PROPHESY · Project

Self-Configuring Predictive Maintenance Platform That Cuts Factory Downtime and Repair Costs

manufacturingPilotedTRL 7

Imagine your factory machines could tell you exactly when they're about to break down — like a car dashboard warning light, but way smarter. PROPHESY built a complete toolkit that collects sensor data from production equipment, uses machine learning to spot invisible wear patterns, and tells maintenance teams what to fix before anything actually fails. It even includes augmented reality glasses so a remote expert can guide a technician through a repair without flying in. The whole system was tested in two real factories, not just in a lab.

By the numbers

consortium partners across industries and research

countries represented in the development consortium

complex real-plant demonstrators for validation

industry partners involved in development

SMEs in the consortium

demo-ready prototype deliverables

total project deliverables produced

60%

industry ratio in the consortium

The business problem

What needed solving

Factory downtime from unexpected equipment failure is one of the most expensive problems in manufacturing — machines break without warning, production stops, and emergency repairs cost far more than planned maintenance. The data that could predict failures already exists in factory sensors, but it's fragmented across incompatible systems, poorly analyzed, and disconnected from maintenance workflows. Manufacturers need a way to turn scattered machine data into actionable maintenance decisions before breakdowns happen.

The solution

What was built

PROPHESY built a complete predictive maintenance platform with 16 demo-ready components, each iterated through v1 and v2 cycles: IoT data collection infrastructure, CPS middleware with edge-to-cloud processing, machine learning models for degradation detection, a data visualization portal with shareable dashboards, augmented reality remote maintenance support, KPI tracking with cost-benefit calculators, a security and data protection layer, a training and knowledge sharing platform, and a service optimization engine that composes all these into turn-key PdM solutions.

Audience

Who needs this

Automotive and aerospace manufacturers with automated production lines experiencing costly unplanned downtimeIndustrial equipment OEMs wanting to offer predictive maintenance as a value-added serviceFood and pharmaceutical processors where unplanned stops mean product spoilage and compliance riskMaintenance service providers looking to upgrade from reactive to predictive service modelsFactory operators running legacy equipment that lacks built-in condition monitoring

Business applications

Who can put this to work

Automotive Manufacturing

enterprise

Target: Car parts manufacturers and assembly plants with automated production lines

If you are an automotive manufacturer dealing with unexpected machine breakdowns that halt your production line — this project developed a predictive maintenance platform validated in 2 real-plant demonstrators that collects IoT sensor data, detects degradation patterns with machine learning, and lets remote experts guide repairs through augmented reality. With 16 demo-ready components including KPI tracking and cost-benefit tools, it targets the metrics you already care about: OEE, MTBF, and end-of-life predictions.

Industrial Equipment & Machine Vendors

mid-size

Target: Companies that sell or lease CNC machines, presses, or packaging lines and want to offer maintenance-as-a-service

If you are a machine vendor looking to move from selling equipment to selling uptime — this project built a service optimization engine that composes turn-key predictive maintenance packages from modular building blocks. The platform includes a data visualization portal, automatic data collection from IoT devices, and a training platform, all developed through 2 iteration cycles (v1 and v2). With 15 consortium partners including 9 industry players, it was designed for real commercial deployment.

Food & Beverage Processing

mid-size

Target: Food production plants with continuous process lines where unplanned stops cause spoilage

If you are a food processor where an unplanned stop means raw materials spoiling and batches going to waste — this project developed edge-to-cloud middleware that works with your existing equipment sensors, plus machine learning models that identify invisible degradation patterns before failure. The security and data protection module was built in from the start, which matters when your production data is commercially sensitive. The system was validated across 2 complex demonstrators in real production environments.

Frequently asked

Quick answers

What would it cost to implement this predictive maintenance system?

The project's EU contribution amount is not available in the dataset, so specific development costs cannot be cited. However, the platform was designed as modular service bricks — you can adopt individual components (e.g., just the IoT data collection or just the AR remote maintenance) rather than the full stack, which allows staged investment. The built-in KPI tracking and cost-benefit tools were specifically designed to help calculate ROI before full commitment.

Can this scale to a large factory with hundreds of machines?

