So far I've examined areas that could be given better bus services without much change to their local networks. For Greater Dandenong I'll be more ambitious. I need to be. Existing routes are too complex. It's hard to justify upgrades without making them simpler.
As mapped on Charting Transport, Greater Dandenong residents are amongst Melbourne's most disadvantaged. Many live away from stations. Buses enable access to employment, education and social opportunities. Their usage is high for the service offered. However often they run only every hour, even at peak times. Some weekend timetables have buses every two hours or not at all.
Local routes are largely unchanged from 30 years ago. The area's last serious bus network review (about 10 years ago) was not acted on. When services were added (eg route 709's extension to Noble Park) they were run over or near unchanged existing routes. This causes confusing and wasteful overlaps like you see in the PTV map below.
Existing Useful Network
As mapped on Charting Transport, Greater Dandenong residents are amongst Melbourne's most disadvantaged. Many live away from stations. Buses enable access to employment, education and social opportunities. Their usage is high for the service offered. However often they run only every hour, even at peak times. Some weekend timetables have buses every two hours or not at all.
Existing Useful Network
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.
Greater Dandenong Useful Network routes are shown below. There aren't many for the area's population. To give an idea of scale, Springvale Rd (902) and Stud Rd (901) are 6 km apart. Most areas in between are residential, served only by infrequent buses. It's a stark contrast to the PTV map, crowded with (mostly hourly) routes.
Key gaps include Heatherton, Chandler, Police and Gladstone roads. Also significant are Jacksons, Westall and Corrigan roads. All except Corrigan Rd have a continuous or near continuous bus route down them. Some roads have multiple routes but none operate at Useful Network frequencies. Multiple overlapping routes allow service upgrades for little cost if they can be simplified.
Expanded Useful Network
The map below shows an expanded Useful Network. Parts of some existing routes go from every 60 to every 20 minutes. Many wide gaps above are gone. Tens of thousands would get a Useful Network service nearer home with this expansion.
Sometimes this is done by rolling overlapping infrequent routes into one more frequent service. For example between Keysborough and Dandenong. Three overlapping routes every hour (812, 813 and 815) become a simpler single route (813) every 20 minutes.
Elsewhere more services are added on existing route alignments. Unavoidably this requires more buses, more drivers and more hours. However the existing patronage along with Greater Dandenong's demographics and social needs make such investment worthwhile. This is different to my previous Useful Networks in areas that had better service to start with so were cheaper to upgrade.
Expanded Useful Network and local routes
Just upgrading existing routes to run more frequently can increase duplication and waste money. Especially in Greater Dandenong where routes often overlap. However making them straighter could mean that some stops lose their bus.
That may be no big deal if there are other stops nearby and you were doing a full area network review informed by survey data and public engagement. But that's not today's exercise. So I'm being more modest, mostly only suggesting changes to neighbourhood routes necessary to retain service.
Even that can provide cascading gains. Moving routes to lessen overlap can enable new stops. Benefits can come from replacing a 6 day route (eg 815) with a 7 day route (eg 812). Or an hourly route (814) with a half-hourly route (850). The 813 and 814 upgrade would boost capacity on Springvale Rd, relieving crowding on the busiest section of the 902 orbital. The 814 / 850 swap increases directness and adds a direct station connection for Noble Park North. Seven day service on all routes is another improvement.
View the expanded network below. Useful Network routes are thick lines. Neighbourhood routes that would change are thin coloured lines. Thin grey lines are not suggested for change. For routes 802/804/862 this is because they are part of another area's network that can be revised later.
Brief comments on changed neighbourhood routes are below. Or delve deeper with the interactive map to compare layers and read more detailed per-route notes.
1. Upgrade Heatherton Rd Route 811 service from every hour to every 20 minutes on weekdays between Springvale and Dandenong. Potentially operate this service as Route 810 and terminate 811 at Springvale.
2. Upgrade Route 813 from every hour to every 20 minutes on weekdays between Dandenong and Springvale Station. Resources partly from reducing duplication. Service forms 814 at Springvale Station.
3. Commence Route 814 at Springvale Station and operate to Dandenong via Police Rd, Waverley Gardens and Gladstone Rd. Upgrade from every 60 min to every 20 min on weekdays with 7 day service. Resources to come partly from deleting Route 813 on Police Rd and savings in Route 850 kilometres.
