Mall Entrance
People go in and out of a building through doors, turnstiles or a revolving door, in the social force model.
- Going in
- Going out
- Door wing or turnstile bar
- Wall
Blue shoppers come from the street at the bottom and go in; orange people leave from the top. Each picks the door with the shortest expected wait. Drag with Erase across the facade to cut a new door, or with Wall to draw walls; Rerun starts again with the plan as drawn.
How is Mall Entrance working for you?
About Mall Entrance
Mall Entrance simulates people going in and out of a building through the doors in its facade. Shoppers arrive on the street side and head inside while others leave toward the street, all walking with the social force model of Dirk Helbing and Péter Molnár: each person steers toward their goal at a preferred speed, keeps a distance from others and from walls, and looks one second ahead to step aside.
Doors are bottlenecks, and two-way traffic through them organizes itself. Helbing, Lubos Buzna, Anders Johansson and Torsten Werner reported in 2005 that at a narrow door the passing direction takes turns, and that where there are several doors, each tends to be used by one direction. Here each person picks the door with the least expected time: the walk to it plus everyone who must use it first, with people coming the other way counted twice, divided by what the door can pass (best-response door choice, as studied by Harri Ehtamo and colleagues). Busy doors then shed people to quieter ones.
Compare one wide door, several narrow doors, separate doors for going in and going out, a revolving door whose four wings turn four times a minute, and turnstiles that let one person through at a time. Set how many people go in and out per minute, draw walls or cut new doors, and read the throughput and mean wait at each door.
How to use Mall Entrance
- Pick the doors. One wide door is a 3 m opening; Several narrow doors are four 1 m doors; Separate entry and exit gives each direction its own 1.5 m door with a wall between the queues; Revolving door turns four wings inside a drum; Turnstiles have two gates for going in and two for going out. Opening time sends many people in and few out through the narrow doors.
- Set the crowd. Going in sets how many shoppers arrive from the street per minute, Going out how many leave from the inside. Blue people go in, orange people go out.
- Read the doors. Under the canvas each door has its throughput in people per minute over the last minute, how many people went in and out through it, and the mean wait of its last 30 passes. The wait is the time from 3 m before the facade to passing it, beyond what walking that far takes. On the Through the doors chart the dashed line is the demand, Going in + Going out: the steady throughput equals it while every arrival gets in.
- Compare runs. At 120 s each run adds a row to the Runs table: the throughput over the last minute, the mean wait and each door's throughput and wait, with the settings that differ between rows. Reset clears the table.
- Change the building. With Pointer set to Erase, rub out part of the facade to cut a new door; with Wall, draw walls or queue barriers. Routes and doors update when you release. Rerun starts the same arrivals again from t = 0 with the building as drawn; Restart clears your walls and doors and starts the layout again; Reset also returns every setting to its default.
- Export. Save a PNG or record a video or GIF. Record 30 s from the start runs the same entrance again from the beginning and records its first 30 seconds, frame by frame. The CSV holds throughput and mean wait every second since t = 0, and the Runs table. A share link carries the walls and doors you drew.
Frequently asked questions
Is one wide door better than several narrow ones?
For the same total width, a wide door lets people pass in both directions side by side, while a narrow door holds one person at a time. In the simulation four 1 m doors carry two-way traffic as long as people spread out and each door ends up mostly one way; push the numbers past what a door can pass and queues grow. Seyfried and colleagues measured in 2009 that the flow through an opening grows steadily with its width.
Why do people take turns at a narrow door?
Once a few people have pushed through from one side, those following them have a clear path while the other side waits, until a gap lets the other direction take over. Helbing and colleagues described these oscillations of the passing direction in simulations and in observations of real doors.
How do people choose a door?
Every second each person who has not reached the facade estimates the time through each door they may use: the walking time to it plus the number of people who will use it before them, over the door's capacity of 1.3 people per second per meter of width. People coming the other way count twice, because meeting them slows a door more. A person switches only when another door is at least 2 seconds quicker.
How many people does a revolving door let through?
In the simulation the drum is 3.2 m across and its wings turn four times a minute, so people have to fit into a compartment and walk with it. With 120 people going in and 90 going out per minute it passes about 100 per minute, well under a 3 m opening, and a queue forms on both sides.
How are the turnstiles modeled?
Each gate is a bar across a 0.75 m opening. It opens for 0.9 seconds when someone waits at it, stays open while a person is inside, and then stays shut for at least 0.6 seconds, which lets about 20 to 25 people through per minute. The two left gates are for leaving and the two right ones for going in.
Is this used to design real entrances?
Engineers use pedestrian models in this family, together with measurements and building codes, to size doors, gates and queues. This tool is an educational simulation, not a substitute for a professional design study; Crowd Evacuation covers the emergency case.
References
- Helbing, D., & Molnár, P. (1995). Social force model for pedestrian dynamics. Physical Review E 51(5), 4282–4286.
- Helbing, D., Buzna, L., Johansson, A., & Werner, T. (2005). Self-organized pedestrian crowd dynamics: experiments, simulations, and design solutions. Transportation Science 39(1), 1–24.
- Seyfried, A., Passon, O., Steffen, B., Boltes, M., Rupprecht, T., & Klingsch, W. (2009). New insights into pedestrian flow through bottlenecks. Transportation Science 43(3), 395–406.
- Ehtamo, H., Heliövaara, S., Korhonen, T., & Hostikka, S. (2010). Game theoretic best-response dynamics for evacuees' exit selection. Advances in Complex Systems 13(1), 113–134.
- Zanlungo, F., Ikeda, T., & Kanda, T. (2011). Social force model with explicit collision prediction. EPL 93, 68005.




