Out-of-the-Box Ideas

Think outside the box!

You are always welcome to submit your own project proposals, even if they do not fit in any of the categories (as long as they are somehow related to software technology or formal methods ;) ) Just contact any supervisor that you think might fit your idea.

Available Project Proposals

Moreover, here are some project proposals that do not fit elsewhere:

  • Comparing MBSE Models for Cyber-Physical Systems -CPSs (Georgiana Caltais, Aimé Ntagengerwa)

    Project Overview

    Model-Based Systems Engineering (MBSE) is an essential approach for designing and analyzing complex Cyber-Physical Systems (CPSs). This project aims to compare different MBSE modeling languages by applying them to the same CPS, evaluating their strengths and weaknesses.

    Project Objectives

    1. Select a CPS to Model:
    2. The system should ideally have both hardware and software components.Possible candidates include the OH-320 or a similar system.
    3. Identify 2 or 3 MBSE Models:
    4. Identify commonly used MBSE modeling languages.Evaluate their suitability for CPS modeling based on some aspects:
      • Expressivity: How well does the language capture a system’s complexity?
      • Validation & Verification Features: What built-in tools exist for correctness checking?
    5. Model the same CPS Using Different MBSE Languages:
    6. Implement the same CPS in multiple modeling languages.Compare the outcomes and assess differences in capabilities and usability of the MBSE models.

    Expected Outcomes

    • A comparative analysis of 2 or 3 MBSE languages for CPS modeling.
    • A demonstration of how these different languages represent the same CPS.

     Connection to the ZORRO Project

    This work contributes to ongoing research in the Zero Downtime for Cyber-Physical Systems (ZORRO) consortium:

    • You will lay the foundation for a literature study on MBSE languages (relevant for industry stakeholders).
    • Your MBSE models will be aligned with an existing conceptual model to:
    • Validate that model's applicability.Demonstrate Fault Tree (FT) generation across different modeling languages.Explore how system models can be split across different modeling languages while maintaining consistency.
    • Your MBSE models may be used to inspire a unifying representation for different MBSE models
  • A Comparison of Optimization Techniques for Solving the Team Formation Problem (TFP) (Yeray Barrios Fleitas, Eduardo Lalla)

    Supervisors: Yeray Barrios FleitasEduardo Lalla

    Keywords: Team formation problem, metaheuristic optimization, algorithm comparison, team diversity, educational science

    Are you happy implementing algorithms and ready to take on the challenge of optimizing team formation? Here’s your chance to dive into heuristic optimization and compare some of the most exciting algorithms out there—like genetic algorithms, particle swarm optimization (PSO), and nature-inspired methods—to see which one best solves the Team Formation Problem (TFP) with real-world data. In this project, you’ll work with actual student data and a mathematical model of TFP to implement and test different optimization techniques. You’ll analyze the strengths and weaknesses of each algorithm and build a comparative analysis that could guide how we form teams in large, dynamic environments like universities and tech companies. This project is perfect for those passionate about algorithm design, big data, and optimization—a combination that’s highly valued across industries!

    Get more information

    You can find more information about the project (methodology, student profile, learning goals and work plan) by clicking this link

  • Build and Optimize a DRAT certificate checker (Peter Lammich)

    Supervisors: Peter Lammich
    In this project, you will design and optimize a drat certificate checker. DRAT is used to certify the outputs of SAT solvers. While the certificate checker is less complex than a SAT solver, it still has to perform some time-critical operations, that can profit a lot from low-level optimizations.

