Register yourself as an individual:

PLEASE NOTE: You will be required to login to register for this event.

INDIVIDUAL AITPM MEMBER

INDIVIDUAL - GOVERNMENT SUBSCRIBER

INDIVIDUAL - CORPORATE SUBSCRIBER

NON-MEMBER

ACADEMIC - REGISTER HERE FREE

Not sure if your organisation is a subscriber?  Check here

Not an AITPM Member? Join now 

Now is the perfect time to join the AITPM and get the AITPM Member rate for this event and access to AITPM's Technical events for FREE, including all the recordings of past webinars - it is excellent value!

Register someone else or a group of people:

REGISTER SOMEONE ELSE

REGISTER A GROUP

AITPM Partners and Sponsors register here:

For AITPM Partners and Sponsors who are entitled to a discounted or free registration, please use this form to register (individuals and groups). 

PLEASE NOTE: For individuals who are also AITPM members,  depending on the type of event,  it may be more cost effective for you to register above individually using your AITPM member login.

AITPM PARTNER AND SPONSOR REGISTRATION

About this event

The AITPM Transport Modelling Network (TMN) brings to you the first in a series of webinars aimed at sharing the wonderful research that is being conducted around Australia at our academic institutions. Here we will listen to three researchers from QUT, University of Sydney and Monash University.  Each presenter will present for 20 minutes followed by a 30 mins Q&A.

Speaker #1: Dr Shamsunnahar Yasmin | Analytics on the changes in vehicle ownership profiles

Followed by the Post-World War II economic upswing, the mobility culture of the industrialized countries around the world have been dominated by private automobile. Australia is no exception. Driving remains the dominant mode of transport among Australian’s accounting for more than 80% of the aggregated trip share. The transport sector is facing an unprecedented challenge of excessive congestion, crashes, air pollution and the associated health burdens. Given the negative externalities of car-oriented mobility, the advocacy of travel demand management strategies has shifted from accommodating a higher number of private cars to the serious consideration of integrated multimodal mobility, sustainable urban transport, and emerging transport alternatives. As such, the major focus of this study is to examine the changes in vehicle ownership profiles (if any) of South-East Queensland Region. Specifically, the household level vehicle ownership decision process will be examined by employing discrete choice modeling framework by using South-East Queensland household travel survey data from two different waves – 2009-2012 travel survey data and 2017-2020 travel survey data.

Speaker #2:  Dr Mark Raadsen | Contributions to the ATRC platform - Open Street Map network building and MATSim traffic simulation

The Australian Transport Research Cloud (ATRC) aims to make it easier for Australian researchers, students, and practitioners to focus on the added value of their work by facilitating a suite of online tools that are otherwise time consuming and or costly to maintain or develop. The University of Sydney contributes with two services: (i) an Open Street Map (OSM) parser capable of converting OSM networks into transport networks, and (ii) an agent-based traffic simulator service (MATSim). These services will be made available within the ATRC but are also available in a stand-alone fashion via our own open-source initiative PLANit (www.goplanit.org) .

Open Street Map contains worldwide data of infrastructure, including roads, rail, and public transport stops. The OSM Parser allows the user to extract such data for any region in the world and create a network consisting of nodes and links that can be used for routing. This is of particular interest when building transport and traffic assignment models. Infrastructure data can be extracted at different levels of fidelity, where the lowest level provides only major roads and rail, while high fidelity returns a detailed network. Various options are available to the user to adjust the included modalities, set a bounding-box, exclude particular infrastructure, and contains intelligence to automatically correct coding mistakes such as public transport stops located on the wrong side of roads, or infer platforms around (train) stations when missing. Network output from the OSM Parser can be converted into MATSim format and when combined with travel demand can be used for (agent-based) traffic simulation. In addition, the OSM parser incorporates extensive logging such that it can also be used to easily detect errors in the OSM tagging itself.

Speaker #3: Professor Hai L. Vu | A new approach to improve destination choice by ranking personal preferences

It is important to have the right choice-sets when dealing with many alternatives in discrete choice models which play a critical role in transport modelling. Popular choice set generation methods include heuristic-based and sampling-based approaches. While these methods have been continuously improved, they do not effectively explain how individuals form their actual choice sets. The best way to know individual choices is to directly ask them about their preferred alternatives, but this is costly and impractical for a large population.

In this work, we propose a novel behavioural choice-set generation approach by ranking personal preferences of destinations using a matrix factorization model with Bayesian personalised ranking. From a large travel survey, we form a user-zone visiting frequency matrix for shopping locations. We then use the model to factorise the user-zone frequency matrix into two lower-rank matrices. The matrix factorization model is optimised by using bayesian personalised ranking. After estimation, the model’s outputs, which are user-factor and zone-factor latent matrices, can produce top preferred shopping destinations for individuals. Our experiment from a large travel survey with thousands of alternatives shows that the proposed framework can increase the hit rate in the choice-set generation stage, as well as significantly improve the predictive capacity of discrete choice models in simulation with small choice-set sizes.

About the AITPM Transport Modelling Network

The AITPM Transport Modelling Network (TMN) is a sub-committee of the Australian Institute of Traffic Planning and Management Ltd (AITPM), and is dedicated to cultivating and catering for the interest knowledge and expertise of transport and land use modelling across Australia. The establishment of the TMN was a core action of the AITPM’s Strategic Plan 2013-2017.

The TMN covers a broad range of transport and land use modelling activities for all modes of transport. This provides an opportunity for individuals from all sectors, including government, academia, research, data collection, software development and consulting to interact, share knowledge and promote good practice within the modelling community and to/from the wider transportation planning & engineering industry. You can read more about TMN here - https://www.aitpm.com.au/about-us/transport-modelling-network

Look forward to seeing you all there!

For any enquires, please contact Nick Veitch

Join the AITPM members from these organisations who have already registered including: