Explore - Experience - Excel

A Prisoner’s Dilemma in Bangalore Traffic – Prof. Rajendra Desai

30 Sept, 2022.

‘Swalpa adjust maadi’ is an oft repeated phrase on Bangalore roads from peace seeking violators of traffic rules. My response to these violators is typically ‘****  ***’ as I steer my car to narrowly miss the violator or come to a complete halt to allow the violator through.

In my commute to and from work, I navigate gridlocks at multiple intersections and a journey that should not take more than half an hour ends up taking 1.5- 2.0 hours. The BMTC bus that takes a U turn from the extreme left of a 2-way road tops the list of aggressors. In order to enter the left-most lane of the opposite side he has to do multiple reversals and the wait time can be 5–15 minutes at that intersection to make the turn!  The potholes, unscientific speed breakers, people crossing randomly…all aggravate this problem.

Prime Minister Narendra Modi during his visit to Bengaluru in June 2022, experienced this mess. He has given the BJP government in Karnataka a deadline of 40 months to decongest the traffic. The Bangalore Traffic Police have recently started a pilot project with Google to optimize the traffic light configuration.

There is hope in all these initiatives and some seem to be delivering results. A website that tracks traffic congestion reports that the average congestion levels had come down from 51% (2020) to 48% in 2021. A congestion level is a measure of the additional time that is taken over a 10 km journey due to the traffic bottlenecks. So, if a journey of 10 kms with normal traffic takes 20 minutes, with a congestion of 50% it would take 30 minutes. However, at 2-3% improvement a year we would need 16+ years to bring this index to a level that would make driving in Bangalore pleasurable.

I explore this situation using the game theory framework* of the Prisoner’s Dilemma for some insights into a possible solution.

In the classical version of the Prisoner’s Dilemma, 2 suspected bank robbers are being interrogated in separate cells. Each is given the option of testifying against the other to get off some jail time. Each bank robber is faced with the choice to cooperate with his accomplice and remain silent or to defect from the gang and testify for the prosecution.

If they both co-operate and remain silent, then the authorities will only be able to convict them on a lesser charge resulting in one year in jail for each (a total of 2 years’ jail time). If one testifies and the other does not, then the one who testifies will go free and the other will get five years (0 years for the one who defects + 5 for the one convicted = 5 years total).

However, if both testify against the other, each will get two years in jail for being partly responsible for the robbery (2 years + 2 years = 4 years total jail time).

Adam Smith’s words “It is not from the benevolence of the butcher, the brewer, or the baker that we expect our dinner, but from their regard to their own interest” suggest that pursuing a self-interest is in the long run best in everyone’s interest.  John Nash (contributor to Game Theory and a Nobel Prize winner) mathematically proved that individual interest is good only as long as it works towards the overall good. The conclusion of game theorists is, in the long run, while cooperation is the best overall solution for both prisoners, if the gains from defection are very high, both prisoners will tend to defect.

Observational data seems to indicate there is no disproportionate benefit of reaching one’s destination substantially earlier. By violating rules and all consideration on the road, the vehicle user can cut down commute time to half seems to be just wishful thinking. The offending vehicle cuts into lanes, the 2-wheeler driver fills up all available space or cuts into a queue to just get a couple of feet ahead may result in time savings only in single digit minutes.

In a single interaction, the best action for each prisoner is to defect to maximize their own gain (0 years prison time).So, if our 2 vehicle drivers believed they are on the road only on that one occasion to get to their destinations, their best bet is to throw all consideration for the other driver to the winds and ensure they get to their destination at the earliest. The result is, as a colleague from Delhi suggested one morning while we were both stuck in exasperating traffic – ‘Don’t give an inch of space to anyone especially the 2-wheeler drivers!’  

However, in the long run, the optimal strategy for both prisoners is to co-operate (not defect or affront) as in multiple interactions, non-cooperation will deteriorate into a zero-sum game. Over a longer period of time, both players will lose (in the aggregate) as in each game one or the other will win with a heavy loss to the other. For the 2 vehicle drivers, day to day driving over a longer period of time will definitely become more painful.

Hence, the recommendation for Bangalore Traffic is, everyone gains if there is more trust, cooperation and belief by all parties. For this there must be belief in being in the game for the long term.  This possibly needs a sense of belonging. With a large percentage of the population being migrant, this sense of belonging may be missing and there is a need to drive it. Does a sense of belonging come from trust, cooperation and belief in the long term or is it vice-versa? Acceptance of the present is a good step forward towards this end.

Driving a sense of belonging may need more than posters of ‘Karnataka Hemmay’ and ‘Namma Bengaluru’. As a friend very perceptively pointed out - Nationalism is hard to drive with mere singing the National Anthem in cinema halls / flying flags on Independence Day etcetra which have already turned ritualistic and devoid of true patriotism. Instead, in happy moments of National pride (wins in the Commonwealth Games) all homes raising the flag / sad moments (at half-mast) may create a stronger sense of belonging.

It may be easier to drive this at a local level with neighborhood communities that can at some point embrace the larger population of Bangalore. Till this happens, the only way for managing Bangalore’s traffic chaos is to introduce a 3rd player in this game – The Bangalore Traffic Police to enforce cooperation through strict monitoring of the traffic.

*A Game Theory framework starts out by identifying the ‘players’ involved in the game. In a ‘Bangalore Traffic Management Game’ these would be the Corporation that maintains the roads (BBMP), the traffic police, the vehicle driver, the pedestrian, the street vendor, the jaywalker…I restricted my exploration in this article to a simple situation with 2 vehicle drivers navigating traffic daily to and from work. This is akin to a 2-player game theory framework understood through the example of a prisoner’s dilemma. Multi-player games require a more complex mathematical approach and continue to be a topic of interest to researchers.

Leave a Reply

Your email address will not be published. Required fields are marked *