BLACK FOREST HACKATHON

🚀 Digital border-breaker: A cross-border service of documents

🔍 Problem Statement: Incoming documents are reviewed manually and checked for completeness and compliance with legal requirements. The current process involves extensive manual activities, the partial manual preparation of correspondence, and dispatch by post. The incoming documents look different, cannot be submitted electronically and cannot be sent electronically to the recipients. As there is an amount of approximately 20.000 requests annually it is a time-intensive task.

🎯 Hackathon Challenge: How might we use a web-based and secure application, that ensures traceability, data integrity and transparency to deliver incoming requests, especially traffic violation fine notices, from various countries (e.g. Switzerland, Austria, Italy) to German citizens. We want to reduce manual processing, increase speed and reduce time spent on incorrect and incomplete requests.

📊 Available Data & Resources:Participants will have access to a template letter, a workflow model and a decision tree for different use cases.

🏆 Success Criteria: A successful solution should solve the basic task of putting address data in a database and produce pdf-documents ready for postal delivery. The procedure must comply with data protection regulations in all cases.

💡 Key Considerations: All incoming documents are (automatized?) classified whether complete, incomplete, or containing errors. Documents should be transferred into a database. This ensures a complete record of which requests were (or were not) successfully forwarded.

🚀 Scan2Configure: Digital Equipment (Pre-)Consultation with Cart, Calculation and 3D 

🔍 Problem Statement: During the outfitting items from our sampling center are selected, documented by hand and then transferred into digital systems. This leads to media discontinuities, errors and ambiguities (e.g. incorrect items, missing variations of a product), time loss and an inconsistent data base between the outfitting, the calculation software (WeKo), construction planning (CAD) and 3D visualization (myWeberHaus Designer)*.​

🎯 Hackathon Challenge: How can we develop a high-preformance web application that uses automated identification to instantly link physical exhibits in the showroom or catalog items with the digital statement of work in the customer file, validate dependencies, and ensure consistent data transfer to the calculation and visualization software*?​

📊 Available Data & Resources:
– Data / Content: Sample master data, sample consultation case (including follow-up processing), code mapping​
– Access to following systems: WeKo (data export/import files, API examples), myWeberHaus customer portal*, demonstration of a house in the myWeberHaus Designer​

🏆 Success Criteria:
– Core user journey defined (application usage) ​
– Scan and add to shopping cart UX suitable for live customer consultations ​
– Reduction of errors during and after the outfitting (potentially reduce follow-up processing e.g. reduction by 80%)​
– Data consistency​

💡 Key Considerations:
– Usability ​
– Data quality, rules and dependencies​
– Integration and interfaces​
– Security and data protection​
– Scalability (e.g. integrating pre-outfitting, the customer journey in the World of Living, furniture exhibitions, etc.)​

🚀 The AI Sewer Detective: Automated Damage Analysis in the Underground 

🔍 Problem Statement: Reviewing thousands of hours of video footage has become a massive bottleneck: experts currently sift through every recording manually, cross-referencing them with a complex damage catalog. Existing external AI solutions are far too slow, as massive data volumes and long upload times lead to week-long delays. We need an in-house solution that understands our infrastructure in real-time.

🎯 Hackathon Challenge: How might we develop a locally-hosted AI solution that autonomously evaluates video streams according to the official damage catalog, detects anomalies, and pinpoints their exact location (meter mark/timestamp) to slash processing time from one week to just a few hours?

📊 Available Data & Resources:
Participants will have access to real-world inspection videos, the official damage classification catalog, and anonymized sample reports to serve as a reference for the required output format. 

🏆 Success Criteria:
A successful prototype demonstrates the automated detection of at least three core damage classes (e.g., cracks, faulty joints, deposits), provides accurate spatial and temporal mapping, and proves the technical feasibility of high-performance local processing without cloud dependency.

💡 Key Considerations:
The solution should be scalable for future data sources (e.g., high-res drone footage), offer a user-friendly interface for final review by technical experts, and be architecturally prepared for the integration of the extensive full damage catalog.

🚀 Once-Only Dog Tax – Code a Platform

🔍 Problem Statement: In Germany’s municipal administration landscape, over 11,000 municipalities independently manage dog tax registration. When a citizen relocates from one city to another (e.g., Hannover to Leverkusen), they must manually de-register their dog in the old municipality and re-register in the new one – submitting the same information twice, waiting weeks for processing each time, and causing administrative overhead. This issue leads to citizen frustration, duplicated effort, wasted administrative resources, and a gap between current practice and the EU’s Once-Only Principle (legally binding since 2023), which aims to ensure that data already held by one authority should not need to be re-submitted by citizens.

🎯 Hackathon Challenge:How might we use SaaS architecture, REST APIs, and automated data transfer protocols to enable seamless dog tax registration transfers between municipalities, so that citizens only need to submit their information once, while ensuring strict data isolation between municipal tenants and full audit compliance?

