Tuesday, October 29, 2019

Timetable Tuesday #46: Chadstone and the temporary Route 700 Shuttle


Weekends are the busiest times for major regional shopping centres. They are when their core working affluent customer base is most likely to be free to shop. Especially in the lead-up to Christmas parking is at a premium and surrounding roads are clogged. 

So people may jump at the opportunity for alternative travel options if they're reasonably available. Especially at bigger centres where the focus is more on experiences (eg cinemas), services (eg whatever women get done to their nails) or small/light items (eg clothing and jewellery) rather than groceries or bulky goods. 

Public transport mode share to large centres is higher than to smaller centres but still quite low. Many shopping centres are remote from stations. Key shopping centre bus routes,  such as 468 to Highpoint and 800 to Chadstone, do not run 7 days. Other routes are infrequent. Nevertheless parking pressures and the retail sector's dependence on young casual employees (not all of whom drive) present opportunities for public transport if a job-ready service is provided

Where the 7 day frequent service is 

The Melbourne Public Transport Frequent Network Map extract below shows where the 7 day frequent service is. It's largely confined to trams and a few lucky train and bus corridors. Most are near the CBD. Only a handful pass regional shopping centres or connect them to key catchments. 


What frequent service do the big shopping centres have

Centres like Highpoint, Box Hill and Southland have 7 day frequent service in two main directions. Eastland, Dandenong, Knox City, Frankston and Chadstone have it in one. The vast majority of centres including Watergardens, Werribee Plaza, Doncaster, The Glen, Forest Hill Chase, Craigieburn, Epping Plaza, Fountain Gate, Cranbourne have it in none. Only a small fraction of people within even 5 km of their nearest large centre have 7-day frequent service to it. 

Major centres need frequent service in at least eight directions

To reasonably connect major inland centres to their immediate surrounds you need frequent service in at least eight directions. Even more directions would be nice. And this would be appropriate for somewhere with a large catchment (eg the CBD). But beyond a certain point the overlaps become so great for increasing distances that it becomes more efficient to add frequent cross routes with stops at intersections to easily feed radial routes. 



Getting to 8 directions is cheaper than you might think. Some centres, like Doncaster and Box Hill, are at or close to this on weekdays. Weekend frequency boosts may be the main things required. 

A simpler bus network with fewer but more frequent routes could help other centres. For example Chadstone currently has a 15 minute service in four directions (north, south, east and west) on its two SmartBus routes as shown below.    



There are large 'holes' on popular corridors where there is no frequent service to Chadstone. Some can be plugged by reforming the existing network to reach our desired eight directions. For example (easiest first):

* Princes Hwy to Monash University: Evenly space 802/804/862 trips for a combined 15 min service. In longer term simplify to two 7-day routes every 30 min each, evenly spaced. 

* Ferntree Gully Rd: Extend Route 693 to Chadstone. Stagger to evenly space with 742 trips for combined 15 min service. (Longer term network reform discussed at Useful Network Part 16). 

* Neerim Rd: Remove Route 623 from Dandenong Rd (other routes eg 900 have since been added). Route it and all 624 trips via Murrumbeena and Neerim Rd with the half-hourly services staggered by 15 minutes to provide a frequent service. Longer term merge routes for simpler service. 

* Murrumbeena Rd/East Boundary Rd: Network reform based on consolidating 822 with the new 627 to provide a simple 15 minute service with other local changes. 

Other corridors with merit involve 767 north along Huntingdale Rd (double frequency of Chadstone - Box Hill portion), 800 via Dandenong Rd (increase frequency from 20 to 15 minutes and 623 east along Waverley Rd (boost frequency). This would deliver eleven frequent corridors with benefits for numerous other smaller shopping centres and the network generally in the south-east. A map is below. 



These simplified corridors would also be more amenable to weekend frequency upgrades. Priority routes, based on existing high usage, could include Wellington Rd (900), Princes Hwy (800, 802/804/862) and Neerim Rd (623/624). 

Existing popularity of buses

So much for what could be a model future network. What about now? 

We saw above the limited frequency of buses to our big shopping centres, particularly on weekends. Timetables can also be unreliable, largely due to a lack of bus priority. This means that cars of one or two people get to delay buses with 40 or 50 riders near centre entrances. 

Despite all this weekend buses to suburban shopping centres are still popular. Some routes record their highest passenger boardings per bus hour results on weekends rather than during the week. The mismatch between service and demand has led to severe weekend crowding. For example Route 900 via Chadstone Shopping Centre has over 80 boardings per bus service hour on both Saturday and Sunday where it runs every 30 minutes. This makes it our most productive weekend bus route and well above the average of around 20. Nearby route 903 also has high usage in the Chadstone section.


Boost existing routes or introduce another? 

There are several ways you could fix this problem. 

The most obvious is to increase weekend frequencies on key routes serving major shopping centres. Candidate corridors for Chadstone were mentioned above. If you were to only upgrade one route it would be likely be the 900 due to its very high boardings per bus hour, especially on weekends. Secondly one might add short trips to augment existing 903 services where they are most stressed. Thirdly you would give 800, 802 and 804 seven day service and boost weekend frequencies on popular routes like the 623 (currently hourly). 

