Transportist: April 2023
Ask Me Anything
In celebration of reaching 1000 followers on my Mastodon account, I will do a live “Ask Me Anything” (AMA) on Tuesday April 4 at 9:00 - 10:00 am AEST. (DST will have ended in New South Wales, and presumably it is on in much of the northern hemisphere) (Remember, if you live in the past, er, the other side of the date line, it will be some time on April 3 for you).
If you want to participate in the AMA, and you are not already on Mastodon, get yourself a mastodon account (e.g. at transportation.social or urbanists.social or anywhere else), and follow me. And if you wind up using Mastodon regularly, help out your server admins (compute is cheap, and way less expensive than it used to be, but it is not free) with some spare change, otherwise big social wins.
Research
Wang, Y. and Levinson, D. (2023) Overlooked Transport Project Planning Process - What Happens before Selecting the Locally Preferred Alternative? Transportation Research Interdisciplinary Perspectives. Volume 19, May 2023, 100809 [doi]
Abstract: This study investigates the transport investment decision process of judging the robustness and viability of the selected option compared with the rejected alternatives. This paper adopted the Ridership to Cost Ratio (RCR), that is analogous to the Benefit Cost Ratio (BCR) and contrasts the unit cost at which the proposed transit alternative can serve one additional transit patron, as the primary measure of assessing the relative advantage of alternatives. We found that on average the ‘do-minimum’ option generates a Ridership to Cost Ratio (RCR) nine times higher than the Locally Preferred Alternative (LPA), and the average RCR produced by the second-best alternative is 86% higher than that of the LPA, indicating substantial opportunity costs of rejecting more economical course of action which could have likely managed prospective demand at much lower costs and delivered more services for more people at the same budgetary outlay. Second, we observed that the primary factor differentiating build alternatives is route alignment. However, ridership estimates for alternatives with different routes show narrow differences, raising concerns about whether the existing user benefit assessment framework can fully exploit potential values credited to each alternative option. Third, the number of alternatives considered in the Draft Environmental Impact Statement has gradually declined across the past three decades, informing that the identified transport problems are prone to be approached from confined terms of reference. Last, most projects only compared the preferred light rail mode against the traditional bus mode in the Transportation Systems Management (TSM) base option, attesting to concerns about selection bias and discrimination in early-stage appraisal and decision-making.
Wang, Y. and Levinson, D. (2023) The Accuracy of Benefit-Cost Analysis for Transport Projects supported by the Asian Development Bank . Asian Transport Studies. 9, 2023, 100104 [doi]
This paper analyses the accuracy of Benefit-Cost Analysis (BCA) of transport projects financed by the Asian Development Bank (ADB). Ex post evaluation of BCAs outside Europe and North America has been largely missing from the current literature. This study covers 59 roadway projects in developing countries funded by ADB and examines the accuracy of BCA results through four dimensions: frequency, magnitude, correlation, and tendency. We found that under an average project completion rate of 95.33%, the average construction cost overrun for 59 roadway projects is 10.71%, equivalent to USD 71.4 million. Grounded on 23 projects disclosing detailed economic analysis, we discovered a systematic tendency to understate both the present values of cost (18 out of 23 projects) and the present values of future economic benefits (13 out of 23 projects) in ex ante BCA. Furthermore, more than half of projects (25 out of 47) underestimated EIRR, and about 52.17% of them (12 out of 23) understated NPV. Since the underestimation of economic benefits is too small to counterbalance the underestimation of costs, the project EIRR is on average 5.4% lower than the initial expectation. Moreover, we discussed ADB’s choice of the social discount rate, shadow price, and counterfactual base scenarios, which significantly influence the accuracy of BCA results and the reliability of decision-making grounded on BCA results. Lastly, the causes of cost overruns and benefits underestimations were analyzed.
Research by Others
TramLab Toolkits won the Victorian Premier's Award for 'Design Strategy' last week. Toolkit 1 lays out the steps for developing gender-sensitive communication campaigns. Toolkit 2 details the process for engaging gender-sensitive placemaking to enhance safety for women. Toolkit 3 details the steps for gathering gender-sensitive data for transport spaces. Toolkit 4 outlines the implementation of gender-sensitive training for public transport service providers and aligned security staff.
Rafael H.M. Pereira alerts me to his new Book "Introduction to urban accessibility: A practical guide with R", published by the Institute for Applied Economic Research (IPEA). The book aims to equip readers with the fundamental concepts, data analysis skills & tools needed to perform accessibility analyses and assess the impact of transportation projects. The book also has a practical approach as a hand-on course for accessibility analysis in R with several reproducible examples and open source data sets and code.
Lijie Yu, Mengying (2023) How subway network affects transit accessibility and equity: A case study of Xi'an metropolitan area. Journal of Transport Geography
Train Outages
Sydney Trains has had a bad month
Prof David Levinson, a transport planning expert at the University of Sydney’s school of engineering, said the communications network used by Sydney’s train network “is far from cutting edge”.
But Levinson said “more importantly, the communications system should not have been designed with a single centralised point of failure”.
“For any major infrastructure, redundancy is required for safety and to keep the system operational,” he said, adding this “is critical for a place like Sydney CBD which depends on rail for the mass movement of people”.
Decentralised systems like the internet or the street network are more robust to widespread failures.
Levinson was scathing of how little information had been explained to the public about the outages, including follow-up issues seen over the weekend.
“From the perspective of customers, the reason why the system failed is less important than accurate information in advance about system closures, their durations, and alternatives, which were not provided.”
Levinson warned repeated outages and slowdowns could occur given the state of the train network.
“These systems take a long time to install, and failures often have multiple causes,” he said, noting that “many systems do have some redundancy, so it sometimes takes two or three things to go wrong for a failure to occur, not just one”.
There was another outage this past weekend. ABC Radio called me at 6:05 am on Monday to ask me to talk about it. I said no, since I dunno what failed, or why there are so many problems so close together, or whether it’s just bad luck or something more systemic.
Parking
University of Sydney transport professor David Levinson said parking that was too cheap led to shortages as residents and workers use up parking spots and make it more difficult for visitors to find a space.
“When scarce goods like road spaces, parking in cities, and clean air are underpriced we have over-consumption (congestion, parking shortages, pollution), which wastes everyone’s time and money,” he said. “Roads and parking should be appropriately priced, just like other scarce market goods.”
Levinson said paid parking was more efficient because it discouraged people from staying longer than they needed and allowed more people to access local destinations.
He said street parking imposed costs on society in the form of wider streets than would be required if people had to park their cars on their own property.
“Why should people who don’t have cars pay taxes that support wider streets, just so other people, some of whom have more than one car per adult, can park their car for free on the street?” he said.
Posts
There are more posts than usual, since I have hired GPT4 as a research assistant, and it is quite productive. (Apparently it has no preferred pronoun, I asked, it replied: “As an AI language model, I do not have a gender or a preference for pronouns. You can refer to me using any pronoun that you feel comfortable with.”).
