PRONTO · Project

Smart Data Analytics to Keep Aging Process Plants Running at Peak Efficiency

manufacturingPrototypeTRL 4Thin data (2/5)

h2020pronto.eu

Imagine you own a chemical plant that's been running for 30 years. Every pump, pipe, and compressor throws off data — temperatures, vibrations, pressures — but nobody's connecting the dots between a wobbly pump and the fact that your whole production line is wasting energy. PRONTO trained a new generation of engineers to build software that reads all those signals together, figures out which equipment is dragging down the whole network, and tells operators exactly where to focus so the plant runs cleaner and cheaper without replacing everything.

By the numbers

30-50 years

Typical lifetime of industrial process plants targeted by this research

Consortium partners across academia and industry

Industrial partners contributing real-world problems and data

Countries represented in the consortium

53%

Industry ratio in the consortium

The business problem

What needed solving

Process plants built decades ago generate massive amounts of data from sensors across mechanical, electrical, and process systems — but this data sits in silos. Equipment health is monitored separately from production performance, so operators miss the connection between a degrading pump and declining plant efficiency. The result: wasted energy, unplanned shutdowns, and suboptimal output from assets that must keep running for 30 to 50 years.

The solution

What was built

PRONTO produced data analytics methods for assessing condition and performance of interconnected process equipment, optimization algorithms that use real-time machinery data to improve resource use across plant networks, and completed industrial case studies demonstrating these approaches. The project also trained a cohort of doctoral researchers under the European Industrial Doctorate scheme with hands-on industry experience.

Audience

Who needs this

Chemical plant operators with aging infrastructure seeking efficiency gainsOil and gas refinery maintenance and operations managersIndustrial automation software companies building plant-wide optimization toolsUtility companies managing distributed generation and processing assetsEngineering consultancies advising process industry clients on digital transformation

Business applications

Who can put this to work

Chemical Processing

enterprise

Target: Chemical plant operators with aging infrastructure

If you are a chemical plant operator dealing with declining efficiency in equipment that's been running for decades — this project developed data analytics methods that combine condition monitoring with process optimization across your entire plant network. Instead of treating each piece of equipment in isolation, the approach connects machinery health data to overall production performance, helping you squeeze more output from existing assets without costly replacements.

Oil & Gas Refining

enterprise

Target: Refinery operations and maintenance teams

If you are a refinery manager dealing with unplanned shutdowns and rising maintenance costs on equipment installed 30 to 50 years ago — this project produced case studies showing how real-time condition data from process equipment can drive smarter resource allocation. By linking equipment health to production optimization, maintenance teams can shift from reactive fixes to planned interventions that keep the whole process network running.

Industrial Automation & Software

mid-size

Target: Predictive maintenance software vendors

If you are an automation software company looking to move beyond single-machine monitoring toward plant-wide optimization — this project's research across 8 industrial partners produced methods for integrating disparate data sources from mechanical, electrical, and process sub-systems. The case studies and trained researchers coming out of this consortium represent expertise you could hire or license to build next-generation optimization products.

Frequently asked

Quick answers

What would it cost to implement these optimization methods?

PRONTO was a doctoral training network (MSCA-ITN-EID), not a commercial product development project. The outputs are research methods, case studies, and trained researchers rather than turnkey software. Implementation costs would depend on hiring the trained researchers or engaging the consortium partners as consultants to adapt methods to your specific plant.

Can this work at the scale of a real industrial plant?

The project included 8 industrial partners who contributed real-world problems and data. Deliverable D.4.4 confirms that case studies were completed and disseminated, indicating methods were tested against actual industrial scenarios. However, full-scale deployment would require further engineering beyond these research demonstrations.

Who owns the intellectual property from this research?

IP from MSCA training networks typically stays with the host institutions — in this case Imperial College London and the partner universities and companies. Since 8 industrial partners were involved in joint supervision of doctoral research, some IP may be co-owned. Contact the coordinator for licensing specifics.

Which industries were these methods tested in?

Based on the project objective, the methods target Europe's process industries broadly — chemical, petrochemical, and similar continuous production facilities. The 8 industrial partners span 7 countries including Germany, Norway, Italy, and Spain, suggesting testing across multiple industrial contexts.

Is this ready to deploy or still in research phase?

This was primarily a research training program that ran from 2016 to 2019. While industrial case studies were completed, the core output is trained PhD researchers and published methods, not deployable software. The results would need to be productized before commercial deployment.

How does this differ from existing predictive maintenance tools?

Most commercial tools monitor individual machines in isolation. PRONTO's approach connects equipment condition data to process network optimization — understanding how one machine's degradation affects the entire production chain. This network-level view is what distinguishes it from single-asset monitoring products.

Consortium

Who built it

The PRONTO consortium is well-balanced for an industrial training program, with 8 industry partners and 7 universities across 7 countries (Germany, Spain, Italy, Norway, Poland, UK, and the US). The 53% industry ratio is strong for an academic training network, ensuring doctoral research stays grounded in real problems. Imperial College London, one of the world's top engineering schools, coordinates. The spread across major European industrial economies (Germany, Italy, Norway) means the methods were shaped by diverse manufacturing contexts. However, with zero SMEs in the consortium, the results may be more suited to large enterprises than smaller manufacturers. For a business looking to adopt these methods, the industrial partners are the most relevant contacts — they have both the research insights and the operational context.

IMPERIAL COLLEGE OF SCIENCE TECHNOLOGY AND MEDICINECoordinator · UK
ABB AGparticipant · DE
AKADEMIA GORNICZO-HUTNICZA IM. STANISLAWA STASZICA W KRAKOWIEparticipant · PL
UNIVERSIDAD DE VALLADOLIDparticipant · ES
INEOS KOLN GMBHpartner · DE
TECHNISCHE UNIVERSITAT DORTMUNDparticipant · DE
ABB ASparticipant · NO
ABB SPZOOparticipant · PL
BASF SEparticipant · DE
CARNEGIE MELLON UNIVERSITYpartner · US
EQUINOR ASApartner · NO
CRANFIELD UNIVERSITYparticipant · UK
NORGES TEKNISK-NATURVITENSKAPELIGE UNIVERSITET NTNUparticipant · NO

How to reach the team

Imperial College of Science Technology and Medicine, London, UK — look for the lead professor in the Department of Chemical Engineering or Computing

Order a report on this consortium See IMPERIAL COLLEGE OF SCIENCE TECHNOLOGY AND MEDICINE profile →

Next steps

Talk to the team behind this work.

Want to connect with the PRONTO researchers or industrial partners? SciTransfer can identify the right contact and arrange an introduction.

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