Another approach, similar to that done for the universities, is to leave existing route timetables as they are and introduce a weekend shuttle service from a nearby station. That would improve connections from the nearby Dandenong line to Chadstone. It could relieve crowding on existing through routes like 900. However its benefits are confined to those who can walk to a Dandenong line station. Plus you're adding another layer of complexity to an already complex network with a part-time route. 

The new Route 700  

Despite above reservations, a dedicated shuttle is the option favoured by the powers that be. A trial Oakleigh Station - Chadstone weekend shuttle bus recently commenced service. It's route number 700, a reused number from the pre-orbital days when Ventura's Mordialloc to Box Hill route was the main service between Oakleigh and Chadstone. 


Route 700's map (below) appears on the new PTV website but not the old one. A written route description appears on the old PTV website


Timetable

Route 700 has some unusual features. The first is that its Melbourne's only weekend-only bus route. The only route that comes close to this pattern is the 695F to Fountain Gate which runs Friday afternoons, Saturdays and Sundays. If you wish to travel from Oakleigh to Chadstone on a weekday you'll still have to negotiate numerous regular routes (which together provide a very frequent though complex combined service). 

Apart from the 695F dedicated weekend routes were phased out about 10 years ago. They date from when regular routes ran Monday to Saturday. In some areas (eg around Frankston) a less direct Sunday route would operate to provide bare-bones coverage at times the regular route didn't. Or the Sunday service may be in the form of a 7-day route being extended beyond its regular terminus to serve an area that had only 6-day routes. Examples include Sunshine West and Melton. 

Secondly 700 is a seasonal route. It started last week. It will run through the pre-Christmas period then through to the January sales. It finishes late January. 

Thirdly it introduces (another) unique public holiday pattern. With two exceptions (681 and 682) all routes that run on Sundays also run on Christmas Day. 700 will increase that number to three. This is because Christmas is not a trading day at Chadstone. 

Next Tuesday (Cup Day) Route 700 will operate to a special timetable with service from 9am to 7pm. This is another variation since it's not a straight Saturday or Sunday timetable that most other bus routes run on public holidays.  

What about operating hours? 700 start at 8am on Saturday and an hour later on Sunday. Last buses from Chadstone leave just before 10pm on Saturday and 8pm on Sunday. These are an hour earlier and later than centre opening hours. I'm guessing that the special holiday patterns reflect Chadstone's core trading hours (some shops eg supermarkets may open outside these). 

700's frequency most of the day is 8 minutes. However it doesn't reach this until an hour or two after first bus. Earlier trips are every 16 minutes. This shows that the route is intended for shoppers rather than workers some of whom would need to travel at these earlier times. However the 8 minute frequency is generously sustained until last bus.   




Run times. They're critical to the success of a timetable. 700's are flat from morning to night. 7 minutes from Oakleigh to Chadstone and 9 minutes from Chadstone to Oakleigh is allowed. The arrival time on one timetable matches with the departure time on the other. Without dwell times the emphasis seems to be on keeping buses moving.  

The constant (and possibly optimistic) run times appear not to take into account likely traffic congestion or longer passenger boarding times when the service is busy. There will need to be buses in reserve to ensure that the advertised regular 8 minute frequency is maintained even if travel times are not. The service could fail if there aren't. 

Seeing how all this works in practice will be very interesting, especially as we approach Christmas. Active service management is essential. Cities that take buses seriously do this. We're less experienced at this; buses here mostly serve low frequency/low capacity local transit and feeder functions with less need to treat them as high frequency/high capacity rapid transit.

Given the volume of irregular/seasonal passengers, both Chadstone and Oakleigh probably need staffing to manage traffic, funnel passengers onto the shuttle (or other routes) and provide next service advice. 

Improved passenger information is also desirable, given that (i) Oakleigh Station is currently a construction site and (ii) it's something missing from major interchanges, eg with individual bus stands lacking maps or directions to other routes. We could learn from experience gained from rail substitute bus services with their heavy flow of people making unfamiliar trips.

Conclusion

This has been our look at the new Route 700. While the university shuttles have been undeniable successes, is the same approach right for Chadstone? Or would it have been better simply to upgrade regular routes like 900 rather than adding (yet another) part-time route and making the network more complex?

And what about those run times? Are they reasonable or not? Please leave your comments below if you have thoughts on these or other topics related to the 700 and Chadstone buses.

PS: Back in 2011 I favourably reviewed Human Transit by Jarrett Walker. It talks a lot about public transport network design topics like we cover here. You can buy it via the link below. The small commission I receive from purchases helps support Melbourne on Transit



Sunday, October 27, 2019

How high productivity can wreck a bus route


Last Sunday I wrote about our ten most productive bus routes - that is those that had a high number of boardings per service hour. They were basically concentrated around universities and in the City of Wyndham (which had a major bus network revamp a few years ago).

Productive buses are great. Their usage means that people are finding them useful. They bolster fare revenue while being cheap to run. If you're a little bus company who runs without government subsidy (like most were up to the 1970s) this is just what you want. 

However is there such a thing as bus routes that are too productive for their own good?

I think there is. 

Especially if you want public transport to do all the good things it can only do by maximising its patronage. Like taking cars off the road, free road space for those who need it, lessen parking pressures, support denser retail and jobs in our town centres, enable more space-effective cities, lower overall community expenditure on transport and make our economy less carbon-intensive.  

