What interventions should Infrastructure Victoria test with its new Melbourne Activity Based Model?

What interventions should Infrastructure Victoria test with its new Melbourne Activity Based Model?
What interventions should Infrastructure Victoria test with its new Melbourne Activity Based Model?

Infrastructure Victoria unveiled a new round of research into its larger programme of work dealing with managing transport demand.  The authority contracted Arup and KPMG to produce the Melbourne Activity Based Model (MABM) and while it is new, it is considered fit for purpose in the strategic context.

The MABM fact sheet published on Infrastructure Victoria's website summarises the model as follows.

The MABM was developed to provide an additional tool for IV and the Victorian Government to test the impacts of transport policy and infrastructure proposals on the behaviour of transport system users.  It is one of a range of tools that can be used for this purpose.

The MABM uses a different framework than existing strategic transports models used in Victoria.  Existing models use the traditional 'four-step' modelling framework.  Planners in the USA designed the four-step framework in the 1950s for the testing of alternative options for highway construction.  Four-step models are mature, tested and well understood by practitioners.  As a result, four-step models are still used extensively around the world.

However, the four-step model framework has limitations for certain types of analysis.  New techniques have been developed and matured over the last twenty-five years which address some of these limitations. 

Some issues that MABM is more suited to addressing thank the traditional four-step model framework include:

  • Understanding the transport's 'customers' and their needs and preferences.
  • Understanding how customers with different socio-economic and demographic characteristics - such as income, household composition and age - respond to changes in the transport policy or new infrastructure.
  • Understanding how fair and equitable a transport policy or investment is, which groups are the winners and losers and to what extent they are impacted.

The report produced by KPMG and Aurp seeks to assess Melbourne's transport network with a 2015 base case and a 2030 future case, looking at the levels of use of the public transport and road networks will change over that period.

It was found that 'over the next 15 years, the population of Greater Melbourne is estimated to grow from 4.5 million people in 2015 to almost six million people by 2031.  An additional 400,000 workers are expected by 2031, increasing the number of daily trips to work within Greater Melbourne by 25%  to almost two million.'

The report also finds that traffic growth will not be experienced uniformly across all regions of Melbourne in line with expected increases in the existing three outer growth corridors and inner-city.

A number of well-known transport projects from the Level Crossing Removal Program and Melbourne Metro to the North East Link have been factored into the modelling for the 2030 future case.

The report notes 'given the overall travel demand growth and changes in transport network supply between now and 2031, travel patterns associated with public transport are [also] forecast to change.   Between now and 2031, the modelling predicts a 75.9% increase in public transport trips across Greater Melbourne, which corresponds to an additional 878,000 public transport trips each day.'

The inner regionals of Melbourne will experience the greatest absolute growth in public trips - the outer regions will experience the largest relative growth due to the low-base figures.  

Despite a large number of new public transport trips predicted this only translates into a very marginal 3.6% mode shift by 2031 where cars are still likely to dominate with an 86.4% mode share of motorised travel versus public transport's 13.6%.

And this relatively modest shift in numbers is acknowledged:

Car dominance is a symptom of the structure of Melbourne's public transport system, with heavy rail services radiating from the city centre.  Melbourne is likely to remain a car dominant city, with driving forecast to account for more than 70% of trips in 2031. 

While this supports public transport use as a significant mode of travel for trips originating in the inner regions (growing from 39% mode share in 2015 to 48% in 2031), for other parts of the metropolitan area, this is not the case. 

Often, bus services are the only viable public transit alternative, particularly for the outer regions, which can struggle to compete with private vehicles in terms of fast journey times.

The greatest challenges are for people living in the outer regions of Melbourne.  Given the large geographic extent of these regions and their low density, public transport will unlikely be a solution to the problems of excess road demand in these areas without significant levels of investment. 

What interventions should Infrastructure Victoria test with its new Melbourne Activity Based Model?
Daily car trip growth 2015 - 2031 - IV, KPMG, Arup


The page on Infrastructure Victoria's website says that 'in 2018 we will look at a range of interventions that can help manage transport demand and make it as easy as possible for people to get around' which is encouraging because in the context of $20 odd billion of new tollway projects having been unveiled during this Government's time in office, it's time to get creative.