Yes. The PROPHESY-CPS middleware was built for large-scale distributed data collection and processing, with edge devices handling local computation and cloud integration for the heavy analytics. The platform was validated in 2 complex real-plant demonstrators, not just lab setups. The architecture supports streaming data from multiple machines simultaneously.

Who owns the intellectual property and can I license this technology?

The consortium of 15 partners across 8 countries jointly developed the platform, coordinated by Netcompany (Belgium). IP ownership follows the EU Horizon 2020 grant agreement where each partner typically owns the results they generated. To license specific components, you would need to contact the relevant consortium partner through the coordinator.

How does this integrate with our existing factory systems (ERP, MES, SCADA)?

PROPHESY was specifically designed to address the problem of data fragmentation and limited interoperability in manufacturing. The CPS middleware infrastructure provides integration capabilities at both enterprise and field levels. The automatic data collection component connects to existing IoT and sensor infrastructure rather than requiring wholesale replacement.

How long would it take to deploy in our plant?

The project name itself stands for 'rapid deployment of self-configuring and optimized predictive maintenance services.' The service optimization engine composes turn-key PdM solutions from pre-built modules, which reduces custom development. Based on available project data, the 2 real-plant demonstrators provide proven deployment blueprints, though exact timelines depend on your specific factory setup and which modules you need.

Is our production data safe with this system?

Security was a first-class concern — the consortium built a dedicated Security, Trustworthiness and Data Protection module that went through 2 development iterations (v1 and v2). This covers both cybersecurity for the CPS infrastructure and data protection compliance. Based on available project data, this was one of 16 demo deliverables with dedicated prototype implementations.

Does this work only for specific machine types or industries?

The platform was designed to be industry-agnostic within manufacturing. The service optimization engine composes PdM solutions from modular building blocks that can be configured for different machine types and production processes. It was validated in 2 different complex demonstrators, showing cross-domain applicability. The machine learning models identify degradation patterns and optimize FMECA activities regardless of specific equipment.

Consortium

Who built it

The PROPHESY consortium is unusually strong for commercial follow-through: 9 out of 15 partners (60%) come from industry, which is well above typical EU project ratios. Led by Netcompany from Belgium, the partnership spans 8 countries (BE, DE, EL, ES, LU, NL, PT, UK) giving it broad European market reach. With 4 SMEs in the mix alongside larger industrial players, 2 universities, and 3 research organizations, the consortium balances cutting-edge R&D with practical commercial deployment expertise. The fact that this was an Innovation Action — not a basic research project — means the entire consortium was oriented toward building something deployable, not just publishable.

NETCOMPANY SACoordinator · BE
SENSAP MICROSYSTEMS ANONIMI ETAIRIA ILEKTRONIKON SYSTIMATON KAI EFARMOGON LOGISMIKOUparticipant · EL
UNPARALLEL INNOVATION LDAparticipant · PT
MARPOSS MONITORING SOLUTIONS GMBHparticipant · DE
MONDRAGON GOI ESKOLA POLITEKNIKOA JOSE MARIA ARIZMENDIARRIETA S COOPparticipant · ES
PHILIPS CONSUMER LIFESTYLE BVparticipant · NL
RESEARCH AND EDUCATION LABORATORY IN INFORMATION TECHNOLOGIESparticipant · EL
JAGUAR LAND ROVER LIMITEDparticipant · UK
Maintastic GMBHparticipant · DE
TECHNISCHE UNIVERSITEIT EINDHOVENparticipant · NL
NETCOMPANY S.A.thirdparty · LU
NOVA ID FCT - ASSOCIACAO PARA A INOVACAO E DESENVOLVIMENTO DA FCTparticipant · PT

How to reach the team

Netcompany SA (Belgium) coordinated this 15-partner consortium. Contact them through SciTransfer for a warm introduction to the right technical lead.

Order a report on this consortium See NETCOMPANY SA profile →

Next steps

Talk to the team behind this work.

Want to explore how PROPHESY's predictive maintenance platform fits your production environment? SciTransfer can arrange a direct conversation with the consortium partners who built and tested it.

Order a report on this topic More in Manufacturing & Industry 4.0