a Decision Support System. The Marketplace will enable dynamic integration with actors in the supply chain.
      1. Comment: Open standards will be favoured and special consideration will be taken to the ones already supported by stakeholder products. The COMPOSITION project will constantly keep an eye on emerging standards for IoT interoperability (e.g. OGC Sensor Things API), which are for sure a key enabler for the IIMS framework. In general, platform will be compatible with many already existing standards so that new starndard-compatible plants can be easily added.

Digital mapping framework - (47)  

View Kanban

For some background information about the digital mapping framework, please see here.

    1. Comment: Viable business framework for deploying the platform embedded in a multistakeholder ecosystem considering IPR, ownership and protection of data.
      1. Comment: 1) Traffic Monitoring in IoT-enabled shopfloors, 2) Intra-factory Optimization & Decision support (i.e. forecasting services, trend analysis, predictive maintenance), 3, 4, 5) Robust module communication & message routing for Intra & Inter-scenarios
            1. Comment: Real-time co-simulation methods will enable key stakeholders to simulate and forecast complex activities. Simulation will target increasing quality and flexibility as well as more efficient management of logistics by also decreasing service time and costs.
          1. Comment: COMPOSITION will create a digital automation framework (the COMPOSITION IIMS) that optimizes the manufacturing processes by exploiting existing data, knowledge and tools to increase productivity and dynamically adapt to changing market requirements.
            1. Comment: Provision of novel decentralised architecture capable of contextually handling shared situational awareness, based on the usage of cyber-physical systems and automation software for continuous real-time monitoring and control of the underlying complex collaborative industrial processes for performance and cost optimization.
            2. Comment: Application of business intelligence for the coordination of mechanisms of collaborative manufacturing process. Use of a Collaborative manufacturing services ontology for the description of both supply and demand entities, and manufacturing processes. Use of intelligent matchmaking algorithms in order to match possible customers/suppliers and fulfil requests with the best available offers.
          2. Comment: Definition of manufacturing process and domain models and implementation of a process-oriented monitoring framework to keep under control forecasted production line states and to allow prescriptive and preventative actions.
          3. Comment: Use of data analytics to provide automated knowledge extraction from big data generated in manufacturing in order to reduce factory and field failures and detect and remedy anomalies earlier.
          4. Comment: Simulation and prediction engine to represent the operation of the manufacturing processes over time so to forecast impacts and reconfigurations of the production process and to predict the probability and timeline of future faults and delays based on a wide variety of analysis techniques such as Descriptive statistics, Linear regression analysis, Markov models , Genetic algorithms for optimization and Correlation heatmaps Simulation and Prediction techniques will be also applied in inter-factory domain for recyclable material management by analysing and correlate factors such as materials, customers, prices, weights, routes and time. Prediction engine will also provide the scientific and algorithmic requirements for the Decision Support System (DSS).
          5. Comment: See Simulation, as in the COMPOSTION project Simulation & Prediction are addressed in an effectively combined way.
        1. Comment: Data acquisition tools to instantiate virtual manufacturing processes.
        2. Comment: The platform will apply smart data processing methods for investigating the dataset in order to find logical links. It will also investigate novel complex event processing approaches.
        3. Comment: Big data analytics in cloud environments to support manufacturing scenarios.
          1. Comment: Machine learning techniques, such as particle filters or hidden Markov models, or continuous/recurrent deep neural network, are going to be employed for analysing big datasets.
        4. Comment: Investigation of advanced HMIs for direct interaction with real-world objects. Consideration of mobile user interfaces that allow accessing crucial immediate information everywhere in the factory. Consideration will also be given to data gathering from ultra low power IoT devices such as wireless sensors where data can then be aggregated and visualised at an appropriate HMI interface. Where possible provision for self-powering these IoT devices using energy harvesting will be taken into account in order to avoid battery replacement. Flexible and dynamic data-driven solutions that can be adapted to different environments and needs.
      1. Comment: Interoperability between installed equipment and infrastructure (machines, environment, people) is critical to capture and process data for the COMPOSITION IIMS components.
          1. Comment: KeyCloak (Authentication, Message Broker)
          2. Comment: Role-based identity management through KeyCloak (Clients, Roles, Users)
        1. Comment: Log-Oriented Architecture through Blockchain Implementation for providing Audit Trail for Manufacturing & Supply Chain data, Matchmaking services (including semantic queries) as part of the Open Marketplace
          1. Comment: Cyber-physical systems usually come with heterogeneous communication protocols, many of them widely used like KNX, ZigBee, ModBus, etc. All the different field buses are integrated in the BMS for data collection and, afterwards, the information is forwarded though the local WLAN. Scalability, re-configurability and inter-operability with existing infrastructure will be a key decision factor in installing such CPS.
          2. Comment: Docker-Based application packaging for standalone & integrated framework
          3. Comment: Integrated Digital Factory Model for Intra-Factory & MarketPlace Open Reference Data Model for Inter-Factory Support (bids, offers and transactions, etc.)
          1. Comment: Digital Factory Models’ instances will gather data in a common format. It is based on well-known standards such as BPMN, B2MML, gbXML and OGC. They are based XML and JSON syntax and provide a high level of simplicity, extensibility, interoperability and openness.
          2. Comment: 1) Security Information & Event Management API, 2) BPMN standard as part of the Integrated Digital Factory Model, 3) RabbitMQ implementation, 4) OGC sensorthings compliant API through Integrated Digital Factory Metadata Model, 5) Part of the Integrated Digital Factory Model
      2. Comment: Design a technical operating system for the support connected and interoperable factories with their stakeholders and for the optimisation of manufacturing and logistics processes.
      1. Comment: Security by design approach. Development of a Security Framework composed by a core set of security mechanisms to guarantee the security, confidentiality, integrity and availability of the managed information for all authorised stakeholders in the supply chain while at the same time maintaining suitable levels of IPR and knowledge protection. The COMPOSITION Security Framework iapplies blockchain technology to provide an audit trail for manufacturing and supply chain data, enabling both product data traceability and secure access for stakeholders. Cybersecurity mechanisms will be developed to monitor and protect against potential threats that could affect the COMPOSITION infrastructure.
      2. Comment: The modular nature of the COMPOSITION system will allow easy and fast scalability for use in both large and small manufacturing sites with high or low complexity. The systems should also be modular such that multiple smaller systems & sub-systems can easily be integrated into larger systems
    2. Comment: Requirements of modern production processes stress the need for greater agility and flexibility, for faster production cycles, increased productivity, less waste and more sustainable production. Human-machine interaction is put in the center, supporting the decision making process.
      1. Comment: see HMI
        1. Comment: An initial choice is going to be made for WSN technology so that the sensors can be retrofitted on, or near to, existing equipment on either a temporary or permanent basis.
        2. Comment: The COMPOSITION IIMS will include a core set of data management and analytics tools to fuse data coming from different disjoint levels of the factory lifecycle.
          1. Comment: The COMPOSITION ecosystem consists of connected IIMSs and it will also provide interfaces and concepts for third parties to offer products and services in the marketplace directory.
      1. Comment: Unsupervised deep learning approaches will be designed and tested to support multiple functions: i) optimization of logistics, ii) distribution of workload and work opportunities over specific areas or entire countries and iii) automated drafting of Life Cycle Assessment (LCA) of products.
      1. Comment: Product life-cycle modelling approach will be supported by a simulation engine to represent the operation of the manufacturing processes over time.

