AI-Supported Escape Room Tasks in Mathematics for Engineering

Invited Workshop Recommended citation: Dias Rasteiro, D. M. L., Suleiman, B., & Kaneko, Y. (2025). AI-Supported Escape Room Tasks in Mathematics for Engineering. In Kangaslampi, R., Langie, G., Järvinen, H.-M., & Nagy, B. (Eds.), SEFI 53rd Annual Conference. European Society for Engineering Education (SEFI), Tampere, Finland. DOI: 10.5281/zenodo.17631736. This Conference Paper is brought to you for open access by the 53rd Annual Conference of the European Society for Engineering Education (SEFI) at Tampere University in Tampere, Finland. This work is licensed under a Creative Commons Attribution-NonCommercial-Share Alike 4.0 International License. AI-Supported Escape Room Tasks in Mathematics for Engineering D. M. L. Dias Rasteiroa, 1 , B. Suleimanb, Y. Kanekoc aPolytechnic University of Coimbra (IPC/ISEC), Coimbra, Portugal, 0000-0002-12286072 b University of New South Wales, Sydney, Australia, 0000-0003-2674-0253 cSabanci University, Istanbul, Turkiye, 0000-0002-1861-5703 Conference Key Areas: Engineering Education Keywords: Mathematics Education, Artificial Intelligence, Escape Rooms, Active Learning, Hands-on, Game-based Learning ABSTRACT This workshop introduces participants to the design of pedagogically rich, AIsupported escape room tasks for Mathematics in Engineering Education. It aims to bridge the gap between abstract mathematical concepts and real-world applications by leveraging generative AI tools (e.g., ChatGPT) to create engaging, contextualised puzzles rooted in engineering scenarios. Participants will explore how escape rooms, when enhanced with AI, can foster active learning, critical thinking, and collaboration among students. The session goes beyond theoretical discussion, offering hands-on experience in developing instructional materials that include theoretical content, worked examples, prompt engineering, and classroom-ready Beamer slides. Attendees will leave with adaptable resources ready to implement in their own teaching, and a shared folder will be provided for ongoing exchange and refinement. This workshop contributes to the digital transformation of Mathematics Education, equipping educators with innovative tools to meet the evolving expectations of engineering students. 1 BACKGROUND RATIONALE AND RELEVANCE Mathematics in Engineering Education is undergoing a significant shift, not only due to digitalisation but also because of students' changing expectations about engagement and interactivity. Escape rooms – especially when worked with AIgenerated tasks – represent a powerful active learning approach that aligns with these expectations. They combine problem-based learning, critical thinking, collaboration, and gamification in a structured narrative environment. Using generative AI (e.g., ChatGPT, Deepseek) to create and refine puzzles brings a new layer of co-design into the pedagogical process, empowering teachers to focus on concept alignment. At the same time, the AI helps with formulation and variety. This approach supports 1 Corresponding Author D. M. L. Dias Rasteiro [email protected] differentiated instruction and motivates students through immersive contexts tied to their engineering fields. The workshop makes the following key contributions to Engineering Mathematics Education: 1. Pedagogical Framework: A structured model for integrating AI-supported escape rooms into mathematics courses for engineering students, aligning cognitive and motivational learning objectives. 2. Design Templates: Ready-to-use, editable materials, including AI prompt templates, student/teacher task packs, and Beamer slides, for immediate classroom implementation. 3. AI Co-Design Protocol: A transparent workflow outlining how educators can responsibly engage AI tools for task generation, including verification, error checking, and audit-trail documentation. 4. Case Studies: Two fully developed exemplars (ODE and Integral applications) demonstrating contextualised, engineering-linked mathematical challenges and their corresponding escape room mechanics. 2 WORKSHOP MOTIVATION AND LEARNING OUTCOMES This workshop aims to explore the integration of Artificial Intelligence in Engineering Mathematics Education through the design of pedagogically rich digital escape room tasks. This year, moving beyond a general discussion of ICT tools and addressing the need for a common platform where Mathematics teachers can share their materials, the focus will be on the hands-on creation of AI-enhanced mathematical puzzles that teachers can adapt to their courses. The goal is twofold: to demystify the use of AI in the classroom and to provide participants with tangible, ready-to-use materials for teaching core mathematical concepts (such as integrals, algebraic systems, or statistical methods). Attendees will leave the session with ready-made templates and contextualised examples they can immediately pilot in their classes. 