Ready for a challenge?

INNOVATION CHALLENGE FOR STUDENTS IN DINGOLFING/LANDAU

Ready for a challenge in the high-tech hub of Dingolfing? Join our 24-hour innovation challenge and tackle real-world problems in the fields of lean, green & digital. Collaborate with an interdisciplinary team to develop cutting-edge prototypes for regional companies, cities and global industry leaders. Expect exciting prizes, exclusive insights into the Ecosystem Smart Ostbayern—a dynamic network of regional business, science and society—and invaluable networking opportunities. After the Innovation Challenge, a party will take place, followed by a free overnight stay in a four-star hotel. The next day will begin with a business breakfast attended by companies from the region. Who knows? You might even land your dream job along the way! Are you up for it?

registration image

Register now!

When: 08.05.2025 to 10.05.2025
Where: Dingolfing and Landau

  • Due to limited capacity, the number of participants for the Innovation Challenge is limited. Secure your place and apply now!
  • You can find the descriptions of each use case below. Take a look to see which challenge you would like to work on and what requirements you already have.
Register by 30 April 2025

Not an IT-Nerd? No problem! Whatever you study – please apply! We are looking forward to having various perspectives on our Use Cases!

Simply complete the survey. Registration will end automatically once all places are filled.

registration

WHAT TO EXPECT AT OUR INNOVATION CHALLENGE

Use Cases

Host and Location

  • This use case is hosted by: SAR
  • This use case is hosted at: SAR office

Motivation

When commissioning new machines and systems, testing the PLC is important to ensure the quality requirements for the entire system and to avoid damage when the system is started up. For complex and widely distributed systems, this IO check involves a high workload for the technicians and thus considerable costs.

Target

Development of a mobile application to support technicians in performing an IO check of PLC-controlled systems with the objective of increasing efficiency.

Approach

  • Introduction, planning, definition of objectives
  • App development and technical implementation
  • Validation of the app using an exemplary system model including PLC
  • Preparation and planning of the presentation

Distribution of Activities

60 % development and programming, 20 % user experience design, 20 % validation

Useful knowledge

  • There are no technical requirements for participants to consider. For information on the candidate profile, see the next point.

Additional information

Candidate profile: Structured work and team spirit, Skills in the development of mobile applications (e.g. Android/iOS, Java, C++, C#), and Basic knowledge of PLCs.

Host and Location

  • This use case is hosted by: DE software & control GmbH
  • This use case is hosted at: DE office

Motivation

High-quality work instructions are essential for efficiency and safety in the modern working world. Traditionally, these instructions are manually created using text and images, which is time-consuming and prone to errors. Especially in noisy production environments where videos without sound are recorded, it is challenging to effectively utilize these videos and combine them with documentation.

Target

We aim to develop a Best Practice Framework for dynamically creating work instructions from videos, applicable to various process types such as assembly, maintenance, machine handling instructions, sales conversations, and care activities. This framework should enable the creation of standardized and high-quality documentation.

Approach

  • Identification and Selection: Analyze relevant process types and select the three most useful processes.
  • Requirements Analysis: Conduct a detailed analysis of the requirements for creating work instructions.
  • Process Description: Develop a generic process description considering various conditions and customer requirements.
  • Prototype Implementation: Implement and test a prototype for at least one selected process using the Halerium AI platform.

Distribution of Activities

30% Requirement-Engineering, 30% Development and Programming, 20% User Experience Design, 20% Exploration

Useful knowledge

  • Basics in GenAI and Prompt Engineering
  • Understanding how to create documentation and work instructions
  • Experience with AI platforms, ideally Halerium
  • Interest in user interface design and user experience

Additional information

Technical Prerequisites • AI Platform Halerium: Based on GPT, Halerium provides a powerful environment for creating and processing work instructions. • Programming Languages: Python, JavaScript Attendees are invited to familiarize themselves with Halerium in a free remote training session. During the challenge, experts will be available as mentors for work instructions and AI. Participants will work with existing pipeline components, design their own workflows, and test their solutions in real industry cases. This offers an excellent opportunity to gain practical experience and expand skills in an innovative and hands-on project.

Host and Location

  • This use case is hosted by: City of Landau
  • This use case is hosted at: City of Landau

Motivation

Which open data is available for the city of Landau an der Isar and how can this data be used to improve the city and the life of its residents?

Target

Identify and collect open data sources related to Landau an der Isar. Develop a unified API to make data more accessible for developers and the public. Propose and implement a practical use case that demonstrates how open data can be used to improve the city.

Approach

  • Data Discovery
  • Data Integration
  • Use Case Exploration
  • Implementation

Distribution of Activities

50% development and programming, 20% data exploration and integration, 20% user experience design, 10% AI/ML model integration (if applicable)

Useful knowledge

  • Cloud Services: Microsoft Azure
  • Programming Languages: Python, JavaScript/TypeScript, SQL, REST APIs

Additional information

Candidate profile: Students or professionals with skills in: Software development (frontend/backend), Data science and AI, API design and integration, Urban data analysis, UX/UI design. Suitable for Bachelor’s and Master’s students in Computer Science, Data Science, AI, or related fields. The project can leverage existing open data portals (e.g., German Open Data Portal, municipal open data initiatives). Potential collaboration with city officials to gain access to more relevant datasets.

