Feb 12 Sometime soon, the Bombay suburb of Bandra, where we live, will see a huge rise in traffic and the consequent congestion. The city's Coastal Road, under construction just off its western shore, is to have an exit onto Carter Road, the sea-hugging road that starts not far from where I sit writing this. The construction for this exit is already well underway and it's hard to look at it and not fear for what lies ahead.
A study done some years ago offers a rationale for the Coastal Road (really, the segment from Bandra to the northern suburbs). When I looked at the numbers it offers, I knew I needed to write a column examining them, or some of them, as best I could. This is that column, in Mint yesterday (February 12): Going by the numbers, a road to nowhere, https://www.livemint.com/opinion/columns/going-by-the-numbers-a-road-to-nowhere-11613062413417.html I really would like to hear your thoughts. Preferably, sometime before I am overrun by a stream of Passenger Car Units. yours, dilip --- By the numbers, a road to nowhere The first giveaways, please note, are large numbers that don’t end in one or more zeros. Their source: A 2012 report to estimate the traffic on Mumbai’s upcoming Versova-Bandra Sea Link (VBSL). For background, the eight-lane VBSL is part of the ongoing Coastal Road project. One stretch of the Coastal Road — the Bandra Worli Sea Link (BWSL) — has been in operation since 2009. Traffic estimates for the VBSL appear in this report: "Independent Verification of Traffic Estimation for the Proposed Versova-Bandra Sea Link." These estimates are the large numbers I mean - estimates, in many cases, about a time decades into the future. Why would such a guess — in the end, a guess is what it is — be 199,584 rather than 200,000? 107,004 rather than 107,000? There’s a possible explanation. A number like 199,584 suggests that coming up with it involved some serious modelling and calculation, in ways that 200,000 does not. And with these particular numbers, these tables where they appear, this report from which I got them — well, it is important indeed to give that impression. The report was written by Prof. K.V. Krishna Rao of IIT-Bombay’s civil engineering department. Available on the website ( https://msrdc.org/Site/Upload/GR/IITTrafficSurveyReportVBSLP.pdf) of the Maharashtra State Road Development Corporation (MSRDC), it spells out the rationale for building the VBSL. Now I don't know for certain if this report played any part in the design of the coastal road. But its presence on the MSRDC website at least suggests as much. It tells us that it is meant to be taken seriously. And that's just what we'll do, going just by the numbers it offers. It's a sequence of five tables from the report that I'm looking at. Essentially, they offer estimates, under varying conditions and in different years, of how much traffic the VBSL will carry. For example, one table is called "Estimated Total Daily Traffic (Buses and Private Vehicles) in PCUs on proposed VBSL." (PCU - Passenger Car Units; let's just call them "vehicles"). The numbers in the tables were generated by using an elaborate model - by the report's own language, an "econometric four-stage transportation planning model" - that was spelled out in an earlier (2005) study called TRANSFORM. The model takes into account factors and concepts like "crowding curve", "trip generation" and "weighted in-vehicle travel cost", among many more, all of which leave no doubt about its mathematical authenticity. This model produced plenty of graphs and analysis in the report, using the data that's in the tables - all those large numbers, remember, some of which don't end in zeroes. Having established its provenance, let's examine some of the things the report tells us. Written in 2012, it offers traffic projections for 2016, 2021 and 2031. With 2016 firmly behind us, let's consider just 2021 for now. The first number that leaped out at me is 175,260, from the table I mentioned above. It is the total daily traffic volume, meaning the number of vehicles, that will use the Coastal Road between the existing Bandra-Worli Sea Link (BWSL) and the exit at Bandra's Carter Road. Why did this leap out? One reason is the terrifying spectre of even a fraction of those thousands of vehicles flooding off the BWSL onto Bandra's narrow and already congested roads. But leave that be. When the BWSL was built, we were told it was expected to carry somewhere between 100,000 and 140,000 vehicles a day (see my article in this space, "The traffic