2.1 Target audience • Higher education faculty & instructors • STEM educators & curriculum designers • EdTech enthusiasts & instructional designers • Anyone interested in innovative math education 2.2 Significance for Engineering Education The hands-on creation of AI-supported mathematical puzzles contributes to a consistent and innovative practice of Mathematics teaching in Engineering degrees. It empowers teachers to create meaningful experiences that help students connect abstract concepts (such as integrals, matrix transformations, probability distributions) with realistic scenarios from Mechanical, Electrical, or Informatics Engineering. Realworld examples contextualise learning, enhancing students' ability to transfer theoretical knowledge to practical situations they will encounter in their future careers. Additionally, the workshop will highlight the need to develop students’ critical thinking when engaging with AI-generated content, incorporating this competence into the activities students will work on. Participants are encouraged to guide their students to question the assumptions, verify the solutions, and improve the clarity of the AI’s output, rather than passively accepting it. This reflective process not only strengthens mathematical reasoning but also fosters digital literacy, promoting the responsible and informed use of AI tools in academic and professional settings. During the workshop, participants will produce complete instructional materials that include theoretical content, illustrative examples, AI prompts for generating task variants, complete worked solutions, and Beamer presentations. The workshop's first author will furnish initial examples to serve as a kick-off. These materials will be immediately applicable in class, adaptable to various engineering contexts, and available in a common folder that will be shared with all participants. 3 WORKSHOP DESIGN The workshop includes the following elements: • A short presentation and scenario-setting: motivation, background, goals (5 minutes) • An example of an AI-generated escape room puzzle in Calculus (e.g., finding the area between curves in a mechanical system) (10 minutes) • Hands-on task: participants create or adapt their escape room puzzle using a shared template and prompt suggestions (30 minutes) • Peer discussion and refinement in pairs or groups (10 minutes) • Wrap-up: sharing outcomes and takeaway materials (5 minutes) The workshop is designed to be highly interactive, encouraging active participation through hands-on group work, peer discussion, and immediate application of concepts. Rather than passively receiving information, participants will collaboratively engage in designing escape room puzzles using AI-generated prompts, analysing and refining their outputs in real time. Breakout moments for feedback and pair-based reflection foster a dynamic learning environment where attendees co-construct knowledge, critically evaluate AI responses, and exchange pedagogical strategies. This format ensures that each educator not only understands the methodology but also leaves with a sense of ownership over the materials created. 4 WORKSHOP RESULTS At the end of this workshop, participants were expected to have developed a comprehensive and adaptable set of teaching materials that could be directly used in their Mathematics courses. These materials would include a contextualised escape room narrative grounded in an engineering application, the associated mathematical problem (e.g., integration, algebraic systems, or statistics), theoretical background content, a fully worked solution, and AI prompts for generating new variants. Additionally, each participant will produce a classroom-ready Beamer presentation, designed to visually support the delivery of the activity and foster student engagement. Expectably, the workshop would also enhance participants’ pedagogical repertoire in three key areas: (1) critical use of AI tools for content generation and analysis, (2) effective integration of active learning methodologies in Mathematics Education, and (3) the capacity to contextualise abstract mathematical ideas within real-world engineering scenarios. Participants would leave with ready-to-use materials and a deeper understanding of how to utilise AI to promote problem-solving, critical thinking, and motivation among engineering students. As expected, the workshop positioned generative AI as a co-design partner to accelerate the production of mathematically non-trivial, engineering-contextualised ER tasks, while retaining rigorous human verification. The design emphasises clarity of task statements, explicit worked solutions, and quality checks (intervals, signs, units, and code derivability). Core mathematical attention included applications of definite integrals and first-order linear differential equations, delivered through authentic engineering scenarios that encourage collaboration, model-based judgments, and reflective debriefs. 