Skip to main content

Decentralised architectures for optimised operations via virtualised processes and manufacturing ecosystem collaboration

Periodic Reporting for period 2 - DISRUPT (Decentralised architectures for optimised operations via virtualised processes and manufacturing ecosystem collaboration)

Reporting period: 2018-03-01 to 2019-08-31

Our core concept is to DISRUPT the traditional automation pyramid by utilising the capabilities offered by modern ICT to facilitate (i) in-depth (self-) monitoring of machines and processes, (ii) decision support and decentralised (self-) adjustment of production, (iii) effective collaboration of the different IoT-connected machines and devices with tools, services and actors (iv) seamless communication of information, knowledge, and decisions from and to the plant floor and (v) efficient interaction with value chain partners. More specifically, within the modular-structured Smart Factory of DISRUPT, each physical element of production (machine, device, sensor, etc.) is monitored and controlled via the IoT by its virtual counterpart. The vast amount of data collected is processed, refined and analysed to obtain knowledge and to detect complex events that, in turn, trigger automated actions fed back to the physical production units at the plant floor or presented to decision makers along with the appropriate tools to handle and utilise them. To that end, DISRUPT offers a set of decision support tools based on three core interrelated and interacting modules: modelling, simulation and optimisation.
To achieve these aims, the DISRUPT project includes the following specific objectives:
• develop a reference implementation of a collaboration platform
• hybridise and synchronise factory automation by combining a bottom-up orchestration of CPSs and production systems with a top-down virtualisation of production processes
• design, deploy and validate a responsive architecture that is both service- and event-driven
• deploy analytics both at the lower and higher level
• provide modelling and design services that allow the abstract representation of actual manufacturing systems
• offer co-simulation tools combined with optimisation algorithms
• develop optimisation methods and tools for finding the optimal or near-optimal solutions to problems of production scheduling, capacity planning and factory throughput, production planning and manufacturing chain optimisation
• implement the entire system as a bundle of cloud-based tools and services
• establish proof-of-concept by demonstrating the above on real industrial cases.
Summarising the status at the end of the project, we have the following results:
-trends in the Automotive and Home Appliances sectors, the expectations from the stakeholders (consumers, industries) and the relevance of components in this context
-definition of business scenarios and goals, KPIs, use case and requirements for the pilots
-proposal of business models
-specification of the internal functionality of all modules
-architecture specification for a flexible data-driven solution to automated vertical and horizontal integration
-specification of meta-models and models to flexibly configure architecture functionality at high level by domain experts by simply changing the models
-innovative integration of modelling and simulation tools, with an emphasis on manufacturing scenarios and use cases
-novel meta-heuristic algorithms for a variety of production scheduling problems
-innovative optimisation models for handling event-driven inbound logistics in a factory
-testing and validation of the above in pilots /CRF, Arcelik)
-integration of the Cypher-Physical Control (CEP) to the messaging bus
-extension of the Messaging bus to handle simulation events and manage simulation activities
-deployment of CPS kernel software for identification purposes
-coverage of all the use cases on the selected pilots (Arcelik, CRF)
-business cases for teh selected pilots (Arcelik, CRF)
DISRUPT has achieved its ambitious goals through the work that is being conducted in this second period. The impact of the project to both the Factories of the Future strategic objectives for digital automation and the trends in the target industries is high and includes the following:
• To achieve increased efficiency of production units and reduced down-time due to malfunctions through the adoption of a modular, decentralised CPS structure, empowered with complex event processing, simulation and optimisation tools as well as the ability to self-adjust;
• To be able to adjust to rapidly changing market needs and unexpected events through advanced capability for agile manufacturing processes, based on hybrid architectures;
• To effectively integrate cyber-physical operations, advanced analytics and decision support tools with the organisation's Enterprise Information Systems towards disrupting the traditional automation pyramid;
• To foster ecosystem-wide collaboration between developers and machine providers so that they address specific manufacturing business cases by offering ready-to-deploy solutions;
• To deliver an interoperable reference platform for more effective coordination of the manufacturing chains;
• To reach increased productivity through the broad use of ICT tools combining modelling, design, optimisation and simulation for day-to-day operations and strategic planning;
• To achieve improved resource consumption profile for processes and products arising from resource-aware optimisation, situational awareness enabled by IoT and CPS and co-simulation allowing more proactive product and process design or ramp-up;
• To advance the knowledge gained from processes and products and communicated in real-time from the plant floor to any management level through interoperable advanced analytics and the integration with ICT tools;
• To exploit big data technologies and modern cloud-based architectures to increase the efficiency of ICT tools and services in future smart manufacturing applications.

Productivity is enhanced in terms of (i) fast reaction to unexpected market changes, (ii) minimisation of differences between the final production and the actual market needs, (iii) efficient production scheduling and planning and (iv) effective (self-) adjustment to unexpected events.
The DISRUPT system enables resource optimisation and minimisation of reaction time and thus lead to a more efficient production line, as per the main economic trends of manufacturing , i.e. (i) the economics of production, related to technologies such as additive manufacturing that make it possible to cost-effectively manufacture products more quickly, in smaller batches and (ii) the economics of the value chain, related to digital technologies that are narrowing the distance between manufacturer and consumer, allowing manufacturers to bypass traditional intermediaries.
DISRUPT has achieved an adaptable system that enables reconfiguration of processes and production units, with the introduction of the appropriate semantics and modelling schemes that ensure interoperability and facilitate interaction. With these functionalities, the DISRUPT platform has significant impact to the time-to-market factor for industries, as this approach minimises the delays not only in the two investigated use cases (home appliances and car manufacturing) but in any other field where DISRUPT can be applied, and thus minimise the economic consequences of these delays.
The business case developped by one of the key DISRUPT partners (CRF) indicates that the approach demonstrated by DISRUPT and applied on a single plant can lead to a mean savings over the next 5 years after industrialisation of nearly 900k€, with a RoI under 2 years.
disrupt-logo.jpg