Posted in Operations & IT Articles, Total Reads: 1550
, Published on 30 August 2013
A very usual scene in metropolitans these days is people stuck in long queues at traffic signals cursing their luck and traffic system. During last two months, I have been driving to Noida from Ghaziabad each day. During these drives I also used to got stuck at numerous traffic signals, on several occasions I noticed that while traffic has been stopped on three sides, one side with green signal has an empty road way. My initial sense was that may be I am travelling through a higher traffic density route. I altered my route for next couple of days, only difference made by this alteration was as I reached the same signal this time, my regular way had a green signal and yet after a passing of a bunch of initial vehicles the road was idle although signal was still green. Another thing worth noting was on every successive signal a same bunch of vehicle gets stalled. That means if say a lot of twenty cars had passed through a signal, you will find that same lot of twenty cars will be waiting at next signal, with few exceptions.
Now this generates a critical operation problem, which can be defined as “while we have long queues of vehicle at one side, there is empty road on the other side of the signal at the same time”. One solution is to introduce smart traffic signals which will be RFID enabled and will calculate traffic inflow and outflow and will make the signals work accordingly, but this will cost millions and also required a good number of manpower to maintain the system. To prevent this huge investment and arrangement let us further analyse this problem from operational point of view, let us compare our traffic system with the production system of a large assembly line. Since our vehicles moves in lots, we can consider one lot of vehicles as one unit of product, every traffic signal as an assembly station and our roads can be equated to conveyors on which our products (lots of vehicles) are running through. If we now see this problem, we will find that lead time of conveyor and machining time of assembly station are not synchronised. In other words, while our unit is in way our assembly station is idle and when our unit reached the station our machine is taking time to reboot, and hence our unit is standing idle.
Coming back to our roads, our problem stands that when our signals were green our bunch of vehicle was either stalled at previous signal or was on the way and when it reached the signal, light was turned red. Now to resolve this paradox we have to first synchronize our traffic signals with each other. Let us take an example to better understand the scenario, let us assume we have four traffic lights on straight road we will name them as A,B,C,D with distance 1 km, 1.5km, 1.25km in between them respectively. Also through observation we came to know that generally vehicle run at an average speed of say 60 kmph. Now we can synchronize these signals accordingly as signal B will turn green exactly after one minute of signal A, signal C will turn green after 90 seconds of signal B and signal D will be 75 seconds ahead of signal C. In this system a vehicle running at approximately at 60 kmph would not have to stop at any of the signal B, C and D.
But this system has a major limitation that while we are trying to synchronize lights with each other we are actually neglecting a major point that at any traffic signal we have three or four ways of incoming and outgoing, therefore there are four signals perfectly synchronized with each other. If we try to synchronize them with the preceding signals, arrangement between these lights (four traffic lights at the same point) will get disturb. Basic idea behind traffic signal is that when traffic from one side is moving, traffic at rest all sides should be stopped and since we cannot change this system, hence we cannot apply our idea to all traffic signals at a single point.
To overcome above limitation, we had a simple solution. We all know that traffic density remains high on particular routes which move from residential areas to commercial areas in morning and vice-versa in evenings. So we can identify these routes and critical traffic signals on these routes and only these traffic signals should be synchronized in a sequence while moving towards the commercial area. Similarly these signals will get synchronized in reverse direction (from commercial area to residential area) in the evening, when traffic density will be high on the route from commercial area to residential area. This will not just eliminate the long queues at peak hours but also answer to energy and environmental issues, but for that we have to manage our roads efficiently and judiciously as conveyors of production systems.
The article has been authored by Vaibhav Maheshwari, LBSIM
The Daily Times (Smart traffic signals planned along East Main Street to ease heavy congestion in Farmington, http://www.daily-times.com/four_corners-news/ci_23365790/smart-traffic-signals-planned-along-east-main-street )
Real-Time Traffic Signal Control for Optimization of Traffic Jam Probability, by cheng-you cui, ji-sun shin, michio miyazaki, AND hee-hyol lee
If you are interested in writing articles for us, Submit Here