📊 Available Data & Resources:
Participants will have access to real German municipal dog tax regulations – they should select any two cities to discover how wildly their tax rules differ! To support their work, we supply teams with regulatory documents and guidelines as background resources. These include EU Once-Only Principle documentation and sample data structure guidance for owners and municipal tenants,  providing essential context for the regulatory framework and data standards. Participants can supercharge their coding with AI assistants like GitHub Copilot, ChatGPT, or Claude. They may choose any  programming language, REST framework, and database they want – SQLite is recommended for zero setup. One rule: solutions must remain local and free – no paid cloud services are allowed!

🏆 Success Criteria:
A successful solution should enable information flow between municipal administrations while ensuring the strong rules for data protection and tax secrecy. A good solution enables information flow; a great solution also ensures compliance with data protection rules. The solution should include automated tax calculations that account for different regulations between municipalities, demonstrating how the same dog can be taxed differently when transferred from one city to another. The solution should include documentation and a UI that can be presented in a demo. Evaluation Metric: Solutions will be assessed on (1) enabling seamless information flow and (2) maintaining full data protection compliance – with excellence in both dimensions defining a great solution.

💡 Key Considerations:
When transferring dog registrations between municipalities, solutions should preserve all historical data by marking cases as de-registered rather than deleting them, maintaining a complete audit trail. Participants are strongly encouraged to leverage AI coding assistants to accelerate development and debug issues.

🚀 Municipal AI Assistant: Taming Documents, Accelerating the Energy Transition

🔍 Problem Statement: Municipalities are tasked with implementing the complex and constantly evolving requirements of the energy transition, often lacking sufficient capacity, time, and specialized expertise. The current legal framework is scattered across numerous sources and is difficult to keep up to date, leading to uncertainty and immense research efforts. To date, conventional AI tools have failed to provide reliable, citable support for these specific needs.

🎯 Hackathon Challenge: Develop an AI application that does more than just answer questions about the energy transition. Your solution should simplify administrative processes by extracting necessary data directly from unstructured documents (laws, contracts, plans) to automatically prepare municipal notices and spreadsheets (templates).

📊 Available Data & Resources:
Participants will receive access to three core types of documents:
1. Curated Content: Laws, plans, contracts, and relevant websites to serve as the foundation for the knowledge base.
2. Sample Municipal Documents: Files to be uploaded and processed as additional context (e.g., official notices, planning documents).
3. Sample Municipal Templates: Excel tables, Word documents, or PDF forms to be (semi-)automatically populated by the AI.

🏆 Success Criteria:
The solution must be intuitive to use, provide accurate and citable answers from the stored knowledge base, and transparently flag any uncertainties. Furthermore, it should reliably identify and repurpose content from additional user-provided documents. Finally, it must assist in filling municipal templates automatically or semi-automatically, achieving a measurable reduction in workload compared to manual processing. 

💡 Key Considerations:
The solution is to be implemented as a standalone web application. It must ensure transparency through the use of citations and a mandatory „I don’t know“ mechanism. Furthermore, it should be designed in a way that future development can fully account for data protection compliance as well as European hosting options and AI models.

🚀 Construction Site Check: Detecting Conflicts Before the Excavator Rolls

🔍 Problem Statement: Municipalities and Badenova plan numerous construction and excavation projects, but information is often scattered across different systems or shared too late. This leads to redundant excavations, delays, and unnecessary costs, while increasing the burden on citizens. There is a lack of a central, easy-to-use platform that consolidates all planned measures and makes overlaps visible at an early stage.

🎯 Hackathon Challenge: How can we develop a digital solution that centrally bundles planned construction and excavation projects and significantly simplifies coordination between the municipality, Badenova, and their partners throughout the entire process? The goal is a functional prototype that captures measures intuitively and identifies spatial overlaps at the street level as well as temporal conflicts within a one-month window. The solution should visualize these overlaps and automatically inform the responsible parties to create a reliable foundation for joint planning.

📊 Available Data & Resources:
Structured baseline data on typical construction projects (address, timeframe, purpose, etc.) will be provided. These serve as a sample for the future data model and should be supplemented in the prototype by additional measures created by the teams. Furthermore, teams will receive sample role and personnel data for guidance, which can be flexibly expanded or adapted. Map or geospatial data are intentionally omitted to allow room for diverse ideas regarding spatial representation and conflict detection.

🏆 Success Criteria:
The challenge is considered successful if the provided construction measures are captured completely and clearly, and if new measures can be added intuitively. The prototype should clearly display temporal and spatial overlaps between different trades and automatically notify relevant contact persons when measures are created or modified. Additionally, it must demonstrate how different user groups can be integrated into the system through a role and permissions concept in future development stages.

💡 Key Considerations:
The solution is to be implemented as a standalone web application. Its architecture should consider data protection and data security as critical factors within the context of municipal users and critical infrastructure, while ensuring an intuitive user experience.