Friday, August 16, 2019

Building Melbourne's Useful Network: Part 16 - Monash Uni, Glen Waverley, Knox and Ferntree Gully Rd

Big suburban universities, especially if they have no trains, are good for bus patronage. Not only do students ride buses to get to class but also for many work and leisure trips. Especially if they live away from home and cluster where the public transport is (like overseas students).

Regional shopping centres, like Chadstone, are also good for bus patronage. Even if trains do run (like Southland) much of their catchment comes via corridors served only by bus. And there's the jobs and visitor traffic generated by suburban hospitals.  

Land is getting scarce in the middle suburbs. If major destinations wish to grow their business they either need to build multi-storey parking (expensive) or shift driving trips that can be walked, cycled or bussed to these more space-efficient modes. 

This applies for employees as much as customers or students. The provision of subsidised or 'free' parking for employees is an extra business expense as it means the space cannot be used for revenue-generating floorspace or even customer parking. And poor public transport access can make a place less attractive to visit, work or shop at.

Existing bus services

A common theme here is that in Melbourne there is often little relation between how often bus routes run and their patronage potential. An urban bus serving busy stations, shopping centres and universities might be less frequent than one through a semi-rural area. At best a major destination like Chadstone or Monash University will have frequent service from a few directions only. And even that is on weekdays only. 

This situation is not very good. But it illustrates the huge potential to easily and cheaply use buses to add people carrying capacity (at a fraction of the cost of building more roads) in increasingly congested and densifying areas like Chadstone, Clayton, Glen Waverley and beyond. State electorates  that would benefit from any bus improvements here include Oakleigh (Steve Dimopolous MP), Mt Waverley (Matt Fregon MP), Ferntree Gully (Nick Wakeling MP), Rowville (Kim Wells MP) and Mulgrave (premier Daniel Andrews).  

Existing Useful Network

Let's look at the existing Useful Network for the Monash University, Glen Waverley and Knox area.  

I explain the Useful Network concept here. It's those routes that are frequent enough and run over long enough hours to be useful for many trips. I've specified a 20 minute frequency on weekdays and 7 day service until 9pm. In other words the coloured lines on the Melbourne Public Transport Frequent Network map with the 20 minute frequency selected.

You can see it on the map below (click it for a better view). Useful north-south routes comprise the 703 and 902 SmartBus routes. Further west there's the 903, but there's a big gap between that and the 703. We sorted that area out, by upgrading the 733 and the 767, back in Useful Network Part 3.

Useful east-west routes are tram 75 (and the connecting 732 bus), the Glen Waverley train line and the Route 900 bus to Stud Park. The latter are more than 3km apart, leaving a large area beyond Useful Network coverage. Eastern areas are also sparsely served with no Useful Network coverage beyond the 901 on Stud Rd (trains to Belgrave and Lilydale being only half-hourly on weekdays, ie less than Geelong). 

As important as coverage is whether there is useful service between the main places that people wish to go.  There are some major gaps. Most conspicuous is the absence of a strong route between shopping, trains, jobs and education (Chadstone, Oakleigh and Monash Uni) and homes to the east along Ferntree Gully Rd. Also missing is a frequent and simple service between Monash University and Glen Waverley, and beyond that to Knox City.  See these gaps on the map below. 

In most cases bus routes already run but they aren't frequent enough. They're typically every 30 minutes each whereas a 15 or 20 minute (or better) off-peak frequency is desirable. Can we improve frequencies without buying buses if we ran our existing fleet harder?  Keep reading. 

An expanded Useful Network

693 interpeak frequency boost

Our first challenge is Ferntree Gully Rd. The existing Route 693 runs straight along it from Oakleigh to Belgrave. It runs every 20 minute during peak and every 30 minutes off-peak. Improving it to a Useful Network standard just requires that its peak frequency run during the day. This could be timed to meet every second train at Oakleigh Station. You'd need more drivers to run the upgraded service but the same bus fleet could be worked harder. 

693 extension to Chadstone

What about access to Chadstone Shopping Centre? Currently 693 terminates at Oakleigh. While there are numerous buses between Oakleigh and Chadstone that people can change to, legibility is poor. And, because 693 is currently only every half hour off-peak, people from Chadstone need to allow a lot of time for a connection to Oakleigh to be sure of catching their intended 693. Since 693 has a lot of catchment, and Oakleigh to Chadstone isn't far, an extension is considered acceptable and would increase the route's appeal despite duplicating other services. It would also restore the 1980s Chadstone destination it had before the 1991 service cuts

Where do we find the resources for this extension? Have a look at the eastern end of Route 693 (below). 

