Yiping Yan

Strategic Modeller | Transurban

 

I am currently working on Transport Modelling area as a senior analyst with Transurban. My previous PhD research was dedicated to build a more sustainable, equitable, diverse, connected transport system through establishing the "why" and "how" of the high private schooling enrolment rate in Australia for the transportation sector and revealing the hidden gender gap in the conventional blue/white collar segmentation for commuters market.

A new typology of workers in South-East Queensland, Australia 

Conventional transport model tends to segment commuters as either blue- or white-collar workers. Some have started to use more elaborate segmentations, more reflective of changes in labour markets, such as increased female participation. But often these have been introduced using deductive research approaches, in ways that can lack academic rigour. The highlight of our research is we proposing a new commuter market segmentation derived via an inductive approach “ unsupervised clustering analysis “ as applied to 2017-20 South East Queensland Travel Survey (SEQTS) data. Commuter types are grouped by occupational, industry, and socio-demographic variables (i.e., gender, age, household size, household vehicle ownership and worker skill score). After comparing the travel behaviour differences across groups and the silhouette width analysis on different cluster numbers(k=2,3,4,5,6,7,8), three clusters seem to produce the optimal result. When only three clusters are formed (k = 3) a market segmentation emerges with one female-dominated type (pink collar), one male-dominated type (blue collar) and one with both genders almost equally involved (white collar). There are nuances as to which workers are included in each segment, and differences in travel behaviours across the three types. Pink collar workers are mostly comprised of female clerical and administrative workers, community and personal service workers and sales workers. They have the shortest median commutes for both private motorized and active transport modes.

Nick Fletcher

Managing Partner  | Vivendi Consulting

 

Nick has 20 years experience helping clients deliver challenging strategies and programs. 
Nick has been a leader in global consulting firms in the UK, Europe, North America and Australia. 

He is a founder of Vivendi Consulting, a firm that helps organisations bring their visions to life using multi-disciplinary thinking, human-centred design and advanced data analytics. 

He has worked for clients including Transport for London, Infrastructure Australia and Infrastructure NSW.

Nick is typically engaged by Transport for NSW to deliver challenging initiatives. He managed the development of Sydney's Rail Future, the first published rail strategy for nearly two decades that underpinned Sydney Metro. 

He has helped Transport for NSW to deliver the first Cycling and Micromobility Strategy, which supported the recent announcement of 30 strategic bike corridors in Sydney, covering 250km (dubbed Sydney's answer to London's super cycleway). 

Most recently, Nick helped Transport for NSW create its first Walking and Placemaking strategy for NSW, which has produced some remarkable strategic insights into how and why people in NSW walk, and what government can do to encourage this.

Using machine learning and big data insights to find what actually makes people walk

Vivendi Consulting worked with Transport for NSW to answer the key question of "what makes people in NSW walk“ a question at the heart of our work to develop a strategy for Walking and Placemaking in NSW. 

This challenging question needs a robust, evidence-based answer. In response we developed the Place Analysis and Walkability Scoring (PAWS) model, using sophisticated machine learning and neural network techniques. We understand this is the first time this data and analytical approach has been used. 

The PAWS model uses billions of anonymised points of mobile phone walking data from 179 towns across NSW. PAWS uses machine learning to analyse actual walking against 117 walkability factors, grouped into:

• Structural factors: features that are 'inbuilt' into a centre and difficult to change.
• Adaptable factors: that can be relatively easily changed.
• Demographic factors
• Environmental factors

Our findings support academic studies that find that 'structural factors' such as housing and employment density, permeability and points of interest account for the majority of walking.

Population demographics explained around a quarter of the total observed amount of walking. 

What is notable is that Adaptable factors such as trees, public transport, traffic speed accounted for less than one fifth of the total observed amount of walking.

This approach has yielded a huge number of additional insights. There are 'tipping' points for some variables: minimum measures above which walking is supported, but below which walking is suppressed. These include: 

• Intersection density 
• Housing density

Another key finding was the value of pedestrians. 

The model shows that:
1. If we want to see people walking more around our centres, the fundamentals are vital.
2. These fundamentals tend to be inbuilt when centres are designed. 
3. Making adaptable placemaking interventions, focused on centres that already have the walking fundamentals in place - is likely to lead to more walking.

Dr Hai Vu

Professor of Transportation and Associate Dean of Research  | Monash University - New Activity Based Modelling

 

Hai L. Vu is a Professor of transportation and Associate Dean of Research in the Faculty of Engineering, Monash University. He is a recognised international leading expert in network modeling and planning with strong research credentials in mobility demand modeling, strategic transport planning, intelligent transport systems (ITS), V2X communications and connected autonomous vehicles (CAVs).

Prof. Vu has over 20 years of research experience in network modelling and performance evaluation of complex networks with research spans several disciplinary areas ranging from signal processing and data mining (AI) to mathematical modelling and optimisation of data and road traffic networks. He has authored or co-authored over 200 scientific journals and conference papers in these areas, and is a recipient of the 2012 Australian Research Council (ARC) Future Fellowship and the Victoria Fellowship Award for his research and leadership in ITS.

Prof. Vu is currently leading the Monash team and research activities focusing on the transport demand modelling, planning and related problems.

Monash Mobility Model (M3) for strategic transport modelling and planning

In this talk I will introduce the new Monash Mobility Model developed by our research group in the Institute of Transport Studies at Monash.

Monash Mobility Model (or M3) is an advanced, comprehensive Activity-Agent-based Framework which fully integrates an activity-based module including activity generation, scheduling and location choice with a multi-modal agent-based dynamic traffic simulation coupling via an automated optimization and calibration processes. 

M3 leverages big data, AI/machine learning, optimization calibration and econometric behavioural models to deliver realistic activity and mobility patterns of a synthesis population in large cities. Monash Mobility Model (or M3) can assist transport planners and policy makers to evaluate various strategic transport planning scenarios and different transport policies including emerging and future mobility services with a great level of detail and accuracy.