4.1 Participant-Produced Exemplars 4.1.1 Case A — Horde Balancer (First-Order Linear ODE) Context. A game-design scenario in which enemy spawn (𝜆) and removal (𝜇) rates yield a population 𝑁(𝑡) governed by 𝑑𝑁 𝑑𝑡 = 𝜆−𝜇𝑁. The task aims to develop a solution by integrating the factor, equilibrium reasoning (𝑁∗= 𝜆/𝜇), transient dynamics, and time-to-target calculations. Student Handout (extract): Game-Design ODE Puzzle – Horde Balancer Scenario: In a survival game, the level designer wants a controlled “horde” mechanic: enemies spawn continuously from portals while existing enemies are removed (killed by players or despawned). The number of active enemies at time t is 𝑁(𝑡).Two parameters tune the system: • 𝜆 = constant spawn rate (enemies per minute), • 𝜇 = per-enemy removal rate (per minute). The game’s wave control panel will only unlock when the in-game enemy count has stabilised close to its intended steady value. Your task is to calculate the population dynamics and determine the time the designer must wait to reach a desired fraction of the steady-state population using 𝑑𝑁 𝑑𝑡 = 𝜆−𝜇𝑁, with 𝜆 > 0,𝜇 > 0. Initial condition: 𝑁(0)= 𝑁0. For this puzzle: 𝜆 = 5 enemies/min, 𝜇 = 0.5 𝑚𝑖𝑛−1,𝑁0= 0. Teacher Pack (key steps): Using the integrating factor 𝑒𝜇𝑡, the general solution is 𝑁(𝑡)=𝜆 𝜇+(𝑁0−𝜆/𝜇)𝑒−𝜇𝑡 N(t). For 𝜆 = 5,𝜇 = 0.5,𝑁0= 0:𝑁∗=10 𝑎𝑛𝑑 𝑁(𝑡)=10(1−𝑒−0.5𝑡). To reach 90% of the equilibrium, solve 1 − 𝑒−0.5𝑡 = 0.9, giving 𝑇 = −2ln(0.1)≈ 4.61 (minutes). Interpretation highlights the role of 𝜆 in setting the level and 𝜇 in governing responsiveness. Suggested escape room mechanics: lock code = rounded 𝑡 in seconds; clue ladder = (1) balance idea, (2) integrating factor, (3) 90% condition. 4.1.2 Case B — Stadium Solar (Integral of Power Curve; Energy Accounting) Context. A stadium-roof PV array with power 𝑃(𝑡)= 120 ∗ sin(𝜋𝑡 12) 𝑖𝑛 𝑘𝑊,0 ≤ 𝑡 ≤ 12 hours after sunrise; a football match window lies within the day and requires approximately 500 kWh. Students compute total/interval energy via definite integrals, interpret units, and discuss sustainability options for any surplus. Student Handout (extract): Exercise: Solar Power for a Football Stadium. A football stadium has installed solar panels on its roof to support sustainable energy use. On a sunny day, the power output of the panels (in kilowatts) can be modelled by the function: 𝑃(𝑡)= 120 ∗sin(𝜋𝑡 12),0 ≤ 𝑡 ≤ 12 , where 𝑡 is the number of hours after sunrise. The stadium consumes around 500 kWh of electricity during a match. Tasks 1. Compute the total amount of electricity. Teacher Pack (results): Daily energy 𝐸𝑑𝑎𝑦 = ∫120 ∗ sin(𝜋𝑡 12) 𝑑𝑡 = 2880 𝜋≈916.7 12 0 kWh. Match-window energy (e.g., 𝑡 ∈ [4,10]) evaluates to 𝐸𝑚𝑎𝑡𝑐ℎ =720(√3+1) 𝜋≈ 626.1kWh. Relative to a 500 kWh requirement, the surplus is about 126 kWh, motivating a brief discussion of storage, load shifting, EV charging, or grid export. Suggested escape room mechanics: lock code = ⌊𝐸𝑚𝑎𝑡𝑐ℎ⌋=626; clues emphasise “energy = area under the power curve” and exact forms before rounding. 4.2 Alignment with the Workshop’s Slides Framework The exemplars conform to the workshop’s Teacher/Student dual-pack specification, include explicit derivations and interpretive checks, and incorporate escape room mechanics (codes, hints, timeboxing, scalable variants). Both tasks satisfy the quality checklist used in the session. Slides and complete work developed during the workshop can be accessed at https://padlet.com/deolindarasteiro/msig-sefi2025workshop-unmmu4xwo7cy1l17. The readers, especially the mathematics higher education teacher’s community, are invited to share their work and use this Padlet as a common resource platform. Mathematical outcomes: correct setup and execution of integrals/ODEs; parameter interpretation; model validation. Transversal outcomes: collaborative reasoning, critical verification of AI-assisted drafts, and succinct mathematical communication. Analytic rubric (0-2 points