Talk:Biham–Middleton–Levine traffic model

Comment

There seems to be something missing in the last paragraph of this article: It would be ideal, however, to formulate a rigorous (something to) predict the end result of any starting position, especially in the intermediate phases. — Kpalion^(talk) 20:09, 17 December 2010 (UTC)

• Thanks for pointing out that mistake! I fixed it now, I think. Purpy Pupple (talk) 04:41, 18 December 2010 (UTC)

Another Implementation

I have made a live wallpaper for Android that implements this traffic model, available at Cellular Life Wallpaper. I don't know if this would be appropriate for the External Links section, or a conflict of interest since I made that app. Hustvedt (talk) 19:00, 14 December 2012 (UTC)

Is the torus model appropriate?

If this model is used to simulate some sort of reality (traffic flow, whatever), then how realistic is it to have points wrap around? Would it not be more appropriate to have the points drop off the "end", whilst new points are created (by your algorithm du jour) at the beginning? 101.174.196.224 (talk) 03:12, 15 December 2012 (UTC)

I'm sticking to my guns here. I challenge the concept of the wrap-around. In what universe are you going to have cars or pedestrians or whatever go through an intersection or interchange or whatever and then immediately go around the other side? That's nonsense.

As a separate but related point, if you do wrap around, then your final conditions for one iteration become the initial conditions for the next iteration. This gives you very limited and very specific initial conditions, which is also nonsense, or at best, highly restrictive.

A model with proper "flow" (that is, traffic dropping off at the "end" and being created (eg, randomly) at the "beginning") makes more sense. It also seems, within the limitations of the model, to simulate reality more closely. In any case it certainly gives more interesting results.

On a totally different note, having the red/blue points take turns is also unrealistic, or at least limited in its simulation of reality. Far more realistic is that both red and blue move at the same time, with some sort of conflict resolution algorithm applied when a red and a blue both want to move to the same blank spot. Again, this seems to more closely limit reality.

This is, of course, all OR, so none of this can appear in the main article. But it seems useful to include as a discussion in a talk page. If anyone is at all slightly interested I can post screen captures of the different models to illustrate these points. Old_Wombat (talk) 23:33, 16 December 2012 (UTC)

I agree that the model sacrifices realism for simplicity. I imagine that the torus model is used so that:

1. The system is deterministic. If you keep adding new random cars and deleting old ones, the system would not be deterministic, which makes analysis harder.
2. We can analyse the steady state tendencies (e.g. a global jam).

But your proposed model with traffic disappearing on two edges and being created on two edges is very interesting. I am interested in seeing your screen captures. As for letting the red/blue points move simultaneously, I believe this has been done in other traffic models that are not called the Biham-Middleton-Levine traffic model. Overall the field of traffic models is still an area of ongoing research. Purpy Pupple (talk) 00:55, 24 December 2012 (UTC)

File:Biham-Middleton-Levine traffic model self-organized to a periodic intermediate phase.ogv +1 to appear as POTD soon

Hello! This is a note to let the editors of this article know that File:Biham-Middleton-Levine traffic model self-organized to a periodic intermediate phase.ogv and File:Biham-Middleton-Levine traffic model self-organized to a disordered intermediate phase.ogv will be appearing as picture of the day on September 28, 2016. You can view and edit the POTD blurb at Template:POTD/2016-09-28. If this article needs any attention or maintenance, it would be preferable if that could be done before its appearance on the Main Page. — Chris Woodrich (talk) 10:12, 14 September 2016 (UTC)

Picture of the day

The Biham–Middleton–Levine traffic model for a 144 x 89 lattice, with a traffic density of 38%. The model has self-organized to a periodic intermediate phase. The red cars and blue cars take turns to move; the red ones only move rightwards, and the blue ones move downwards. Every time, all the cars of the same colour try to move one step if there is no car in front. This video has been sped up such that only one in four frames is shown.

See the disordered intermediate phaseFilm: Dllu

File:Biham-Middleton-Levine traffic model self-organized to a globally jammed phase.ogv +1 to appear as POTD soon

Hello! This is a note to let the editors of this article know that File:Biham-Middleton-Levine traffic model self-organized to a globally jammed phase.ogv and File:Biham-Middleton-Levine traffic model self-organized to a free flowing phase.ogv will be appearing as picture of the day on August 19, 2017. You can view and edit the POTD blurb at Template:POTD/2017-08-19. If this article needs any attention or maintenance, it would be preferable if that could be done before its appearance on the Main Page. — Chris Woodrich (talk) 01:41, 8 August 2017 (UTC)

Picture of the day

The Biham–Middleton–Levine traffic model for a 144 x 89 lattice, with a traffic density of 28%. The model has self-organized to a free-flowing phase. The red cars and blue cars take turns to move; the red ones only move rightwards, and the blue ones move downwards. Every time, all the cars of the same colour try to move one step if there is no car in front. This video has been sped up such that only one in four frames is shown.

See the globally jammed phaseFilm: Dllu