The area is more sparsely populated than the western end of the 693. 693 overlaps with trains near four stations east of Ferntree Gully. Meanwhile, because many routes terminate at Belgrave or Upwey, even quite local trips, such as between Belgrave South and Upper Ferntree Gully, require a change to a train. A way to resolve this is to terminate 693 at Ferntree Gully or Upper Ferntree Gully and extend local routes (running less frequently than the 693) to these locations. Then more local trips could be made on the one bus. And because routes are shorter local buses could be rigorously coordinated with trains, including being held back if trains are late. In terms of funding, bus kilometres could be freed by deleting the largely duplicative Route 694, as recommended in the Knox Mooroondah Yarra Ranges Bus Service Review of 2010.  Shortening 693 here frees up time for the Oakleigh - Chadstone extension.

Monash University - Glen Waverley - Knox City 737 upgrade

Glen Waverley trains operate every 15 minutes off-peak on weekdays. Currently there are two bus routes between Monash University and Chadstone (737 and 742). Both operate roughly every 30 minutes but go different ways. This makes access to Monash from Glen Waverley (which has significant surrounding density and shopping) less legible than it might if there was only one 'turn up and go' route. 

The 737 from Glen Waverley continues to Knox City (and beyond). Its half-hourly frequency makes it unattractive for CBD trips via Glen Waverley Station. More frequent alternatives include taking the 901 to Ringwood or the slow 732 option via the 75 tram. Access to Monash University Clayton from Knox City is also inferior to access from Stud Park, which has the 900 every 10 - 15 minutes. 

Route 737 is also very well used. Apart from the 733 (discussed here) and the 767 (discussed here) no other bus route with a basic 30 minute frequency carries over a million passengers per year. Patronage numbers more than justify a 737 frequency upgrade. 

There are many benefits of a 737 upgrade. A 15 minute frequency between Monash University and Knox City could allow 'every train' timed connections at Glen Waverley during the middle of the day. It would improve connectivity between education and jobs in the Monash precinct and the Knox area which  is notorious for its poor public transport. And, in conjunction with the 703, there would be 8 buses per hour connecting the Glen Waverley line with Monash University at Syndal Station.  

Where can we find the resources for a 737 frequency upgrade? Have another look at the Ferntree Gully Rd area network. Every half hour are buses on Route 742 between Chadstone, Glen Waverley and beyond to Ringwood. Route 742 has no unique coverage. The main overlap is along Ferntree Gully Rd, where 693 (that we've just upgraded) runs. There is another overlap between Chadstone and Oakleigh. That's also redundant given the previously discussed 693's extension to Chadstone. Finally there's an overlap up Springvale Rd with routes 850, 885 and 902. 

What if we were to operate Route 742 as a Glen Waverley to Ringwood route, with trips carefully timed with trains at Glen Waverley? 742's Chadstone to Glen Waverley trips could then be rolled into Route 737 to provide a 15 minute service between Monash University, Glen Waverley and Knox City. That's double today's frequency. It could connect well with trains at Glen Waverley and mesh with the frequency provided on  the  901 SmartBus. Every second trip would continue to Croydon, maintaining the current frequency. Upgrade of this portion could be considered in a future review. 

Instead of two routes every half hour at irregular intervals, Ferntree Gully Rd would get one route (693) every 20 minutes meshing with trains every 10 minutes at Oakleigh. Fewer buses per hour but simpler and neater. There would be better service to Chadstone and local access improvements in the Ferntree Gully area with some local route extensions. If crowding is an issue then short Wheelers Hill - Oakleigh trips could be added during peak times to provide a combined 10 minute service. That's the sort of frequency that would really appeal to commuters and boost patronage.  

Paying for it

We've already discussed the 693's funding. It requires more bus and driver hours but shouldn't materially change peak bus requirements. Some economies may be possible around Belgrave, where  bus patronage is lower than in the areas we're improving. 