Certain routes have incredibly high boardings per hour figures on weekends. These are the 900 (over 80 boardings/hour) and the 733 (around 60 boardings/hour). Both serve Monash University Clayton, with 900 roughly east-west from Caulfield and 733 north-south from Box Hill. When you get to very high numbers the service deteriorates, with delays due to boarding/alighting and, in extreme cases, passengers being unable to board.   

Why is productivity so high? Both routes combine very favourable passenger demographics with low service levels. For instance 900 is half-hourly while 733 is hourly on a Sunday. 

You could quadruple 900 to every 7.5 minutes and triple 733 to every 20 minutes on a Sunday and still meet Infrastructure Victoria's 20 boardings per hour productivity threshold. That's guaranteed - ie it assumes that the vastly better frequency will not attract a single extra passenger. 

In the 'real world' things behave differently, especially if the route serves a fertile catchment. The better the service, the better the patronage. Though it's not necessarily a 1:1 relationship. For example you might only get a 40% increase in patronage for a 100% increase in service. 

Although there have been times when we've beaten that. For instance in the few years after serious local bus upgrades started in 2006 we got roughly a 25% increase in patronage for a similar increase in service. And that's largely just with service upgrades, not with network reform that can be more cost-effective. 

Getting back to the 900 Sunday example, if we quadruple service from every 30 to every 7.5 minutes, we increase it by 300%. Even if elasticity is quite low (33%) we will have doubled patronage (100% increase). The productivity will have dropped from over 80 boardings per hour to a still impressive over 40 boardings per hour. There will be some knock-on gains flowing to intersecting routes (as trips involving transfers to the 900 will be easier) along with achieving all the other things associated with high patronage mentioned before.  

Let's say you were to keep increasing service. You will reach a point where frequency is already so good that big improvements won't attract more passengers. Eg the difference between every 3 minutes and every 4 minutes would probably not be noticed. But the 3 minute service requires 33% more buses to run than the 4 minute service. That's hard to justify unless crowding requires the capacity the higher frequency provides.

When frequency gets stupidly high your boardings per hour might fall to make the route unproductive. Unless you can find some other means to justify this (eg wider benefit of everyone arriving by bus and not driving) you would back off, run articulated buses if required for capacity and deploy the drivers on other routes that need a frequency boost. Eventually you'd upgrade your original route with a bus lane, busway or light rail, with the operational benefits of higher productivity.

The graph below might explain this better:

The left scale is productivity in boardings per bus hour (red line). The pink area is below Infrastructure Victoria's 20 boardings per hour while the green area is above it. Our network average is a little higher (faint line). 

The right scale is patronage (blue line). Across the bottom is frequency. That's simplified; a more precise measure could be service quantity which also accounts for operating hours. However this has its own complications given that seperate graphs are really needed for day type (eg weekday, Saturday, Sunday). 

Especially in areas where the catchment is favourable for buses and/or not many people own cars productivity peaks well before patronage. Routes 733 and 900 on weekends are extreme examples with very high productivity despite their infrequent service. Even with modest patronage elasticity you can vastly boost frequency, still have a highly productive route and lift average network performance. But only to a point, as mentioned before. 

Each route will have a different curve shape. And the frequency scale doesn't apply everywhere.  University shuttles (eg 301, 401, 601) combine extremely high productivity with extremely high frequency. In contrast routes in sparsely populated areas may not be viable at any frequency but are run due to a policy decision to retain coverage. And some might not attract many more passengers even with doubled service, meaning a sharp fall in productivity to below viable levels.

Fortunately this is hardly a worry for most of our most productive 40, 50 or even 100 routes that are likely to respond favourably to a service upgrade, especially if efficiency-enhancing network reforms are done simultaneously. If we do those we might be able to increase frequencies on selected routes with the existing bus fleet and a less than proportional increase in driver hours required. 


Below is an attempt to cluster different types of routes according to their productivity and patronage. 



Cluster A: Very productive routes with high occupancy per bus but less overall patronage and service. For example they might not be very frequent and finish at 9pm or earlier. Their catchment is favourable for bus usage and is likely to respond well to service upgrades. 

Weekday examples with at least 40 boardings per hour include 150, 151, 160, 167, 170, 180, 190, 192, 201, 237, 270, 279, 302, 406, 410, 423, 424, 494, 495, 497, 508, 529, 533, 536, 537, 570, 630, 703, 733, 737, 813, 814 and 893. 

Some routes are more crowded (ie productive) on weekends than weekdays. This is because they might drop to half or even quarter frequency even though demand for travel does not drop in the areas they serve (eg major shopping centres). Such routes may qualify for Cluster A on weekends but not weekdays where service is better matched to demand. 

Productive routes with at least 40 boardings per hour on Saturday and/or Sunday include 150, 279, 302, 508, 623, 630, 631, 733, 767, 800, 900, 903 and 907. It is likely that busy segments of longer routes like 216/219, 220, 901 and 902 would also qualify.  

Cluster B: These routes rank amongst Melbourne's best used. They have better service than Cluster A. While productivity might be slightly lower than Cluster A (especially for the more frequent examples) it is still above average.  