conundrum: speeding up to a standstill", 21 December 2018, https://www.livemint.com/Opinion/94qFJz07pRNbNEL2nZ271I/Opinion--The-traffic-conundrum-speeding-up-to-a-standstill.html). Let's use the higher number, 140,000. If 140,000 vehicles use the BWSL daily, how will over 35,000 more - 25% higher - suddenly appear on the 2km VBSL stretch to the Carter Road exit? And yet that's not even the real problem with this number 175,260. In that 2018 column, I wrote about the actual usage - as opposed to expectations - of the BWSL. My calculation from the declared toll collection was 57,000 vehicles a day - not 100,000, not 140,000 - through the (then) nine-year life of the BWSL. But wait, even that 57,000 isn't correct. In fact, the history of the BWSL has been one of slipping usage. So in 2017-18, an average of just 32,000 vehicles a day used it. If only 32,000 vehicles use the BWSL daily, how will over 140,000 more suddenly appear on the 2km stretch to the Carter Road exit? But wait, there's still more to ask. According to the table, the 175,260 figure applies if the stretch between BWSL and the Carter Road exit is free. But what if there is a toll? What if a driver is charged Rs 5/km for that stretch, thus Rs 10 since it will be about 2km long? The model considered just that scenario. According to the table, this Rs 10 toll will drop daily usage to just 77,244 - less than half. Think about that for a moment. The report says that if motorists are asked to pay Rs 10, instead of nothing, more than half will not use this road. In fact, the table goes further. If the toll rises to Rs 20, we learn, daily usage will fall by another 50%, to just 33,360. Remember, these are motorists who have paid Rs 70 - the current toll - to use the BWSL. Is it credible that half of them will refuse to pay an additional Rs 10 for the convenience of reaching Carter Road; that if it's Rs 20 for that convenience, three-fourths will refuse to pay? Yet after laying out these numbers, the report has this inexplicable conclusion: "[T]he estimated traffic on VBSL ... is insensitive to the toll." How do 50% and 75% drop-offs suggest "insensitive"? But wait, there's still more to ask. 175,260 vehicles will use the (free) BWSL to Carter Road stretch daily. But going further north past Carter Road, 184,392 will use the Carter Road to Juhu stretch; 163,284 will use the Juhu to Versova stretch. How were these numbers generated and how should we reconcile them? What they suggest is that the model assumes a large number of vehicles will drive only between Bandra and Juhu. Is this a reasonable assumption? But wait, I have one final dilemma. There are more, but I'll stop with this one. Two of the tables refer only to peak-hour usage of the VBSL. One accounts for private vehicles, the other for buses. Together, they tell us that during peak hours, 14,249 private vehicles and 356 buses will use the BWSL to Carter Road leg. Think about that for a moment, again: 356 buses, compared to over 14,000 cars. Let's assume 2 people in each car - the average is actually about 1.75 - and 80 in a bus, which is a seriously crowded bus. That means 28,000 people in cars, and about the same number in buses, will use that stretch during peak hours. If that doesn't furrow your brow, consider how Mumbai's road commuters actually get about the city. In that December 2018 column, I mentioned that 28 lakh commuters use buses daily, compared to 4 lakh who use cars. (We're not even considering suburban trains, which carry upwards of 40 lakh daily). We look at that seven-fold difference, and what do we do to ease these commuters' travel? We build a transport facility that caters unabashedly not to the great majority of them, but to the smallest section among them: car owners. And to justify it, we commission reports filled with models and technical phrases and numbers. Indeed, the very first sentence in this one's conclusion tells us that the VBSL "will significantly ease the congestion and enhance the mobility." How? And no less pertinently, for whom? -- My book with Joy Ma: "The Deoliwallahs" Twitter: @DeathEndsFun Death Ends Fun: http://dcubed.blogspot.com -- You received this message because you are subscribed to the Google Groups "Dilip's essays" group. To unsubscribe from this group and stop receiving emails from it, send an email to dilips-essays+unsubscr...@googlegroups.com. To view this discussion on the web, visit https://groups.google.com/d/msgid/dilips-essays/CAEiMe8rN544ZKZEzAkm3ZAGTu9toJi9GO7mJHEfGM7jW4fjL0w%40mail.gmail.com.