AIPlan4EU · Project

Ready-to-Use AI Planning Tools That Automate Complex Scheduling and Decision-Making

digitalPilotedTRL 7

aiplan4eu.fbk.eu

Imagine you run a factory or a logistics company and every day you face hundreds of decisions: which machine runs what job, which truck goes where, when to restock. Right now, most companies solve this with spreadsheets, gut feeling, or expensive custom software. This project built a universal toolkit that lets any company plug in their specific scheduling problem and get back an optimized plan — without needing a PhD in AI. Think of it like a universal remote control, but instead of TVs, it works with any AI planning engine out there.

By the numbers

consortium partners across Europe

countries represented in the consortium

industry partners validating real-world use cases

SMEs involved in development and testing

total project deliverables produced

iterative release versions of the final UPF platform

50%

industry participation ratio in consortium

The business problem

What needed solving

Most companies with complex scheduling and planning problems — factories juggling production orders, logistics firms optimizing routes, agrifood businesses timing perishable supply chains — either rely on manual planning or invest heavily in custom optimization software. Existing AI planning tools are powerful but fragmented: it is hard for non-specialists to find the right technique, there are no shared standards, and encoding business rules into a planner requires scarce expertise.

The solution

What was built

The project built the Unified Planning Framework (UPF) — a planner-agnostic API that lets businesses access multiple AI planning engines through a single, standardized interface. The final implementation was released in three versions and integrated into the European AI On-Demand platform. Across 27 deliverables, the team produced working prototypes, a final platform, standard interfaces to common industrial technologies, and guidelines for practitioners.

Audience

Who needs this

Logistics companies struggling with vehicle routing and delivery scheduling optimizationManufacturers needing automated production scheduling across multiple machines and shifting prioritiesAgrifood companies managing time-critical decisions in perishable supply chainsSoftware integrators building planning features into enterprise applicationsOperations managers at any company where manual planning creates bottlenecks

Business applications

Who can put this to work

Logistics and Supply Chain

mid-size

Target: Mid-size logistics operators and 3PL providers

If you are a logistics company dealing with route optimization and delivery scheduling across dozens of vehicles — this project developed a universal planning API (the UPF) that connects to multiple AI planning engines through a single interface. Instead of building custom optimization software from scratch, you can plug your fleet data into the UPF and let it select the best planning engine for your problem automatically. The consortium included 9 industry partners who validated real-world use cases.

Manufacturing and Production

SME

Target: Discrete manufacturers managing complex production schedules

If you are a manufacturer dealing with production scheduling across multiple machines and shifting priorities — this project built a planner-agnostic platform that reasons about your constraints and generates optimal production plans. The final UPF implementation was released in three iterative versions and tested against real industrial use cases. With 6 SMEs in the consortium, the tool was specifically designed to be accessible without deep AI expertise.

Agrifood and Precision Agriculture

any

Target: Agrifood companies optimizing harvest, processing, and distribution timing

If you are an agrifood company dealing with time-sensitive decisions about when to harvest, process, and ship perishable goods — this project explicitly targeted agrifood as a key application area. The UPF platform lets you encode your domain rules (shelf life, weather windows, equipment availability) and get back automated scheduling decisions. The tool is available as a standalone resource or through the European AI On-Demand platform.

Frequently asked

Quick answers

What does this cost to implement?

The UPF platform was developed as an open resource integrated into the European AI On-Demand (AI4EU) platform. Based on available project data, the tool is designed to be freely accessible through the AI4EU platform, though integration into your own systems would require development effort proportional to your use case complexity.

Can this handle industrial-scale operations?

The project went through iterative development with two major UPF releases (at months 12 and 30) and a final platform extension released in three versions. The consortium included 9 industry partners and validated the platform on use cases both from within the consortium and from external innovators recruited through cascade funding. This suggests readiness for industrial-scale deployment in supported domains.

What about IP and licensing?

The UPF was developed as a general, planner-agnostic API designed for integration into users' own systems. Based on available project data, it was built for the public AI4EU platform, suggesting open or permissive licensing. Contact the coordinator at Fondazione Bruno Kessler for specific licensing terms.

How long would integration take?

The project developed standard interfaces between the UPF and common industrial technologies specifically to lower the integration barrier. The platform provides a uniform API that abstracts away the complexity of individual planning engines. Based on deliverable descriptions, the team prioritized user-centered design and concrete guidelines for practitioners.

Does this replace our existing scheduling software?

No — it complements existing systems. The UPF acts as a middleware layer that connects your business logic to specialized AI planning engines. It was designed to be integrated into users' existing systems through a standard API, not to replace your ERP or MES.

What planning problems can it actually solve?

The project covers automated planning and scheduling for decision-making scenarios where you need to reason about when and how to act to achieve objectives. Explicitly mentioned application areas include agile manufacturing, agrifood, and logistics. The platform supports multiple planning engines that can be selected based on your specific problem type.

Is there ongoing support after the project ended?

The project closed in December 2023, but the UPF was integrated into the European AI On-Demand platform (AI4EU), which continues to operate. The project website at aiplan4eu.fbk.eu and the AI4EU platform remain available resources. For technical support, Fondazione Bruno Kessler in Italy led the 18-partner consortium.

Consortium

Who built it

This is a strong, industry-balanced consortium with 18 partners across 8 European countries. What stands out for business buyers is the 50% industry ratio — 9 out of 18 partners come from industry, with 6 being SMEs. This means the technology wasn't built in an ivory tower; half the development team were companies that need this to work in practice. Led by Fondazione Bruno Kessler in Italy, one of Europe's top applied research centers, the consortium spans Belgium, Switzerland, Germany, France, Ireland, Italy, Sweden, and the UK, giving it broad European market coverage and multi-sector validation.

FONDAZIONE BRUNO KESSLERCoordinator · IT
AIRBUSparticipant · FR
EASYMILEparticipant · FR
UNIVERSITAT BASELparticipant · CH
INSTITUT NATIONAL DES SCIENCES APPLIQUEES DE TOULOUSEthirdparty · FR
DEUTSCHES FORSCHUNGSZENTRUM FUR KUNSTLICHE INTELLIGENZ GMBHparticipant · DE
NETWORK RESEARCH BELGIUM SAparticipant · BE
F6S NETWORK IRELAND LIMITEDparticipant · IE
F6S NETWORK LIMITEDthirdparty · UK
OREBRO UNIVERSITYparticipant · SE
SAIPEM S.P.A.participant · IT
CENTRE NATIONAL DE LA RECHERCHE SCIENTIFIQUE CNRSparticipant · FR
MAGAZINO GMBHparticipant · DE
UNIVERSITA DEGLI STUDI DI ROMA LA SAPIENZAparticipant · IT
UNIVERSITA DEGLI STUDI DI BRESCIAparticipant · IT
PROCTER & GAMBLE SERVICES COMPANY NVparticipant · BE

How to reach the team

Fondazione Bruno Kessler (FBK), Trento, Italy — a leading applied research center. Search for the AIPlan4EU project lead at FBK for direct contact.

Order a report on this consortium See FONDAZIONE BRUNO KESSLER profile →

Next steps

Talk to the team behind this work.

Want to know if this AI planning platform fits your scheduling or logistics challenge? SciTransfer can arrange a direct introduction to the development team and help you assess fit — contact us for a tailored briefing.

Order a report on this topic More in Digital & ICT