Wu, Hao, and Levinson, D. (2021) The Ensemble Approach to Forecasting: A Review and Synthesis. Transportation Research part C. Volume 132, 103357 [doi] [Author Link]
Highlights
Review and synthesize methods of ensemble forecasting with a unifying framework.
As decision support tools, ensemble models systematically account for uncertainties.
Ensemble methods can include combining models, data, and ensemble of ensembles.
Transport ensemble models have the potential for improving accuracy and reliability.
Abstract
Ensemble forecasting is a modeling approach that combines data sources, models of different types, with alternative assumptions, using distinct pattern recognition methods. The aim is to use all available information in predictions, without the limiting and arbitrary choices and dependencies resulting from a single statistical or machine learning approach or a single functional form, or results from a limited data source. Uncertainties are systematically accounted for. Outputs of ensemble models can be presented as a range of possibilities, to indicate the amount of uncertainty in modeling. We review methods and applications of ensemble models both within and outside of transport research. The review finds that ensemble forecasting generally improves forecast accuracy, robustness in many fields, particularly in weather forecasting where the method originated. We note that ensemble methods are highly siloed across different disciplines, and both the knowledge and application of ensemble forecasting are lacking in transport. In this paper we review and synthesize methods of ensemble forecasting with a unifying framework, categorizing ensemble methods into two broad and not mutually exclusive categories, namely combining models, and combining data; this framework further extends to ensembles of ensembles. We apply ensemble forecasting to transport related cases, which shows the potential of ensemble models in improving forecast accuracy and reliability. This paper sheds light on the apparatus of ensemble forecasting, which we hope contributes to the better understanding and wider adoption of ensemble models.
I will be giving a seminar to the Sydney University Civil Engineers group today: Tuesday September 21, 2021 from 17:00 to 18:00 AEST. Zoom link for the seminar: https://uni-sydney.zoom.us/j/89457527160
Title: The End of Traffic and the Future of Access
Abstract:
Two decades into the new millennium, transport is becoming interesting again. New technologies are taking root and society is responding; together, these phenomena are changing how people access places and exchange goods. This talk about the future of transport in cities discusses the implications of automation, electrification, sharing, and COVID-19 on travel demands and transport policy.
Bio:
Prof. David Levinson teaches at the School of Civil Engineering at the University of Sydney, where he leads the Network Design Lab and the Transport Engineering group. From 1999 to 2016, he served on the faculty of the University of Minnesota where he held the Richard P. Braun/CTS Chair in Transportation (2006-2016). Levinson has authored or edited several books, including The 30-Minute City: Designing for Access, The Transportation Experience, and Metropolitan Transport and Land Use: Planning for Place and Plexus, as well as numerous peer reviewed articles. He is the editor of the journal Findings.
Join a growing Faculty and be part of a University that places amongst the world’s best teaching and research institutions.
Located on the edge of Sydney’s bustling central business district, close to beaches, parks, public transport and shopping districts
Seeking outstanding Academics to provide leadership and help create a world-class, internationally recognised Faculty for research and education excellence
About the opportunity
The School of Civil Engineering is at the forefront of civil engineering education and research. Our systems approach to teaching allows students to graduate with the much sought-after design, research and problem-solving skills needed to create and manage sustainable built and natural environments.
Our research strengths lie in structural engineering, geomechanics and materials, environmental fluid mechanics, transport engineering and complex systems. Our expertise and facilities continue to meet the issues associated with critical infrastructure, sustainability, climate change, water management and the natural environment.
The School has a long tradition of engagement with industry which has led to successful collaborations with many leading national and global companies and government organisations. This coupled with our excellence in analytical and experimental research has played a prominent role in the development of Australian and International Standards relating to many aspects of civil and structural design.
The School of Civil Engineering at the University of Sydney is searching for faculty members at all ranks and in all areas of Civil Engineering. The school seeks to increase the diversity of its faculty and encourages women to apply.