Examples include busier ex-Met corridors eg 200/207, 234, 246 and 250/251 which typically have 7-day service until midnight and better frequency than routes in Category A. You would also include weekday services on SmartBus routes. The SmartBus orbitals are a mixed bag as all three have low-productivity segments. 

This is the 'sweet spot'. At least on weekdays all of Melbourne's most used bus routes are in this cluster. Their quality of service is the best on the network and their productivity is above average. You need to bring as many routes as possible into this group if you wish to maximise network patronage and the number of passengers carried with your vehicle fleet. 

Cluster C: We don't actually have any of these in Melbourne, at least with the frequency scale shown. Our two super-frequent routes (401, 601) are university shuttles and their productivity is excellent. Our stronger uni shuttles (301, 401, 601) are really Cluster Bs but at a higher frequency point whereas the low frequency Deakin Uni 201 and 768 are Cluster A ('could do better'). 

If you were to adjust the frequency scale to reflect what would be considered extraordinary for a sparsely populated area, then you would find some Cluster C examples. For example 582 and 695 whose 20 to 30 minute frequencies appear very high for their catchments. The same applies for parts of the 901 and 903 orbitals that provide weekday 15 minute frequencies and service until midnight to some low density industrial, residential and rural areas that don't need it. Efficient planning would transfer service from Cluster C routes to others that need it more.  

Getting from A to B (but not C)

Very high productivity in a bus route isn't everything. 

By not operating routes with favourable catchments as frequently as they should be we are denying ourselves millions of passenger boardings per year and many hours of productive revenue service.

We need to efficiently get our productive high patronage potential routes from A to B by reforming networks and boosting service. If a route is already operating at 40 or more boardings per hour it is possibly under-serviced. 

Just mind the curves above. 

PS: Back in 2011 I favourably reviewed Human Transit by Jarrett Walker. It talks a lot about public transport network design topics like we cover here. You can buy it via the link below. The small commission I receive from purchases helps support Melbourne on Transit




Friday, October 25, 2019

Building Melbourne's Useful Network: Part 25 - Buses for our booming "ethnoburbs"


Much has changed about our outer suburbs. They're dense, they're multicultural and they're growing fast. For all the talk about CBD apartments and urban consolidation, outer suburbs still account for most of metropolitan Melbourne's population growth.

They're no longer just about young homebuyers; renting is now also big business. Rentals outnumber homes owned outright by as much as four to one, with consequences for an area's socio-economic mix.

With more than half their residents born outside Australia, some of our big outer growth areas have been dubbed 'ethnoburbs'. Melbourne examples include Point CookTarneit and Craigieburn West, whose network I discussed earlier this year.

Because they're in politically safe seats politicians can take them for granted. This is why local upgrades don't feature in this marginal seat list . The quality of recent local political representation has possibly worked against western areas like Tarneit and Melton. And when infrastructure (like road/rail grade separations) does come projects are later, uglier and cheaper than what marginal voting suburbs in the east and south would get.

Planning doctrines that favour segregated land uses and coarse road grids has meant that outer suburbs generate huge outflows of traffic via a few critical bottlenecks.  What passes as local jobs have been zoned out of residential areas into isolated industrial estates with minimal transit access and pedestrian connections. Many jobs are unstable and unemployment is higher than average.

Road systems in growth areas are often stressed. Their populations participate in work or education at higher rates than the state population (due to both the skilled migration program and a younger demographic). This increases travel demand across all modes for most types of trips.

1950s to 70s suburbs were built with local primary schools within walking distance of nearly everyone. Many closed in the '90s as birth rates fell, populations aged, enrolments shrunk and governments slashed. Today's new suburbs are getting schools but they're more widely spaced and clustered to include pre-primary, primary and secondary years. This, the rise in private schooling and parental concerns about children walking alone has meant that nearly all take motorised transport to get to school.


It's the same with other services such as libraries. The outer suburban City of Melton has just two  libraries for its 150 000 residents (or one per 264 square km). Whereas the inner suburban City of Yarra has five libraries for its smaller area and population (one per 4 square km). Again the sparsity of services is generating demand for motorised trips. Add that to the raised population density like we're seeing in these new suburbs and you get increased congestion and difficulty accessing jobs.


Most outer suburban homes have garaging for at least two cars. We tend to assume that there is one car per adult in these parts. After all, you need a car to live in these parts. How else could one get to work to pay the mortgage or even the car loan? And that is largely true in some suburbs like Mickleham in the north and many in the outer south-east.

However other growth sububs, like Craigieburn West and Tarneit have many adults without cars, as this Charting Transport analysis shows. And we're seeing more renters in areas like these as well.  These demographics have implications and opportunities for bus usage and demand, as we'll see later.


What hasn't changed about the outer suburbs? The typical bus service that most of them get. The basic formula is still a bus every 40 minutes along an indirect route because developers didn't complete the road network or leapfrogged their estates. And buses can be a long time coming.

To be fair though, while residents of Mandalay Estate with their one bus per day may beg to differ, it's improved slightly from the 40 year backlog suffered by the City of Knox that to this day still has main road routes also with just one bus per day. Disparate bus operators back then were like dogs, marking their territory by providing a token but basically useless service that subsequent governments never saw fit to improve or scrap.