Host and Location

  • This use case is hosted by: Karl Mossandl GmbH & Co. KG
  • This use case is hosted at: Offices of Karl Mossandl GmbH & Co. KG

Motivation

Actual status: • Belt weigher available on the conveyor belt for the raw material • Manual weighing possible due to existing weighing program or visualization

Target

Target status: • Automated, definable (e.g.daily, weekly, ...) creation of a weighing certificate for the raw material from our gravel plant and sand drying plant • Watchdog (restart) or error message in the event of problems

Host and Location

  • This use case is hosted by: HORSCH LEEB Application Systems GmbH
  • This use case is hosted at: Offices of HORSCH LEEB Application Systems GmbH

Motivation

Agricultural machines are frequently maintained in field by both dealers and owners. During the vegetation period, the windows of operation are quite limited. Therefore, it is crucial to quickly identify the correct spare part to minimize downtime.

Target

Identify spare parts with the help of the smart phone camera and AI.

Approach

  • You are receiving 50 spare parts from Horsch with article numbers. Develop a tool within 24 hours to identify and assign the parts to their article numbers. Use pretrained neural networks. Adapt and train them with additional pictures of our specific parts. Identify parts from various perspectives and conditions, including new, used, and assembled states. Develop a solution for parts that appear too similar and cannot be distinctly identified.

Distribution of Activities

30 % development and programming, 10 % user experience design, 60% AI training and integration

Additional information

Technical prerequisites (Visual Studio / VSCode, GitHub, Copilot). Candidate profile: Programming (Python, (C++), GIT), Basic AI understanding (e.g. YOLO)

Host and Location

  • This use case is hosted by: City of Dingolfing
  • This use case is hosted at: Landshut University, TZ PULS, Bräuhausgasse 33, 84130 Dingolfing

Motivation

The city of Dingolfing supports its citizens with a municipal funding program for the implementation of private climate protection measures. To do so, citizens must submit a funding application to the city using an Excel form. The application is then checked by an employee in the town hall. If all requirements are met, the application is entered into a database and a funding notification is sent to the applicant. This process ties up staff and reduces time for strategic tasks.

Target

The aim of the task is to develop a web-based, automated and smart application process for Dingolfing's climate protection funding program. An AI can independently determine whether an application meets all requirements. A dashboard will also be used to enable comprehensive real-time evaluations of all funding applications.

Approach

  • Applications should be able to be submitted directly via a web application and a corresponding grant notification should be generated automatically by the program. The program should be able to independently check various eligibility criteria, such as the date on the offer. The program also creates a clear database of all applications received and enables real-time evaluation of the various program items. For example, it should be possible to call up how many electricity storage systems have been installed in the city and their total output in kilowatt hours.

Distribution of Activities

20% website development, 20% development and programming, 20% user experience design, 20% AI training and integration, 20% data analysis

Useful knowledge

  • Technical requirements: Open source programs: All programs used should be openly accessible, usable and secure for the administration.
  • Candidate Profile: - Experience with web applications, AI or data processing - Ability to create a clear user interface - Basic understanding of energy technology, climate protection and renewable energies - Bachelor's degree or Master's degree

Additional information

As part of the Dingolfing climate protection funding program, over 700 applications have been submitted to date and €780,000 in funding has been approved. All information on the funding program can be found at: https://www.klimaschutz-dingolfing.de/foerderungen/dingolfinger-anreizprogramm-klimaschutzoffensive The project is based on the previous Excel application form and the grant notifications.

Host and Location

  • This use case is hosted by: BMW
  • This use case is hosted at: BMW Bildungszentrum: Mengkofener Straße 13, 84130 Dingolfing

Motivation

1) Error avoidance through typing/reading errors -> Improved inventory results. 2) Cost and resource savings (work materials, working time for counters and inventory staff, replenishment of containers)

Target

The container are immediately digitally recorded correctly during the count, reconciled, and transferred to Excel. The manual counting and typing effort will be eliminated.

Approach

  • Plant 2.2 -> Can be expanded to all BMW plants for the inventory as best practice

Distribution of Activities

75% development and programming, 25% design of the user experience

Useful knowledge

  • Technical requirements: Linking of PowerApp or similar application with Excel and possibly Sharepoint.
  • Candidate profile: Anyone who can program well and has skills in linking PowerApp or a similar application with Excel and possibly Sharepoint.

Additional information

Currently, the count results are manually recorded by hand on paper. These are then typed into an Excel file and manually checked for accuracy there.

