In Pune, currently we have a phase where the DCP is keen on removing / banning right turns to improve flow of traffic. The problem junctions are on two major roads (Ferguson College Road and JM Road) and can be seen in this google map symbolised by a blue question mark and red caution traingle respectively - click here for the map. The plan proposed by the DCP is to put a divider at the junctions and convert the road in to a dual carriage way.
Is this a good idea? Will it work? Does it work? Lot of questions -
I thought I will share some relevant information in this regard with everyone and hope to get more ideass and views on this from everyone on this group.
This is a very complex issue. In fact there are dozens of examples where right turns have been removed to improve safety as well as traffic flow. In short, the DCP may have the right idea and partly a correct approach (in that he is ready to seek public consultation) - but there is no evidence as yet to show that the idea is based on scientific thinking, until such time, its fair to consider this as an ad-hoc measure.
Essentially the equation is demand / capacity. The solution thus is to reduce demand or increase capacity.
Typically, demand is reduced by way of congestion charging, road pricing, tax on vehicles, removal of subsidy (on petrol), limiting parking and removing paid parking. Sometimes the equation is tilted by prioritising buses, high occupancy vehicles and NMT.
Typically, capacity is enhanced by more road building. And we have dozens of case examples making this evidently an useless option. The term used to explain the failures is 'induced congestion'.
Now there can atypical measures of changing the equation. Within the UK health service we had a crisis a short while ago over waiting times/lists. Initially this was attributed to lack of manpower. Root cause analysis showed that it was to do with queue management. Now take our railway ticket booking counters - we have 10 windows and 10 queues. But try remembering going to a Bank in the Western world, they have 10 counters but one queue!! Which works better? Research shows the latter works better.
This applies to our roads when we decide to use road space optimally. Recently I circulated a link to an article describing giving an experimental explanation of how traffic jams happen even without bottle necks. This is simply because every road has a given capacity for a specified speed limit. Motorways have a sped limit of 70 miles per hour, but at that speed, stopping distances / 2 second gap is 315 feet. When more cars exceed capacity but attempt the above speed, chaos begins, usually there is a rear end collision and this takes out a lane causing further delay. To improve reliability and also get more people to their destinations safely and quickly, motoerways now have two implements - 1) Ramp / slip road meters (signals) - these will restrict number of vehicles on motorway at a given point in time. 2) Variable speed limit - thus when it starts getting saturated, the speed limit is dropped to 50 (or even 40), this reduces the 2 second gap to 175 or 120 feet respectively, allowing more cars to travel reliably.
Now a motorway connects 2 cities with no habitat in between, hence traffic can be handled more easily.
The city is rather different. Its a given that after every 100 yards there will be a cross road or a T junction. Our planners need accepting this fact. Because right now all they seem to want to do is try and create a flow which is non-stop (akin to motorways) - to do this they use flyovers, subways, viaducts, underpasses. But the DCP is using 'ban the right turn' method. This works when - traffic is diverted away from a busy segment to an unused or underused segment. Most people think if they take a bigger arterial road they will get faster to their destination - this is not true (and we will all have experienced this). In the process, the majority try and use the arterial road and saturate it but smaller parallel feeder roads are oft lying vacant/under-used.
The solution then can be like how the banks manage their queus, by removing a certain road, pedestrianising it or in this instance, creating one way gyratory systems or banning right turn to divert some of the vehicles to unused sections - a detour, but overall traffic flow improves, journey becomes reliable and people reach their destinations more quickly (I suspect this also has a positive impact on pollution). Sometimes, this is done by sync of signals - a method used is phasing signals - i.e. green at junction one, the next junction becomes green by the time cars reach it and so on. But by phasing, what they mean is that junctions 2,3 and 4 will have incremenatally longer green phase to improve throughput. Often when traffic is made to detour on smaller unused road to make the most of available capacity, this raises concerns over pedestrian safety - hence these measures may have to be used with traffic calming methods.
But it can all go very wrong when traffic is diverted from one junction on to another busy road or busy junction - this is like 'aapne angan ka kachra dusare ke angan me dala'.
Thus, There is nothing wrong in the method as long as -
- Pedestrians are not compromised (stop right turn with dividers if need be but provide staggered crossings)
- If the overall flow is improved.
My real problem with the proposal is -
- I worry pedestrians will not be a priority
- I am not sure traffic flow will improve -
- By removing junction at Kamat on JM road, he has created a detour of sorts, but to where? If vehicles will be allowed a U / Right turn at at McDonalds just 150 m further down the road - will this work at all?
- Then perhaps we must let traffic move straight on connect with FC only via Alka Chowk/Garware
- In any case, a detour will put pressure on FC, and those taking a detour will want to eventually get to JM
- If the Tukaram Paduka Chowk right turn is removed - where and how do vehicles get to JM?
- Thus my worry is the DCP may improve situation at Gandharva chowk but because he is diverting traffic to other busy sections and junctions rather than under-used sections of the local road network - he will have chaos somewhere else.
- What will he do then, stop right turn at next junction? Propose another flyover?
Here is a nice article on A Review of Queue Management Strategies - http://www.its.leeds.ac.uk/projects/primavera/p_queue.html
The other thing is, he has not even finished dealing with junction, why is he going for another?
Also, has he got baseline stats to compare with post-trial period? This link partly shows what indicators should be looked at when overhauling a system http://www.its.leeds.ac.uk/projects/primavera/bpm.pdf
ASJ
Forward this Blog entry
Comments
Traffic flows how to improve
Reference my article which talks about visualizing similarities of City Traffic with examples in nature. I have compared the growth of a city to that of a tree in here
That is an interesting article you have referred:
“ This applies to our roads when we decide to use road space optimally. Recently I circulated a link to an article describing giving an experimental explanation of how traffic jams happen even without bottle necks”
However you go about pseudo solutions to improve which cannot be improved as per your own premise.
My request is Please determine the limit of no of vehicles you can reasonably accommodate in your cities existing or planned infrastructure, than getting lost in rambling thoughts.
Speed limits affect road capacity
www.driving-india.blogspot.com
Number of vehicles that a city can tolerate
ASJ
I agree with your calculation to find the capacity of a road. You rightly pointed that this figure varies with the assumption you make about the speed limit.
Any design or planning is dependent on the ingenuity of the designer to figure out how best he can make these assumptions. However I do not agree with your statement
“Problem is I doubt our planners ever had a clue and thought about these things when making roads, flyovers etc.”
The planners should be having qualified advisors. They are not the ones who actually make the roads. In any case best design proposals should be selected from a set of competitive designs.
There are some gross estimates which can be made to get at maximum minimum figures relating to the design. For example how many cars can be parked fender to fender, in all the roads of a city? It is easy to calculate right? It is equal to total area of all roads divided by average car area. What is the limit of registered cars a city should have? Again a gross estimate would be a suitable fraction (or is it a multiple) of the above car numbers limit. There is a limit to number of autos allowed in Bangalore. I do not have it at my finger tips. Is there a limit to the number of cars that can be registered in a city? If there is none we should have it
The point I make is very simple. We want the cities comfortable to live in. Let us be aware of the main threats against this wish of ours. I try and put some
1. The population of our country increasing alarmingly
2. The Caliber of the average citizen.
3. The rich and poor divide
4. The craze like "think big to become big".
5. lack of quality thinking
6. Keep adding please!
I invite the readers to read This article too. Probably it may be of interest!
Banning U turn on Bannerghatta Road has lead to Chaos