Consequences for buses

What are the implications for buses in our booming "ethnoburbs"? While they don't necessarily carry the most total passengers, they rank amongst Melbourne's most productive, as measured by boardings per bus hour figures obtained from the Department of Transport. Crowding on them has put paid to the idea that outer suburbanites don't ride buses. And the high boarding numbers don't always just apply during peak but can affect off-peak and weekend periods as well.

We discussed our most productive bus routes on Sunday. Let's recap the numbers. 

Heavily university oriented routes (the 601, 301 and 401 shuttles plus the longer 733 and 630) account for five spots. All the rest are "ethnoburbs" routes. That is 495, 529, 180, 167 and 150. Only 529 existed ten years ago with 150, 167 and 180 starting barely four years ago. Suburbs served include Point Cook (1 route), Craigieburn (1 route), and Werribee/Tarneit (3 routes). 630 and 733 are the two oldest routes but neither has had a substantal off-peak service boost in the life-spans of almost all of their student (and even lecturer) passengers. 

What about the next 10? Same story. University serving routes (406, 900, 703, 201) are again prominent. Ethnoburb routes (160, 151, 494, 170, 192) again occupy five places. The only non-university / non-ethnoburb routes to feature is the 318, which is a peak-only commuter route whose number is inflated by not having return trips. All these have 50 to 55 boardings per bus hour. 

The next 10 (43 to 49 weekday boardings per bus hour) are a little different. This time the tally is university serving (505, 424, 737), ethnoburb (533, 190) and peak-only (740). High performing regular routes include 907, 536, 410 and 508. While the last four do not serve the fringe "ethnoburbs" as defined before, all have significant student and/or low income populations with a high propensity to use buses. More on these in a future post. 

Are there any omissons?

The low boardings per hour figures on long routes need to be treated with caution as parts may be very productive and other parts less productive. If such routes were split parts may rank highly, displacing some listed above. However none are in the fringe suburbs discussed here.

Patronage data also has limitations. For example it's important to know the extent of off-peak and weekend versus peak patronage. Routes in areas with high employment participation may have low usage interpeak. Those in areas with high unemployment may be busy off-peak while routes with lots of teenagers and students may have high weekend usage. This affects priorities for service upgrades.


A new service standard

We need to rethink how we plan bus services in our new dense neighbourhoods, especially those with lower car ownership and a higher proportion of renters.

The current 'hourly until 9pm minimum standard', is probably a reasonable 'safety net' for a low density suburban area. However it is short of what a useful service is for an active lifestyle involving modern working, education and social activities. SmartBuses are more useful but none have been added since 2010, during which time Melbourne gained about a million people and half a million jobs. Much of that growth is in the locations discussed here.

Such areas that are densely populated yet have few community services within walking distance of most people need something better. The boardings per hour results of routes in areas like Tarneit, Point Cook and Craigieburn are now amongst the highest in Melbourne. This indicates that these areas are likely underserviced by current timetables.

In such areas we need a new set of service standards for our busier routes in denser outer suburbs.

Two routes in the Wyndham area are a template for what is desirable. These are 170 and 180 from Werribee to Tarneit. Both have a daytime frequency of every 20 minutes Monday to Sunday. This is approximately double the frequency of most routes in growth areas and meshes in well with local train services. Weekend frequency is better than SmartBus, though span is less. The spread of such services is discussed in The rise of the 20 minuter. It's a handy concept worth extending. 

Operating hours are also important given flexible working hours. The 2006 minimum standards were a great step forward at the time. However their 9pm evening finish and late weekend start (8 or 9am) still make many trips a challenge. Extending spans by a couple of hours at each end of the day would make a substantial difference to the usability of popular local routes. Routes like 180 and some others like 495 go some way with service continuing after 11pm on Fridays and Saturdays but not on other nights.

While adding extra hours at quieter times of the day may reduce a bus route's productivity (because the extra hours may have a lower than average number of boardings), the appearance of long hours routes (eg 190, 279, 302, 900, 907, 908) in the 50 most productive routes list shows that wide spans and high productivity are not mutually exclusive if intelligently applied. The sorts of areas where they could work are exactly where our existing productive routes are. And because the buses are already there, sitting in depots unused, none need to be purchased unless you need to add peak trips.

To sum up, bus routes in high propensity outer areas should run 16 to 18 hours per day with a base 20 minute daytime service 7 days per week. Peak service could be higher where needed, with both the 495 and 529 being very successful with their 10 to 15 minute service.

Which routes? 

Where might you apply a new service standard? I would start with already busy "ethnoburb" routes. With few exceptions they currently only operate every 40 minutes during non-peak times. Here's a list with suggested upgrades influenced by productivity data.

495: Freq upgrade Mon - Sun daytime from 40 to 20 min / extra 3hrs/day span
529: Freq upgrade Mon - Fri peaks to 10 min, Mon - Sun day from 40 to 20 min / extra 3hrs/day span
180: Freq upgrade Mon-Fri peaks from 20 to 10 min / extra 3 hrs/day span
167: Freq upgrade Mon-Fri peaks to 10 min, Mon - Sun day 40 to 20 min / extra 3hrs/day span
150: Freq upgrade Mon - Fri peaks 20 to 10 min, Mon - Sun day 40 to 20 min / extra 3hrs/day span
160: Freq upgrade Mon - Fri peaks 20 to 10 min, Mon - Sun day 40 to 20 min / extra 3hrs/day span
151: Freq upgrade Mon - Fri peaks 20 to 10 min, Mon - Fri day 40 to 20 min / extra 3hrs/day span
494: Freq upgrade Mon - Fri peaks 20 to 10 min, Mon - Fri day 40 to 20 min / extra 3hrs/day span
170: Freq upgrade Mon-Fri peaks from 20 to 10 min / extra 3 hrs/day span
192: Freq upgrade Mon - Fri peaks 20 to 10 min, Mon - Sun day 40 to 20 min / extra 3hrs/day span
533: Freq upgrade Mon - Fri peaks to 10 min, Mon - Fri day from 40 to 20 min / extra 3hrs/day span
190: Freq upgrade Sat - Sun day to 20 min

All rank in Melbourne's 30 most productive routes. Being largely the product of recent network reforms, most are direct along good alignments. Where they are indirect (eg 167, 529, 533) it is because they serve frontier locations, may be the only route in the area and need to deviate to provide coverage. I discuss the latter two Craigieburn examples here.


The off-peak upgrades make use of existing buses. The peak upgrades are the dearest, requiring new bus purchases. However they would be efficiently used on productive routes like these with bidirectional peak patronage patterns. 20 extra buses, for example, could provide thousands of seats to local stations during each peak period, without the land acquisition, development, local congestion and opportunity costs of expanding at-capacity station parking.


Conclusion

Our outer suburbs have changed. The way we've specified buses hasn't very much. But it needs to. Our 'ethnoburbs', especially those with dense housing and low car ownership, are a major opportunity to be growing public transport patronage if the right services are specified and provided.

PS: An index to other Building Melbourne's Useful Network items is here

PPS: Back in 2011 I favourably reviewed Human Transit by Jarrett Walker. It talks a lot about public transport network design topics like we cover here. You can buy it via the link below. The small commission I receive from purchases helps support Melbourne on Transit



Tuesday, October 22, 2019

Timetable Tuesday #45: 216/219/220 bus reform - the good, the bad & the unfinished


We interrupt normal Tuesday programming to bring news of upcoming reform to a major cross-city route group involving what we now know as Routes 216, 219 and 220 between Sunshine, Footcray, CBD and Brighton Beach/Gardenvale. 

Their predecessors operated as seperate routes on the west and south side of the city. See old maps here. Then they were merged to create three through routes in the early 1990s. At their longest extent (before the 2000s) one could catch these routes from the Brighton area to Melton on Saturday afternoons. 

Through-routing pros and cons

Before we go further I should mention the pros and cons of long through routes versus short split routes. Through routes are more economical and provide less duplication of service in a city centre. For example if you worked in CBD north and lived in a southern suburb you could jump on a through-routed bus and get a one seat trip home.

Whereas if routes were split in the middle of the CBD you'd have to change buses. Or maybe not, if the split was at CBD north. However that would inconvenience those who lived in the northern suburbs who would need to change to get to CBD central. To overcome that you'd run their buses to terminate in CBD south. The problem with that is you'd have heaps of overlapping buses in the CBD and the service wouldn't be very efficient. Not to mention the difficulties of terminating buses in areas where land is scarce.


So through-routing seems both attractive and efficient. And it is, provided delays aren't high. But as soon as traffic builds up and buses get delayed then punctuality on long routes deterioriates. Hence calls to split routes to isolate delays. This might help but it's a second-rate solution. If you value the efficient transport that buses can provide then through-routing via exclusive bus corridors is the best of all worlds. Like we aim for with CBD trams. Otherwise you keep having to increase bus run times, slow passengers' trips and make providing high frequency more expensive (which has happened - compare today's slower schedules with old timetables here).

Something that's desirable with a through route is that both sides have similar patronage. This is so you can pick a frequency that's adequate for both without having to terminate too many buses in the city. Otherwise, if you run the same frequency all the way through you get overcrowding on one side and fresh air on the other. Unfortunately the 216/219/220 suffered from this, with the western half underserviced and the southern part overserviced. We touched on the latter back in June.

Recent splits

A few years back, due partly to concerns over traffic levels with the construction of the new Metro tunnel, Routes 216/219/220 were split. Not in one place, but in 216's case, two. 

216 used to start from Burnside near Caroline Springs. It stayed there for years even after the new town centre popped up. When it hit the highway the 216 overlapped the 456 and didn't connect evenly with trains. I wrote about all that in 2010.  This got fixed in November 2016 when the 216 was split at Sunshine. The western portion became the 426 from Sunshine to Caroline Springs Town Centre. It, along with 456, now forms a simpler, evenly scheduled Ballarat Rd service coordinated with trains. 

Another split for 216 (and also affecting 219 and 220) was in the city. Routes from the west terminated in the city. Those from the south terminated at the Alfred Hospital. People could get a tram to continue their trip into the city. It was confusing because there were actually two of each of these route numbers on opposite sides of Melbourne.