The School of Civil Engineering at the University of Sydney is prioritising a search for faculty members who are able to contribute in the areas of:
El-Geneidy, Ahmed and Levinson, D. (2021) Making Accessibility Work in Practice. Transport Reviews [doi] [first 50 free download]
Accessibility, the ease of reaching destination, is the most comprehensive land use and transport systems performance measure (Levinson & Wu, 2020; Wachs & Kumagai, 1973; Wu & Levinson, 2020). Accessibility has been applied in planning research since the 1950s (Hansen, 1959), and still today, we find major barriers to adopting it in practice (Handy, 2020). Advances in computing and software have enabled researchers to generate complex measures of accessibility with higher spatial and temporal resolutions moving accessibility research at a fast pace, while the implementation of accessibility, in practice, lags (Boisjoly & El-Geneidy, 2017). Even simple measures, such as the cumulative opportunities measures of accessibility, confront challenges in adoption.
Allen, Jeff, Farber, Steven, Greaves, Stephen, Clifton, Geoffrey, Wu, Hao, Sarkar, Hao, and Levinson, D. (2021) Immigrant Settlement Patterns, Transit Accessibility, and Transit Use. Journal of Transport Geography. 96 103187 [doi]
Abstract: Public transit is immensely important among recent immigrants for enabling daily travel and activity participation. The objectives of this study are to examine whether immigrants settle in areas of high or low transit accessibility and how this affects transit mode share. This is analyzed via a novel comparison of two gateway cities: Sydney, Australia and Toronto, Canada. We find that in both cities, recent immigrants have greater levels of public transit accessibility to jobs, on average, than the overall population, but the geography of immigrant settlement is more suburbanized and less clustered around commuter rail in Toronto than in Sydney. Using logistic regression models with spatial filters, we find significant positive relationships between immigrant settlement patterns and transit mode share for commuting trips, after controlling for transit accessibility and other socio-economic factors, indicating an increased reliance on public transit by recent immigrants. Importantly, via a sensitivity analysis, we find that these effects are greatest in peripheral suburbs and rural areas, indicating that recent immigrants in these areas have more risks of transport-related social exclusion due to reliance on insufficient transit service.
Fig. 3. Bivariate maps of transit accessibility and density of recent immigrants.
Catbagger n. Someone who tries to put the cat back in the bag. I.e. someone attempting a futile act too late, which may have been prevented but cannot be reversed.
In Australia are currently experiencing an exponential increase in COVID-19 cases as part of the Delta Wave. This is sad, and results in a few deaths daily (a rate, mind you, that is low enough other countries use it as a level at which lockdowns are lifted, rather than imposed). The rise is due to any number of mistakes that went previously. Iwon’tre-litigatethepast. Instead, I posit that had those mistakes not been made at that time, lessons from those mistakes wouldn’t have been learned, and a similar mistake would have then been made shortly thereafter. This is not an apology for incompetence, and I am sure none of the Transportist readers would have made those mistakes had they been in charge, but is an acknowledgement that like COVID-19, incompetence is endemic and no one competent person can be everywhere simultaneously, and everyone relies on systems that are only as good as their weakest link.
If not for some outbreak, people would not (over)-react, leaving the same conditions in place for a later outbreak. While on average one prefers to avoid mistakes, it is only by mistakes that lessons are learned, pre-planning is imperfect, and we can plan and prepare for any number of eventualities that would never occur at great cost, leaving us worse off than those who react to the eventualities that do actually occur without having wasted resources preparing for those that don’t.
So while it may be psychologically or politically important to blame individuals who should have done this instead of that, or have learned from the mistakes of others, (and obviously the best people do better than the worst, by definition), and hopefully select slightly less incompetent administrators, that merely would have delayed the mostly inevitable outcome in terms of cases and deaths. And until there were COVID outbreaks, or very obvious prospects of COVID outbreaks, no vaccine would have been developed, no vaccine would have been manufactured, and no one would have gotten vaccinated, leaving everyone vulnerable to a COVID outbreak.