Host and Location

  • This use case is hosted by: BMW
  • This use case is hosted at: BMW Bildungszentrum: Mengkofener Straße 13, 84130 Dingolfing

Motivation

ABiz Learning & Training Center, qualification programs, workshops and room ​occupancy are controlled via an Excel list. Free slots are occupied here for inquiries. Outlook calendar entries are also sent to users.​

Target

The content, the processes, the visualizations must be digitized - in order to automate the ​current “manual work”​

Approach

  • Plant 2.2 -> Scalable for all TEM plants worldwide -> MEX, HU, USA, China, DE​. The tool can also be used by the city for any event - or for internal organization / DGF Ecosystem​.

Distribution of Activities

50% development and programming, 25% user experience design, 25% AI training and integration. (appraisal)​

Useful knowledge

  • Technical prerequisites: APEX is a good fit for our project. An AI-supported decision-making aid in execution would be an advantage.
  • Candidate profile: Anyone who is well versed in APEX programming and can integrate AI help on this platform.

Additional information

On base of the process can be implemented digitalization. Prozess is Known.​

Host and Location

  • This use case is hosted by: BMW
  • This use case is hosted at: BMW Bildungszentrum: Mengkofener Straße 13, 84130 Dingolfing

Motivation

High Quality training data is mandatory for AI training. Unfortunately, data is expensive and is not available in sufficient quantities. Therefore, we want to generate data synthetical to represent different classes of errors adequately. This approach should overcome typical obstacles in industrial AI integration.​

Target

Engine production at BMW Landshut. End-of-Line quality assurance.

Approach

  • We can use various methods, including generative AI (generative adversarial networks or diffusion models) and classical image processing methods, automated with Python.
  • Metaverse- or simulation-based approaches should not be considered.​

Distribution of Activities

60% development and programming, 20% Image Processing, 20% AI training and integration.

Useful knowledge

  • Technical prerequisites: AWS Cloudroom / JUMA, GitHub
  • Candidate profile: Basic Python Programming, Image Processing & Machine Learning Skills

Additional information

During the challenge, experts will be available​.

Host and Location

  • This use case is hosted by: BMW
  • This use case is hosted at: BMW Bildungszentrum: Mengkofener Straße 13, 84130 Dingolfing

Motivation

We are seeking innovative algorithms that can analyze operational data from production facilities and detect early signs of potential disruptions. The goal is to develop a system that can prevent unplanned downtime and increase productivity. Key aspects to be addressed include the utilization of AI models or algorithms for automated pattern recognition in the data, the capability to provide real-time warnings about impending equipment failures, and the seamless integration into the maintenance workflow. Scalability and robustness of the algorithms to cater to the requirements of diverse sectors is also a crucial factor.

Target

Early detection of problems in facilities allows for immediate countermeasures to be taken to prevent major equipment failures.​

Distribution of Activities

Development & programming =75%​, AI training & integration = 25%​

Useful knowledge

  • Technical prerequisites: Programming language Python​
  • Candidate profile: ​Bachelor students, Ability to abstract​, Understanding of Data Processing & AI-Knowledge​

Host and Location

  • This use case is hosted by: BMW
  • This use case is hosted at: BMW Bildungszentrum: Mengkofener Straße 13, 84130 Dingolfing

Motivation

Containers in industrial logistics can often be routed more efficiently if the exact type of container is known beforehand. However, there are hundreds of different container types and differentiating them fully automatically is a challenging task.​

Target

The goal is to optimize BMW‘s logistics processes by enabling more efficient container routing depending on the type of container.​

Approach

  • Use deep computer vision techniques to determine the exact type of container in a supervised manner.​

Distribution of Activities

20% conceptual design, 70% programming, 10% documentation​

Useful knowledge

  • Technical prerequisites:​ Software engineering (Python), Image processing (pillow, cv2), Machine Learning (pytorch)
  • Candidate profile: IT students

Host and Location

  • This use case is hosted by: i-LogiX - Capgemeni
  • This use case is hosted at: TZ-PULS, Bräuhausgasse 33, 84130 Dingolfing

Motivation

Freight forwarding companies face high operational costs due to inefficient route planning, prolonged idle times, and suboptimal resource allocation. By leveraging AI-powered optimization, we aim to enhance the efficiency of truck, trailer, and driver deployment while minimizing downtime and costs.​

Target

Develop an AI-driven route optimization solution to streamline delivery schedules and maximize resource utilization.

Approach

  • Using a provided dataset, we will apply AI algorithms to optimize route planning, considering time windows, delivery cycles, and fleet constraints. The system will aim to reduce idle times, enhance scheduling efficiency, and lower overall operational costs.​

Useful knowledge

  • Technical prerequisites​: AI-based optimization algorithms (e.g., Machine Learning)​; Data processing & analysis (e.g., Python, NumPy)​; Cloud-based or on-premise computing infrastructure​
  • Candidate profile: Experience in AI/ML optimization techniques; Proficiency in programming (Python, SQL, Java)

Additional information

The project focuses on real-world logistics challenges with practical AI-driven solutions. Participants will work with structured datasets to develop a scalable and efficient optimization model for transport logistics.​ Due to the time restriction, initial approaches or the development of initial solution modules are of course also valid!