Split formalised with another to come 

Some bus changes, like the new Chadstone - Oakleigh shopper shuttles are promoted with fanfare. Others are not so much announced as found out about. Such as the 216/219/220 changes discussed today. To be fair they're almost a month off (17 November 2019) and formal passenger information will no doubt emerge (UPDATE: It has).  However new timetables, gleaned from scrolling the bus route list on the old PTV website, tell us the things we need to know now.

What's the gist? 

In a nutshell the split will be formalised with some new route numbers in the south. 

Service will be transferred from the quiet south to the busy west, where it is most needed. 

Another split at Sunshine will allow 219 trips to be folded into a simplified upgraded 216, which will provide a better-than-Smartbus frequency between Sunshine and the city. The Sunshine South end of the 219 will be replaced by the new 429. There will be no replacement at the Gardenvale end though other routes are nearby. 

Route 220 from Sunshine will see a much-needed Sunday frequency boost from 30 to 20 minutes.  This is at the expense of corridors in the south losing Monday to Saturday frequency, reflecting their low patronage. 

The main points are summarised on the map below. 

The good, the bad and the unfinished

The good

The upgraded and simplified Route 216 from Sunshine and its timetable is excellent. It discards the obsolete Saturday afternoon drop in frequency with the 15 minute morning frequency running all day. The Sunday frequency increase from 30 to 20 minutes is also welcome.  This upgrade fully delivers  the Sunshine - City SmartBus proposed for this corridor last Friday (Useful Network Part 24) on all but branding and numbering.  

Another advance is Route 220's Sunday frequency upgrade. Needed and requested for years, this is the timetable that delivers. The route however remains the same; ideas to transform 220 into a stronger Ballarat Rd 'Megabus' route also appeared on Friday (Useful Network Part 24). 

While not ideal for everyone, the frequency reduction in the south does bring service provision closer to (low) demand. The network reform goes part way to removing poorly used duplicative services in the area.

The bad

Route 429 started with high hopes with its basic frequency harmonised with trains at Sunshine (something that the old 219 was unable to do). It overlaps Route 428 for two-thirds of its length. There may have been opportunity to provide a good combined frequency over the common section. However, especially in the evenings this was not done and there are four consecutive departures where buses on each route leave almost simultaneously. 

428 dep from Sunshine (Mon-Fri): 4:48 5:18 5:40 6:15 6:55 7:30 8:10 9:00pm
429 dep from Sunshine (Mon - Fri): 4:58 5:28 5:58 6:28 6:58 7:28 8:08 8:58 9:58 10:58 11:58pm 

Getting perfect coordination is likely harder with the routes being run by different companies. Nevertheless it was achieved with nearby routes 426 and 456 involving the same operators. However it would probably be better if 428 and 429 were amalgamated into one simpler and more frequent Route 428 as discussed last month (Useful Network Part 21).

Both 603 and 604 continue their 20 minute frequency until last service (after midnight). Not much of the 603 south of Elsternwick is very far from a station with trains also every 20 minutes until midnight. And there remains generous Sunday evening service that not even SmartBuses get. Given the area's demographics and the parallel train line it is hard to see these trips getting much use. Meanwhile other suburbs remote from trains with higher social needs and better patronage potential get no service much after 9pm.


The unfinished

The changes somewhat simplify bus services in the south. They should also save some money to fund needed upgrades in the west. However, particuarly south of Elsternwick, they don't reflect the wholesale network reform that the area needs to cut inefficiency and get people onto buses. 

The southern part of the 603 (in particular) has amazing operating hours and frequency given its proximity to trains and catchment demographics. It would be worth monitoring its patronage performance. If usage is low buses could be shifted towards other routes that deserve the service more. Even in the Brighton area there are significant network gaps including the lack of a continuous route down Hampton St, excessive in number but poor in frequency east-west connections and the absence of an Elsternwick - Nepean Hwy - Southland route.


I had a go at reforming St Kilda, Brighton and Sandringham bus network in Useful Network Part 8. A network like this would reduce duplication near trains (such as what the 603 change exacerbates) while providing more frequent and direct service where trains don't run.

Also unfinished is the matter of the Alfred Hospital terminus. Is this the best place for buses to finish? Or is there scope to use buses to provide connections currently unavailable such as to the north? I discuss how one might straighten routes like 603 or 604 to provide a connection to Burnley St/Victoria Gardens, Richmond here (Useful Network Part 22). 

View the proposed timetables 


Visit Krustylink for how they looked in the '80s (including the old 600-series)

Conclusion

These changes are a welcome first step forward that deliver increased service where it's needed. 

They indicate that maybe, just maybe, this government is starting to acknowledge the worth of network and timetable reform, having spent its first term fending off radical changes for both train and bus. And it wouldn't be a moment too soon! 

What do you think about what's proposed? And should there be a second stage of reform to deliver simpler and better used buses in the Brighton area? If you have views please leave them in the comments below.

PPS: Back in 2011 I favourably reviewed Human Transit by Jarrett Walker. It talks a lot about public transport network design topics like we cover here. You can buy it via the link below. The small commission I receive from purchases helps support Melbourne on Transit



Sunday, October 20, 2019

Melbourne's 10 most productive bus routes and what we can learn


A bus route that carries lots of passengers over its length is not necessarily the same thing as one that's productive. Tack several short routes together and you'll get a single long route that carries more passengers than short routes alone would. This is why, measured on gross patronage, our long orbital SmartBuses (eg 901, 902 and 903) top the chart for the most passengers carried. 