For authors, considering peer review: Are you biased so as to be more likely to accept papers that cite you? Are others (generally) similarly biased?
I am/Others are 46.3%
I am not/Others are 24.4%
I am/Others are not 0%
I am not/Others are not. 29.3%
At least no one admitted to being more unethical than the population as a whole (choice 3). About half the people admitted bias (choice 1 and 3), indicating that the people who thought “Others are not” (choices 3 and 4) are hopelessly naive. I tend towards choice 2 for myself, at least I hope I am not.
How much time would you be willing to sacrifice at the end of your life (your life would be X units of time shorter) to forego 1 month of lockdown for yourself?
0-1 hour 44.4%
1-24 hours 17.8%
1-6 days 11.1%
7 or more days 26.7%
Now these are Twitter polls, so sampling bias is rife, and questions cannot be particularly sophisticated (lockdown means different things to different people in different places, what about lockdown for other people, etc.), but it does suggest that many people think that lockdown makes their life worse off in a way that suggests their benefits (reducing COVID cases) need to be countered with their costs (diminished quality of life). It also suggests that some other people really like lockdown, and if there had been negative numbers, some people who elected for choice 1 might have given up time at the end of their life to preserve lockdown longer. This I think gets pack to the Plants vs. Animals dichotomy I developed last year.
The World Society for Transport and Land Use Research presents awards at each Symposium. The conference took place last week, and I had the honour of bestowing the Best Paper and Best Student-Led Paper awards (and the honour of being the namesake of one of the awards). Congratulations to the winners who are:
Traffic-Land Use Compatibility and Street Design Impacts of Automated Driving in Vienna, Austria Emilia Brucke and Aggelos Soteropolis, Technical University of Vienna
HONORABLE MENTION BEST STUDENT-LED PAPER
Traffic Noise Feedback in Agent-Based Integrated Land-Use/Transport Models Nico Kuehnel (Technical University of Munich), Dominik Ziemke (Technical University of Dresden, and Rolf Moeckel (Technical University of Munich)
The University of Sydney Faculty of Engineering is running two events in Research Week aimed at recruiting prospective Higher Degree by Research Students. The event on September 9 has a breakout room led by Mohsen Ramezani discussing Transport Engineering.
8 September 2021 – 1:00 pm AEST
The Value of a PhD in Engineering, Computer Science or Project Management
Associate Professor Kalina Yacef, Associate Dean Research Education
Breakout rooms – Our Research
Aerospace, Mechanical and Mechatronic Engineering- Dr Donald Dansereau
Biomedical Engineering – Dr Omid Kavehei
Chemical and Biomolecular Engineering – Dr Farshad Oveissi & Professor Fariba Dehghani
Civil Engineering – Associate Professor Daniel Dias Da Costa & Associate Professor Yixiang Gan
Computer Science – Dr Shuaiwen Leon Song
Electrical and Information Engineering – Professor Xiaoke Yi
Project Management – Dr Nader Naderpajouh
Admission, Scholarships and the Student Experience
Lyndon McKevitt, Research Education Manager
9 September 2021 – 7.00pm AEST
The Value of a PhD in Engineering, Computer Science or Project Management
Associate Professor Kalina Yacef, Associate Dean Research Education
Breakout rooms – Our Research
Aerospace, Mechanical and Mechatronic Engineering – Professor Stefan Williams
Biomedical Engineering – Dr Omid Kavehei
Chemical and Biomolecular Engineering – Professor Fariba Dehghani
Civil Engineering – Dr Mohsen Ramezani
Computer Science – Dr Chang Xu & Dr Qiang Tang
Electrical and Information Engineering – Professor Xiaoke Yi & Dr David Boland
Admission, Scholarships and the Student Experience
Lyndon McKevitt, Research Education Manager
David Levinson, Transportist 
You must be logged in to post a comment.