Technical Assets
Building block 9:
Augmented Reality interface
Building block 3:
Decision Support Framework (DSF)
Building block 4:
Financial Analysis
Building block 5:
Prognostic and health management toolkit
Building block 6:
Predictive Maintenance
Building block 7:
Refurbishment & Re-manufacturing Planning
Building block 8:
Data Analytics
Building block 1:
Smart sensors & digital retrofitting
Building Block 2:
Cybersecurity for IoT
1. Physical Layer
Building block 1:
Smart sensors & digital retrofitting
An adaptive sensor network and digital retrofitting infrastructure attached to the refurbished or remanufactured machines will retrieve data and accelerate predictive maintenance tasks.
Main innovations:
RECLAIM includes IoT controllers with hardware acceleration capabilities. A network of low-cost programmable logic IoT devices will deliver high performance analytics and health monitoring for operation profiling and predictive maintenance tasks. Those IoT devices will offer a sweet spot between performance, flexibility and power consumption.
Building Block 2:
Cybersecurity for IoT devices
Cybersecurity endpoint protection will be embedded both into digital retrofitting infrastructure design. Secure IoT devices protect sensitive and personal data.
Main innovations:
Security by design mechanisms will secure user and device authentication, encryption, intrusion detection/intrusion prevention, and overall enhanced cyber-secure operation.
2. Real Time Decision-Making Layer
Building block 3:
Decision Support Framework to optimise lifetime-extension strategies
The Decision Support Framework (DSF) is designed to support and improve refurbishment and re-manufacturing of machinery decisions. DSF will identify and propose strategies based on different criteria such as the impact and value of refurbishment, extension to asset life, optimal timing, machine condition and possible upcoming failures, production planning, and resource allocation.
The DSF will bundle all the tools in the real-time decision-making layer together into one, easily navigable tool. It will have attributes from both knowledge and model-driven type decision making tools, including scoring mechanisms, rule-based decision making and AI algorithms. Data mining algorithms will help propose decision trees, genetic algorithm, and ensure the extraction of valuable information from IoT data.
A visual analytics suite to capture and translate insights will provide users with actionable strategies, alternatives process models, KPIs visualisation and real-time health assessment of different production aspects.
Main innovations:
Flexible knowledge- and model-driven DSF which is adaptive and reliable in real-time momentary situations to
- improve competitiveness;
- maximize productivity;
- increase resource use efficiency; and
- increase awareness of resource use efficiency deviations for the existing or future control process units.
Building block 4:
Cost modelling and financial analysis toolkit
This component provides an effective cost estimation tool for cost and financial impact. It will take into account all types of life extension strategies and activities – helping to estimate the resources needed for each activity. The modelling will be linked to incoming data generated, providing real time life cycle cost estimation.
The toolkit will be developed to perform these functions across multiple industries, expanding their benefits and impact to European manufacturing.
Main innovations:
Monte-Carlo Simulation statistical modelling and discrete event simulation will deliver a precise cost and financial analysis to support reliable decision-making on refurbishment and re-manufacturing strategies. Multiple real time cost implications may be visualised based on monitoring of the equipment health status.
Building block 5:
Prognostic and health management toolkit
A component-level prognostics and health management tool will be developed to increase equipment lifetime, productivity and service quality. RECLAIM will use shop floor data in order to calculate overall equipment efficiency and extract other meaningful information for prediction and prevention capabilities.
Main innovations:
Capturing data from devices to improve the decision-making process for predictive maintenance, leveraging the interactions and relationships between device data and expert data.
Creating integrated equipment degradation and quality probability based on system level and influence diagrams using both expert knowledge and operational data.
Building block 6:
Fault diagnosis and predictive maintenance digital twin
A factory environment digital twin will monitor and predict performance and status of factory assets. This will provide all the information needed to perform proper maintenance planning, optimising production throughput and reducing stoppages.
The system will monitor patterns in real time and compare them with historical data, to autonomously identify repeated scenarios and create rules to handle them.
Main innovations:
Training, testing and adoption of predictive maintenance algorithms to better predict future outcomes.
Building block 7:
Refurbishment & re-manufacturing toolkit
Production planning optimisation using IoT data will create high value information for monitoring production as a precursor to deploying improvement and control steps. Smart sensors (building block 3) together with system constraint and behaviour recognition ensure the best possible outcomes.
Main innovations:
Machine learning techniques will make long-term optimisation of production planning possible – preventing failures, malfunctions and abnormalities, as well as obtaining better predictive performance.
Building block 8:
In-situ repair data analytics
Industrial analytics are used to identify and recognise machine operational and behavioural patterns to make fast and accurate predictions and act with confidence when needed.
A visual analytics suite will use
- perception (monitoring) elements on the shop-floor and
- comprehension (inspection, exploring) thanks to an extensive network of sensors.
Main innovations:
Tailored solutions for existing data structures added to new batch and streaming visual analytics will give powerful repair and drill-down analysis.
3. User Layer (Facilities for User)
Building block 9:
AR-enabled multimodal interaction mechanisms
A novel way to visualise and localise information on equipment refurbishment and re-manufacturing operations.
Using a network of sensors and proposals from the decision support framework, technicians will be able to vision an augmented reality of several streams of data.
During refurbishment and re-manufacturing, the system will provide animated 3D stepwise instructions on disassembly and reassembly required, as well as support in the form of on-the-job remote assistance with real-time audio-visual communication and 3D annotation to technicians during the procedure.
Main innovations:
Real-time localisation and 3D augmented reality with natural language, hand gesture and gaze input interaction algorithms and real-time AR annotation for remote assistance.