The Andrews Government has already rejected out of hand one of IV's key recommendations - a new road pricing regime - and had a knee-jerk reaction to being wedged by the opposition by piling on the North East Link - politics and transport planning as usual in other words.  So it's refreshing to see one of the new institutions the Andrews Government has set up is doing what it was intended to do - provide independent advice.

Whether the major parties listen and adapt their transport planning policies around this public advice is, of course, a whole different ballgame.

What interventions should Infrastructure Victoria test with its new Melbourne Activity Based Model?
Road utilisation in the traditional peak hour in 2031 - IV, KPMG, Arup

That said, here's a laundry list of interventions which would be interesting to see modelled.

Heavy Rail

The previous state government released the heavy rail network development plan 4-5 years ago and for the most part, all new rail projects that have been announced or appeared in other public domain planning documents have loosely followed it, perhaps with the exception of staging (Mernda rail extension), alteration (Melbourne Airport Rail link) or in the case of removal (the North East Link will include a Doncaster busway which effectively kills off the idea of a rail line for at least another few decades).

The four-stage plan provided a vision of Melbourne where trains would be secotrised (to provide independent operations, significantly lowering delay and disruption risks) through two new tunnels - Metro 1 (under construction) and Metro 2 (Clifton Hill to Fishermans Bend, which is now recognised as per the Fishermans Bend Structure Plan process as running from Clifton Hill to Newport).

It also provides for a reconfiguration of the City Loop which would see two of the four tracks become a through line and join the Craigieburn and Frankston lines.  Likewise the Sandringham and Upfield lines would join and form a north-south line with some hard infrastructure works at North Melbourne to separate lines and the Alamein/Glen Waverley Lines would join and run straight through to the West  providing for a second east-west line (the first will be created by joining Pakenham/Cranbourne with Sunbury and eventually Melton).

Secotirsation's primary benefit is increasing capacity on a single line, and rolling out state of the art signalling throughout all lines - like what's being done between Sunshine and Dandenong as part of the Metro 1 project - as well as continually grade separating level crossings will allow far greater frequency of service.  The condundrum for increasing frequencies fundamentally changes from a network (track) problem to one of procuring the right amount of rolling stock.

Therefore, IV would do us all a public service by testing the following:

  • assume an accelerated programme to ensure Metro 2 starts sometime soon during or after the main works for Metro 1 are complete as well as conducting the City-loop configuration - this might complete by 2035 (and outside the 2030 future case), but just assume at least one of those extra heavy rail projects is complete and open for service
  • what's the impact on modal share in the outer suburbs (Metro 2 links two of the outer suburban growth areas - Mernda & Werribee) as well as the inner suburbs?
  • what's the impact on major freeways - West Gate, Monash & Eastern?

Time for some creativity

As per the heavy rail network plan, the Alamein and Glen Waverley line become eastern branches of a second cross-town line to the West - what happens when the Alamein line is extended south with a stop at Chadstone and then interchange at Oakleigh on the Metro 1 line?  What happens when that line is extended further to become the Rowville line?

The original Rowville Rail study explicitly excluded this link and we therefore have no idea what the impact of a train station at Chadstone shopping centre is.  Take a look at the road utilisation image from the report above - that kind of Alamein extension would drive right through the heart of the predicted congested areas in 2030.

And getting even more creative.

We're no strangers for advocating for researching and building an orbital rail network for Melbourne.  This article from March might serve well for some inspiration. 

The radial heavy rail network with all the above network expansions will have an enormous amount of new capacity built in, assuming all lines run on ten minute off-peak and sub-ten minute peak frequencies (with maximum throughput only governed by the number of trains available for each line & thanks to full sectorisation).

What happens to travel patterns when Melbourne's rail lines are all knitted together outside the core of the city, through the suburban job and leisure/shopping clusters like Monash University & employment centre, Box Hill, Deakin University, Doncaster, Highpoint, Sunshine and Northland which could also see significant new residential and job spaces created?

That's our public transport 'freeways' covered - now let's look at everything else.


Our tram network is the backbone of local and one-transfer from the rail network travel in the inner-city, but the network runs embarrassingly slow.  If you've been living under a rock, you might have missed that the inner-city is an enormous growth engine, not just in jobs, but for where people want to live. 