  • Conversational AI Bot (CaiMate) for Assessing Team Roles and Personal Attributes in Education (Yeray Barrios Fleitas)

    Supervisors: Yeray Barrios Fleitas

    Keywords: Generative AI, Belbin roles, personality traits, interactive chatbots, team diversity, educational science

    Ever wanted to build a chatbot that’s more than just a friendly assistant? Imagine an AI bot that can help students uncover their strengths and roles within a team, guiding them through self-assessments to determine their Belbin team roles, personality type, learning style, and conflict resolution approach. This project lets you create a powerful conversational bot that empowers students to understand themselves better and enhance their teamwork! Using Natural Language Processing and AI-driven conversation flows, you’ll design a chatbot that engages students in personalized conversations, helping them identify their unique traits. This project will teach you to design smooth conversation flows, apply validated assessment frameworks, and provide actionable feedback based on students’ responses. If you’re passionate about conversational AI, educational technology, and making a meaningful impact on team-based learning, this is the project you’ve been waiting for.

    Get more information

    You can find more information about the project (methodology, student profile, learning goals and work plan) by clicking this link 

  • Debug information for verified programming (Edoardo Putti, Peter Lammich)

    Supervisors: Edoardo Putti, Peter Lammich

    Once you proved your program correct there is no need to debug it! Unfortunately all tools for performance analysis require debug information. We want to generate debug information for our verified program so that we can optimize them further. In this project you will learn how to use LLVM to generate debug information so that profilers and debuggers can be used with our verified toolchain.

  • Predicting team performance: Developing a predictive model using machine learning (Yeray Barrios Fleitas)

    Supervisors: Yeray Barrios Fleitas

    Keywords: Team Tuckamn model, Team Dynamics Assessment, Dysfunction Detection, tracking systems, AppScript

    Are you ready to dive into predictive modelling and machine learning? Imagine having the power to predict which teams will thrive in a project-based learning environment and which might need extra support! This project will allow you to build a data-driven tool that helps educators improve student collaboration by identifying key success factors based on team diversity, skills, and composition. Using powerful ensemble techniques like Random Forest and Gradient Boosting, you’ll create a model that could shape the future of team dynamics in education. If you’re passionate about data scienceteam analysis, and making a real educational impact, this project is perfect for you. Along the way, you’ll sharpen your machine-learning skills, tackle real-world challenges in team prediction, and gain hands-on experience in educational data science—skills that will set you apart in the job market!

    Get more information

    You can find more information about the project (methodology, student profile, learning goals and work plan) by clicking this link

  • Prove your Theorem with an Interactive Theorem Prover (Peter Lammich)

    Supervisors: Peter Lammich
    Interactive Theorem Provers (ITPs) provide a way to develop computer checked (mathematical) proofs. In this project, you will learn to use an ITP, and apply it to formalize some interesting theory. The theory to be formalized will be chosen in coordination with the supervisor, to suit your interests and be realistically achievable for a BsC project.

  • Tracking Team Health: Developing the Tuckman Stage Metric for Monitoring Team Dynamics (Yeray Barrios Fleitas)

    Supervisors: Yeray Barrios Fleitas

    Keywords: Team Tuckamn model, Team Dynamics Assessment, Dysfunction Detection, tracking systems, AppScript

    Ever wondered how teams evolve from the chaotic forming stage to becoming high-performing units? Imagine having a tool that can monitor a team’s health at every stage, identifying when they are thriving or might need extra support. In this project, you’ll bring the Tuckman team development model to life by building a system that tracks where each team stands on their journey to success. You will explore what makes teams tick by researching the Tuckman stages and identifying the characteristics defining each. Then, using validated metrics, you will create a unique scoring system to measure each team member’s position on the curve and combine these into an overall team health score. Plus, you will design ways to detect early signs of dysfunction, helping teams stay on track and educators intervene at just the right time. If you’re passionate about team dynamics, data analysis, and building meaningful tools, this is your project! Not only will you gain hands-on experience with real-world data, but you’ll also develop skills in monitoring systems and team health analysis—skills that are highly valued in tech, consulting, and HR. Ready to make an impact on how teams work and grow? Let’s get started!

    Get more information

    You can find more information about the project (methodology, student profile, learning goals and work plan) by clicking this link 

Contact