Fire Evacuation Drill Game Generated from Real Building Floor Plans
A serious game that turns a school, office or hospital floor plan into a playable 3D building where staff and students practise fire evacuation: finding exits, avoiding smoke, helping others and meeting at the assembly point. An intermediate final year project joining level generation from plans, pathfinding and training analytics for safety officers.
Intermediate
A small team, or one strong student willing to learn something new.2 semesters · 3 students
7 to ship
3 optional extrasSuggested stack
What you should ship
- A floor plan import tool that takes a vectorised or traced plan (walls, doors, stairs, exits) and generates a walkable 3D level with collision and navigation mesh.
- A drill mode where a player starts in a random room, a fire starts at a random location, and smoke spreads room by room over time, blocking some routes.
- Signage, fire doors, extinguishers and the assembly point placed from the plan, with the ability to mark exits as blocked for a scenario.
- Scoring on time to reach the assembly point, route choice compared with the shortest safe route, and mistakes such as using a lift or re-entering the building.
- A route replay heatmap over the floor plan showing where players went, hesitated or turned back.
- A safety officer view that aggregates drills across a group and shows rooms where people most often choose unsafe routes.
- A study with at least 15 occupants of the real building comparing their game performance with their knowledge of exits before playing.
If you have time left
- Non-player occupants who evacuate using simple crowd rules, creating congestion at stairs.
- A mode for fire wardens that tests sweeping rooms and reporting to the assembly point.
- Automatic wall and door detection from a scanned floor plan image using computer vision.
The problem: fire drills that teach one route
Fire drills in schools, offices and hospitals are infrequent, disruptive and predictable. Everyone walks out the usual stairs at a known time. Nobody practises the case that matters: the usual route is full of smoke and you are in an unfamiliar part of the building. Visitors, new staff and students often do not know where the second exit is. Safety officers know this but can rarely run realistic drills with blocked routes.
Buildings in growing cities, including many in Pakistan, have been converted and extended over time, so exits are not always where people expect. Practising in a copy of the actual building, rather than a generic one, is what makes training transfer.
What already exists and the gap
- Pathfinder by Thunderhead Engineering simulates building evacuation for fire engineers designing buildings, not for training occupants.
- AnyLogic's Pedestrian Library models crowd movement for planning and analysis.
- XVR provides virtual reality training for fire and emergency responders.
- Generic fire safety games and VR demos teach extinguisher use or show a standard office.
The gap is occupant training in a copy of your own building, generated from its floor plan, with scenarios where routes are blocked and analytics that tell the safety officer which rooms are risky. Engineering simulators model people; this project trains them.
What you build
Safety officers import a floor plan by tracing walls, doors, stairs and exits in a simple 2D tool (or from a vector file), then place fire equipment and the assembly point. The generator builds a 3D level, one floor or several, with a navigation mesh. Players (staff, students) start in a random room when the alarm sounds, with smoke spreading from a random origin. They must find a safe route out and reach the assembly point. After each drill they see their route on the plan next to the shortest safe route. The officer view shows results across all players.
The technically interesting part
Two problems make this FYP-worthy. First, level generation from plans: turning 2D line data into walls with door openings, stairs connecting floors and a correct navigation mesh, reliably enough that a messy traced plan still produces a playable building. Second, the dynamic safe-route problem: as smoke spreads, the shortest safe route changes, so you need a graph of rooms and doors with time-varying costs, a way to compute the best route at each moment, and a fair comparison between what the player did and what was possible given what they could see.
Show evaluators a plan imported live, the smoke spreading, and the route comparison screen.
How to evaluate it
- Generation: time to import a floor from plan to playable level, and a count of defects (missing doors, unreachable rooms) found by an automated reachability check.
- Pathfinding: correctness of the safe route against hand-checked cases, and computation time per update.
- Training value: at least 15 real occupants take a short exit knowledge quiz, play three drills, then repeat the quiz; report the change and time-to-exit improvement across drills.
- Officer usefulness: interview the building's safety officer about whether the heatmaps revealed anything new.
Suggested plan
- Weeks 1 to 4: get permission and a floor plan for one real building; design the plan data format.
- Weeks 5 to 9: tracing tool and 3D level generator for one floor, with navigation.
- Weeks 10 to 14: player controller, alarm, exits and assembly point; first playable drill.
- Semester 2, weeks 1 to 5: smoke spread, blocked routes, multi-floor stairs, safe-route computation.
- Weeks 6 to 9: scoring, replay heatmap, officer view.
- Weeks 10 to 14: study with occupants, report and demo.
Path to a product
The first customers are school chains, universities and hospitals that must run drills anyway and want evidence of preparedness. Charge a setup fee per building for import and a yearly subscription per building for drills and reports. Fire safety consultancies are a strong channel since they already visit buildings and advise on drills.
Scope warning
Do not build realistic fire physics, a VR version or a full crowd simulation. Do not promise automatic plan reading from any scan; manual tracing is fine for the core project. Do not publish detailed plans of real buildings; keep them private to the owner.
Learn what you need on Sefism
- Unity Manual and Scripting API : NavMesh, procedural meshes and scene building.
- Red Blob Games : pathfinding and graph search explained with interactive examples.
- DSA roadmap : graphs, shortest paths and dynamic costs.
- Game Programming Patterns : update loops and spatial partitioning for smoke spread.
- Data analytics cheat sheet : summarising drill results for the safety officer.
Get help with this idea
Turning real plans into levels is fiddly, so agree early how much of the import is automatic and how much is traced by hand. Bring your proposal and we can agree the scope, pick a stack your team can finish with, and design an evaluation your panel will respect. You build it; we help you make the calls.
New here? Create a free account first.
Ideas and guidance, not finished projects
These are project ideas and scoping guidance, published free for students to use as a starting point. I do not build, write, or sell final-year projects, and I do not complete coursework for anyone. Take an idea, make it yours, and build it.
Not sure this one fits you?
I help students pick an idea that matches their skills and their deadline, then scope it down until it is finishable. Guidance only, never done for you.
Stuck on something specific?
Writing only gets you so far. If you want an answer to your situation rather than the general case, book a session and we will work through it together. Sessions are free for approved Sefism members, and a few slots open each week.
Follow along
New writing, resources and project ideas land here first.