As for 737's upgrade, this comes straight off the 742. Off-peak Glen Waverley to Chadstone on the 742 takes about 40 minutes (withdrawn trips), the same as Knox to Monash on the 737 (added trips).  Peak run times will be longer and 742's timetable is sparser. However frequency doesn't necessarily need to be doubled. Depending on passenger loading you might wish to finish some earlier trips at Glen Waverley Station. However it would still be desirable to have a 10 minute service to the Monash precinct when most people are travelling. You would also need to balance resources between this and the short trips from Wheelers Hill on the 693 mentioned before.

Loose ends

Almost every bus network reform, especially one that seeks to rely on cost-effective improvements like this, has some loose ends. The 742, which this plan deletes, sometimes deviates off Ferntree Gully Rd to go nearer Monash University. That's convenient for students but it delays through passengers on their way to Oakleigh Station. 

The expanded Useful Network just has the upgraded 693 along this portion. Options could include (i) making people walk further, (ii) deviating all 693 trips, making it less direct or (iii) deviating some 693 trips, making it less legible. 737 could be terminated there but this would harm the ability to connect from other routes at the main interchange so is not recommended. 

Other loose ends include the removal of 742. Some parts of Springvale Rd would lose their direct bus to Monash. An attempt should be made to provide as convenient as possible bus-bus interchange facilities, including short walking distances and responsive pedestrian lights at the Brandon Park intersection.  That change and a similar one at Wellington Rd would also assist those going to Chadstone Shopping Centre. 


This would have to be one of the biggest 'bang for the buck' Useful Network upgrades featured. It upgrades already popular routes and connects major trip generators such as Chadstone, Monash University, Glen Waverley and Knox City. There is also a large increase in the residential catchment near a Useful Network service, especially near Ferntree Gully Rd. 

What do you think? Is this an improved network? Is it worth the hassle of some of the 'loose ends' created? Please leave comments below.

PS: Want to see expanded Useful Networks in other areas? An index to them all is here


Michael Angelico said...

Ah, you've done 733, now you're doing my other hobby horse! :) Gotta love the middle south eastern suburbs with their nice wide straight roads for buses to run on.

I agree on almost all of what you've said - duplicate routes along FTG Road makes no sense (especially when they run every half hour and five minutes apart), and with 742 being duplicated elsewhere it makes sense to trim it back and strengthen the eastern half of it with higher frequencies.

I used to do the Monash-Glen Waverley trip fairly often, and I found that the connection from 703 to the train at Syndal was pretty good - I'd usually arrive on the platform to see the PIDS showing three or four minutes. Even so the 737 trip was quicker.

Re extending 693 to Chadstone - there's plenty of buses that go there already, and I generally think it's a bad idea to indulge people's unhealthy obsession with the single seat journey. Especially when it means making a long route even longer.

On that note, I'd really like to see Oakleigh-Chaddy-Holmesglen given a short, high frequency shuttle - a route 801 maybe?

Peter Parker said...

Thanks Michael

I thought long and hard about extending 693 to Chadstone but decided to do it because:
(i) It would soften the blow of deleting the 742, which does go to Chadstone
(ii) The 693 doesn't end up being longer as I shorten it in the east to compensate
(iii) Chadstone is such a major trip generator I don't mind a small amount of duplication
(iv) Oakleigh - Chadstone service is not yet super frequent at all times (especially weekends where you might have uneven gaps with the half-hourly SmartBuses and hourly on some other routes)
(v) If we wanted to simplify things you'd look at other duplicative corridors first - eg the new 627 & 822, the messy 802/804/862 or the terrible 624 in conjunction with the 623 (which duplicates part of 900).

As for the service km difference it's not much. Since the 742 is already half-hourly (on weekdays) we're just adding just one trip per hour with the upgraded and extended 693. I reckon it's worth it. But in peak (especially am peak) I'd run short trips finishing at Oakleigh.

As for the 801 idea I think we can just simplify information and upgrade the SmartBuses for a good frequency without an extra shuttle. Though the Chadstone deviation is so time-consuming that it would be desirable to have something that runs straight from Oakleigh - Holmesglen to save time. But that splits frequency from the 903 and the faster time might be offset by more waiting. Still north-south connectivity is a major structural flaw of the bus/tram network - it's either infrequent or too slow for a large chunk of east.

Having said that would you be bold enough to take the 903 out of Chaddy, upgrade both it and 900 to a 10 min 7 day frequency, and have your 801 every 10 min to go into Chaddy? That's an extra transfer that won't be popular. But it would speed through travel.