each; total 8): • Setup correctness (model, variables, and assumptions). • Calculus execution (methods, algebraic accuracy, exact forms). • Graphical/model reasoning (units, limits, sanity checks). • Code derivation and explanation (traceability from maths to lock code). Rapid formative checks: a 2-minute peer cross-check of the final code and the last derivation step; one ‘sanity check’ question per team (e.g., “Is 𝑁∗ dimensionally consistent?”). Reflection prompt: “Identify one assumption your team made and describe how altering it would change the code.” Maintain an audit trail of AI involvement (initial prompt, AI draft, human verification and edits). Include a one-line AI contribution note in the Teacher Pack (e.g., “Initial narrative generated with GPT; mathematics verified and edited by instructors”). This practice encourages transparency and reinforces methodological rigour in AIsupported design. 5 DISCUSSION The workshop demonstrated that AI could serve as a powerful co-designer in developing educational materials, accelerating idea generation and variation while maintaining educator oversight. The workshop’s participants noted improved confidence in adapting AI-generated material and a stronger awareness of the need for verification, transparency, and contextualisation. Furthermore, integrating escape-room mechanics (locks, clues, sequential reasoning) improved perceived student engagement and comprehension of applied mathematics. However, while generative AI enhances productivity, it introduces challenges. This includes ensuring mathematical rigour, maintaining alignment with learning objectives, and mitigating biases or inaccuracies in AI outputs. The collaborative setting of the workshop allowed these issues to be surfaced and constructively addressed. 6 LIMTATIONS AND FUTURE WORK This study is limited by its exploratory, workshop-based nature and the absence of long-term data. In particular, the workshop did not track how students’ learning or engagement changed over time after using the AI-supported escape room activities. The evaluation primarily relied on qualitative feedback from participants rather than systematic classroom trials. As future work, it could be helpful to focus on empirically evaluating AI-generated escape room tasks in real classroom settings to measure their impact on learning and motivation. In addition, it could be interesting to explore automated analytics to assess task complexity and student reasoning, extend the approach to interdisciplinary STEM contexts, and develop ethical frameworks to mitigate bias in such AI-assisted educational design. REFERENCES Díaz-Ramírez, J. (2020). Gamification in engineering education: An empirical study. Heliyon, 6(9), e04972. https://doi.org/10.1016/j.heliyon.2020.e04972 Freeman, S., Eddy, S. L., McDonough, M., et al. (2014). Active learning increases student performance in science, engineering, and mathematics. Proceedings of the National Academy of Sciences, 111(23), 8410–8415. https://doi.org/10.1073/pnas.1319030111 Gruber, M., Ziefle, M., & Janson, A. (2025). Digital educational escape game design for STEM higher education. Frontiers in Education, 10, 1497291. Hernandez-Martinez, P., & Keane, T. (2024). Learning mathematics and its relevance through a digital storytelling assessment task at university. International Journal of Mathematical Education in Science and Technology, 1-17. Walters, L., Green, M., Goldsby, D., & Parker, D. (2018). Digital storytelling as a problem-solving strategy in mathematics teacher education: How making a math-eo engages and excites 21st century students. International Journal of Technology in Education and Science, 2(1), 1-16. Pinto, C.M., Nikolopoulos, C., Cruz, M., Queiros, R. and Petridis, K. (2024) Empowering engineering students’ math literacy with digital escape rooms. In Proceedings of the 2024 IEEE World Engineering Education Conference (EDUNINE) (pp. 1-6). Ortiz, M, Chiluiza, K., Valcke, M. (2016) Gamification in Higher Education and Stem: a systematic review of the literature. In Proceedings of EDULEARN16 Proceedings, (pp. 6548-6558). Rasteiro. D. M. L. D. et al. (2024) "Integrals Applications: A STEAM Activity to Teach/Learn Mathematics in Higher Education," 5th International Conference in Electronic Engineering, Information Technology & Education (EEITE), Chania, Greece, 2024, pp. 1-5, doi: 10.1109/EEITE61750.2024.10654423. Veldkamp, A., van de Grint, L., & Knippels, M.-C. P. J. (2020). Escape education: A systematic review on escape rooms in education. Educational Research Review, 31, 100364. https://doi.org/10.1016/j.edurev.2020.100364