We're thankfully starting to see the middle ring of suburbs start to see intensive development, but central Melbourne as a sacred cow of activity is unlikely to be thrown off course - and the tram network allows those locals to move about the inner-city and the fringe middle-ring suburbs in some cases.

To give some context, here's an anecdote.  For the past three weeks I've been travelling to the Victoria gardens area of Abbotsford/Richmond a lot - a peak direction commute from the north to the city and then a counter-peak commute from the city out to Victoria Gardens.

In the morning peak, clearways only operate city-bound on Victoria Street between Victoria Gardens and Hoddle Street and only operate outbound in the afternoon peak.

There are two tram route choices - the 12 (which originates/terminates at Victoria Gardens) and the 109 which runs right out to Kew and Box Hill.  In the afternoon peak, to travel from Victoria Gardens to Hoddle Street - about half the journey toward the city & Parliament station - it's taken on average 13 minutes, on one day it took 18 minutes.  To travel approximately 2km. 

This is running pace and a shocking indictment on how cars are prioritised over trams in the inner-city.

There's no doubt many other anecdotes in other parts of the tram network - especially where tram tracks run down an undivided street (the majority of the network) - but what's the impact on the transport network if, at best, all tram routes which are in this narrow road scenario all become permanent clearways (and on-street parking replaced off-street) or in a compromise, what's the impact of all tram routes become peak-hour clearways (regardless of peak direction)?

The tram network suffers the same fate as the heavy rail network - connections (excluding the Chapel Street and Moonee Ponds-Footscray tram) can only be made in the CBD - what would the impact on the transport network be if the old inner-city rail route were to re-open as a tram route linking Royal Park with Clifton Hill and perhaps an entirely new connection from South Yarra via Domain and into Fishermans Bend?


The MABM fact sheet notes that 'it is also more suited to modelling behavioural responses to complex changes to the transport landscape that are likely to occur in coming years and decades' with examples including the emergence of connected and autonomous vehicles.

Well here's some low hanging fruit for the suburbs and the last mile part of the public transport network that is currently dominated by cars in the growth areas and some of the more established suburban areas: 'autonobuses'.

Granted, modelling the impacts on behaviour with autonobuses will be difficult and no doubt requires many more assumptions to be made, but early modelling will be valuable as a baseline as the technology and regulatory regime around autonomous vehicles develops.  One suspects this would tie in with Infrastructure Victoria's other pet area of research: road pricing regime change.

And of course, the bus as we know it today, much maligned as it is (whether fairly or unfairly), should also get a lot of modelling love.  It's with some warped and outdated egalitarian ethos that roads in Victoria are treated as a free-for-all with limited priority for road-based public transport, but let's model the behavioural change by doing away with the last century road's policy thinking.

The fixed route bus is likely to be here for many decades to come, especially as the regulatory regime and technology for autonobuses evolve, and in the near term it is still the mode that can be rolled out the fastest and still the mode that really ought to have a much bigger focus in the city's public transport network.

So what happens when major arterial roads see a greater rollout (and policing!) of bus lanes - in the same way that trams might get permanent clearways?  How will fixed route buses running on high frequencies on trunks integrate with autonobuses in future?  How much more people moving capacity do we get out of existing road network is there in the widescale priority given to public transport services?

Alastair Taylor

Alastair Taylor

Alastair Taylor is a co-founder of Urban.com.au. Now a freelance writer, Alastair focuses on the intersection of public transport, public policy and related impacts on medium and high-density development.

Comments (1)

Help contribute to the Urban community by leaving your comments about this article
What would you like to say about this project?
George D
Things they should test and model:

Pedestrianising more of the inner city. There will be 1 million people there each weekday soon. It's currently becoming a crowded mess.

Cycle lanes that actually get people safely from one part of the city to another and are suitable for ordinary people (not men in lycra like myself).

Doncaster bus-rapid-transit in the motorway corridor.

Permanent bus lanes and parking clearances on more arterials.

More connections.

And consideration to adoption of driverless technologies, which will start to roll out over the next decade.
Not helpful

Reply to this comment

What would you like to respond to this comment?