4. Implement supporting local network changes including (i) Route 850 to Noble Park instead of Dandenong Station to replace Route 814 on Jacksons Rd and provide new station connection, (ii) Route 811 to replace Route 814 on Regent Avenue, Springvale, (iii) Route 848 to replace Route 813 on Hansworth St, Mulgrave, (iv) Route 812 to operate more directly to Parkmore and replace Route 815 to Noble Park, (v) Route 824 to operate via Darren Rd and Corrigan Rd to improve coverage, and (vi) simplification in area north-west of Dandenong Station. More details in the interactive map.
5. Provide 7 day service and lengthen operating hours on routes 800 and 885. Also consider 7 day service on Routes 802 and 804, likely in conjunction with network simplification involving Route 862.
Lower priority but still desirable are service upgrades for Westall Rd and Jacksons Rd. These could be achieved by (i) extending all Route 824 trips to Keysborough to benefit Westall Rd and (ii) upgrading the revised Route 850 from every 30 to every 20 minutes for Noble Park North. The latter is best done in concert with network simplification in the Glen Waverley area.
Conclusion
More than the others, this network will cost money. New buses will be needed. And more drivers on more shifts. But there will be more passengers. You only need to see how full buses on Springvale Rd get (photo above) to see evidence of people flocking to a more frequent service in the area. Even on evenings and weekends when a 30 minute service operates. Due to historical patterns of underservice, there would be few places in Melbourne where you would get more patronage for the service added.
What if you wanted to do something and didn't have the money? While they wouldn't meet the Useful Network criteria, boosting the streamlined routes 810, 813 and 814 to every 30 minutes would still double service in many areas. Due to the resources freed up from removing duplication (typically involving hourly routes) the implemention cost would be much lower than the 20 minute Useful Network approach. 30 minute frequencies aren't as marketable as 20 minute services, but usage is still likely to significantly grow given existing patronage and social needs.
This exercise was based on finding the best means to provide the most people with a Useful Network service without greatly inconveniencing existing passengers. It has concentrated on local destinations like Springvale, Waverley Gardens, Noble Park, Keysborough and Dandenong. Trip generators outside the area have not been considered. A wider review could consider improved connections to Clayton, Monash University and Dandenong South with further patronage growth potential. As well it could enable simpler routes and better service around Corrigan Rd, Kingsclere Av and Police Rd.
What do you think? Do you have other ideas? Are there things you see are wrong with the above? Please let us know your thoughts in the comments below.
Post script 31/5/2019
One of the few things for buses in the 2019 Victorian state budget was a new route in Keysborough. The funding for this increases the likelihood that it will be possible to upgrade Greater Dandenong buses with relatively little extra expenditure.
PS: An index to all Useful Networks is here.
2 comments:
Have you tried to calculate the fleet implications?
Also it looks like you've changed or ignored some of the ways that buses currently access the Dandenong Station interchange - is that intentional?
David - I haven't calculated the fleet implications for this one, though more buses would be needed, at least in the 20 minute version. Whereas my Friday items for other areas have been cheaper as they already have more frequent peak buses that could just be used for more of the day.
A 30 min network would greatly reduce the additional bus requirement. Cutting it to zero is likely possible if one redrew the whole network including closing stops where parallel routes are close. Something for a local review to consider but would bring other complications.
I was mindful that some route lengths are more efficient than others if you want to run a particular service frequency. And that there can be severe inefficiencies if train frequencies limit what bus frequencies you can run and still connect. I suspect that those issues are lessened in Greater Dandenong because (i) trains run every 10 min so you can have local buses every 20, 30, 40 or 60 min and still harmonise, and (ii) because so many routes terminate at Dandenong you may be able to interline with other routes if it turns out that an individual route's length is inefficient for the frequency you want (eg the worst case would be a route that you want to run every 20 minutes taking 41 minutes from end to end, whereas 34 minutes would be efficient with recovery time).
Yes, I did make some changes to approaches to Dandenong. They're fairly rough and one could argue their merits. But I wanted the 800 - being the main Princes Hwy route - to run along as much as the highway as possible. I wanted more simplicity around Hemmings St. And it was a trade-off as to whether you run only to the station or to the station via the shops (as it's a bit of a walk).
Post a Comment