From Application Recommendation to Smart Engineering Tool
2 November – 2 December 2026 · for student teams · prize pool €1,500 · registration until 23 October
AutomationML Application Recommendations describe how AutomationML can be applied to concrete engineering problems. But what happens when such a recommendation is transformed from a document into an interactive, executable and machine-interpretable engineering tool?
The AutomationML Hackathon 2026 continues the 6th AutomationML Conference. Student teams have one month to turn the Application Recommendation Drive Configurations into a working engineering tool, and to show what the next generation of engineering standards could look like.
When
Registration until: Fri 23 Oct 2026
Kickoff webinar and start: Mon 2 Nov 2026
Submission deadline: Wed 2 Dec 2026
Final presentations (online): Mon 7 Dec 2026
Who
Teams of up to five students from universities in the AutomationML community. Every team has an academic supervisor.
Prizes
1st place: €800
2nd place: €500
3rd place: €200
The Challenge
Develop a software application based on the AutomationML Application Recommendation Drive Configurations (M-CAD aspects).
The goal is an interactive drive-chain configurator that demonstrates how an AutomationML standard can become more than a static document: a reusable, verifiable and extensible engineering resource. Your application should let users configure a drive chain interactively and represent the resulting engineering model in AutomationML. It should not only create AutomationML files, but also understand, validate and reconstruct the engineering information contained in them, while providing a foundation that can evolve together with the standard.
What Your Solution Should Do
Minimum requirements
Every submission must cover these four points to be eligible for a prize:
- Interactive drive-chain configuration based on the concepts defined by the Application Recommendation.
- AutomationML roundtrip: configurations can be exported to AutomationML, imported again and reconstructed for further editing.
- Reuse of the AutomationML libraries provided with the Application Recommendation.
- Machine-executable validation rules derived from requirements and constraints in the Application Recommendation, with test cases covering valid as well as deliberately invalid configurations.
Graded depth
Beyond the minimum, the jury looks at how far you take the following:
- Attribute Library integration: selecting appropriate standardized attribute types from the AutomationML Attribute Library for the modeled information.
- Traceable validation: linking each implemented rule, where possible, to the statement or section of the Application Recommendation it comes from.
- Extensible architecture: future extensions of the Application Recommendation, additional classes, attributes, rules or related AutomationML libraries can be integrated without fundamentally redesigning the application.
- Completeness of the supported drive-chain elements, and robustness of the roundtrip.
Go Beyond the Core
Teams are encouraged to explore additional ideas. Possible directions include:
- graphical modeling of motion profiles
- mechanical–electrical integration with other AutomationML Application Recommendations
- engineering plausibility checks or calculations
- support for time-series or measurement data
- integration into the AutomationML Editor
- component or catalogue integration
- AI-assisted configuration and validation
- natural-language explanations of validation errors
- machine-readable representations of standard rules
- identification and implementation of useful extensions to the existing Application Recommendation
- novel approaches toward interactive or smart standards
These are examples, not limitations. Creative approaches are explicitly encouraged.
Technology
Teams may choose their own technology stack. For the backend, Python or C# are recommended. The frontend technology is completely open: a web application, a React-based interface, a desktop application or an AutomationML Editor plug-in are all possible.
The focus is not on a particular framework, but on AutomationML correctness, engineering functionality and a convincing, extensible software concept.
Target format. All AutomationML models are created in CAEX 3.0 (IEC 62424:2018). This is the version that allows attributes to reference standardized attribute types from the AutomationML Attribute Library.
AI coding assistant. Every registered team receives access to Claude as an AI coding assistant for the duration of the hackathon.
Hackathon Materials
- AR Drive Configurations (M-CAD aspects): standard page with abstract, scope and citation
- Download package (ZIP): the Application Recommendation document, its AutomationML libraries and example models
- Drive Configurations, hands on: three interactive pages that show the concepts of the AR in the browser. Use them for inspiration; your solution is expected to go well beyond them.
- Best Practice Recommendation: Units
- All AutomationML specifications, including the AutomationML Attribute Library
All inputs for the hackathon are published on this page, so that you can get familiar with them before the kickoff. Further material will be added here before 2 November; questions are answered at the kickoff webinar.
Deliverables
By 2 December 2026, each team submits a public GitHub repository under the MIT licence containing:
- a working software prototype, with a way for the jury to try it out (a hosted demo, a container, an installer or clear build instructions: your choice);
- the source code;
- example AutomationML models demonstrating the roundtrip;
- the implemented validation rules and their test cases;
- a short description of the architecture, the implemented functionality and how future AR extensions could be integrated.
In addition, each team gives a short final presentation with a live demonstration, online, on 7 December 2026.
Evaluation
Submissions that meet the minimum requirements are evaluated by the Technical Advisory Council of the AutomationML association.
| Criterion | Weight | What the jury looks at |
|---|---|---|
| AutomationML & Standard Conformance | 20 % | Correct and meaningful use of AutomationML concepts, libraries, attributes and the Application Recommendation. |
| Engineering Functionality | 20 % | Quality and completeness of the drive-chain configuration and the AML roundtrip. |
| Smart Standard Concept | 20 % | How effectively the solution turns knowledge from the Application Recommendation into machine-interpretable, executable validation logic. |
| Architecture & Extensibility | 15 % | Separation of domain model, AutomationML representation, validation logic and user interface; how easily future AR extensions or further AutomationML concepts can be added. |
| Software & User Experience | 10 % | Usability, robustness and overall implementation quality. |
| Innovation | 15 % | Additional ideas that advance the use of AutomationML, the Drive Configurations AR or the concept of interactive standards. |
Timeline
Until Fri 23 Oct 2026
Team registration
Mon 2 Nov 2026
Kickoff webinar, start of the hackathon
2 Nov – 2 Dec
Hacking month. Teams work on their solutions with their supervisors.
Wed 2 Dec 2026
Submission deadline, 23:59 CET
Mon 7 Dec 2026
Final presentations and live demos, online at the AutomationML Technical Advisory Council, and award of prizes
Who Can Participate
- The hackathon is aimed at students from universities within the AutomationML community.
- Teams consist of up to five students; doctoral candidates cannot join a team.
- Every team needs a supervisor from its university, who is named at registration.
- Teams may work collaboratively throughout the hackathon and have access to AutomationML resources, existing examples and supporting material.
Registration
Registration is open until Friday, 23 October 2026. Register your team with the form below, or send the same information by e-mail to the AutomationML office at office@automationml.org. Registered teams receive the kickoff webinar details and their access to the AI coding assistant.
Your mission: turn an AutomationML Application Recommendation into a working engineering tool, and show us what the next generation of engineering standards could look like.