Digitalisation pathways - (18)  

View Kanban
    1. Comment:

      The concept of the hyperconnected factories is approached in the interfactory part of the project with connections between different links of the supply chain, using an agent-based marketplace. 

    2. Comment:

      The concept of the autonomous factories is approached in the intrafactory part of the project with connections between different links of the value chain. Agent marketplace and automated bidding process which enable automated negotiation and transaction.

        1. Comment:

          A part of information at shopfloor level may be fed to a MOM via a texteditor for the final user. 

        1. Comment:

          Implemented at both central pilots. 

        2. Comment:

          Implemented at both central pilots. 

        1. Comment:

          Implemented at both central pilots. 

        2. Comment:

          IoT enabled connectivity with intrafactory systems. 

        1. Comment:

          Applicable. 

Diverse - (3)  

View Kanban
    1. Comment:

      Cooperation with other FoF and especially FoF-11 projects. Interactions with EFFRA, as three partners are members of the association. Partners are members of various clusters and promote advances in the project to them. COMPOSITION is participating in the FOF-11 workgroups “WG DA – 02 - Blockchain / Security” and “WG DA – 04 - Data analytics”. Finally, COMPOSITION is leading the workgroup “WG DA – 07 – Marketplace”. It is also participating in the WG-Ineteroperability and WG-Business Modelling. 

    2. Comment:

      The project contributes to the Industrial Ontology Foundry with a Supply Chain use case (CERTH/ITI) and a Predictive Maintenance (ATL) use case. Members of the consortium are participating in the WG-Maintenance. 

      Partner TNI-UCC is also contributing to standardisation activities. 

logo_COMPOSITION
Results