But they're not our most productive routes. They're dear to run with hundred of drivers assigned. And if you were to ride end-to-end on them you'll find both busy and quiet patches.   

Our most productive routes, defined as those with the most number of boardings per bus operating hour, are shorter. They run directly through continuously populated catchments. People will be waiting at nearly every stop. If you were to board you might find yourself standing during busy times and see consistently high use off-peak. And the timetable might not include much service when (a) there are usually fewer passengers eg late at night or (b) there are lots of passengers but they're mainly people without cars who will pack your weekend buses no matter how infrequent.

If you were a private bus operator operating in the pre-subsidy/pre 1970s years you would want your routes to be like that to survive. However the operator's patronage imperative was lost when government funding came in and the taxpayer assumed the risk of low patronage. While suburban bus operators remain privately owned businesses, their routes operate under a regulated regime without the same sink-or-swim competition and ease of entry as (say) a suburban pizza shop. 

Routes and service levels largely determine a system's patronage, fare revenue and operating costs. Even with public subsidies, public transport planning can be business-minded while seeking to efficiently deliver on social goals. Indeed one might regard that as a duty of effective transport planning agencies given their use of taxpayer funds. We saw glimpses of this during the early PTV era when they radically reformed bus networks and had ambitious patronage goals. 


The concept of PTV as an independent transit planning agency proved short-lived with its functions soon folded back into a revived Department of Transport. Service optimisation and reform got sidelined with little substantial network reform in the last couple of years. Though you'd think the government, as the steward of public funds and with significant growth pressures, would jump at opportunities to pursue network efficiency and effectiveness.

Bus network reform here has always been in fits and starts. It will no doubt return. The trick is to keep it sustained long enough for all areas to get revamped networks, not just a few. 

Our productive bus routes

If you wanted to reform buses the first step is to learn from our existing successful bus routes. Here's our ten most productive bus routes, based on weekday boarding per hour figures supplied by the Department of Transport. 

601 Monash University (Clayton) - Huntingdale Station (FUS)
301 La Trobe University - Reservoir Station (FUS)
495 Point Cook South - Williams Landing Station (RN)*
401 North Melbourne Station (FUS)
733 Oakleigh - Box Hill (RU)
529 Craigieburn Station - Craigieburn North
180 Werribee Station - Tarneit Station (RN)
167  Tarneit Station - Hoppers Crossing Station (RN)
150 Tarneit Station - Williams Landing Station (RN)
630 Elwood - Monash University (RU)

(*) Disclosure: In a previous professional life I designed Route 495 and had a hand in some others listed.

To give an idea of numbers, the 601 Monash shuttle had a massive 219 boardings per hour (August - October 2018) with 301 and 495 following at over 70. Others were between 58 and 66.

What's missing?

Some high profile routes are missing from this top ten. Again I repeat that this list is the most productive routes, not the most used routes. Still one might think certain well-known routes should be there.

While SmartBus orbitals have some very busy sections (eg 903's Oakleigh South to Doncaster, 902's Springvale South to Nunawading and 901's Ringwood to Frankston) there's enough quieter sections to depress the boardings per hour average. If the orbitals were split into busier and quieter sections then some of the former may feature on the list. 

The same is probably true for routes like the 216, 219 and 220 that have busy western sections but quiet eastern sections. Plus the 10 to 30 rankings are filled with routes in established areas (eg 406, 410, 508 and 703) that continue to attract patronage but have been pushed down in ranking by upstarts in growth areas. 

Types of productive bus routes

All but one of the top performers can be placed into three categories: frequent university shuttle (FUS), regular route that serves a university (RU) and regular route in an area with a recently reformed network (RN). 

Three of the four top places are frequent university shuttles. These are express weekday-only shuttles between a major university campus and its nearest railway station. All feature a frequent service - typically every 3 to 6 minutes but never worse than every 10 minutes. They are a major success story of recent  bus service planning. Second-placed 301 was the newest to start, commencing in 2016. Because these routes are short a single bus can do multiple trips in an hour. 

Universities are a proven patronage-getter with two other university routes, the 733 and 630, also featuring highly. Both are medium length all-stops routes serving Monash University Clayton. Their off-peak frequency is not particularly high, being 30 and 20 minutes respectively on weekdays and down to 40 or 60 minutes on weekends. Peak frequency is in the 12 to 15 min range for both.

Don't have a university nearby? They certainly help but are not a requirement for a bus route to be a top performer. The other major category of productive routes are products of PTV-era 'greenfield' networks. For example the 495 in Point Cook (2013) and Tarneit's 150, 167 and 180 (2015). All those areas got their old indirect network erased, to be replaced with new, often more direct routes running to new stations. 

Like the busiest part of the 733, routes like the 150 and 180 run directly along main roads between stations on roughly parallel rail lines. This means that both north and south-bound buses can be carrying peak loads. High usage in both directions is why circumferential rail feeder buses can have higher boardings per hour than tidal radial routes like the freeway expresses from Doncaster, that, though busy in one direction, can be quiet in the counter-peak direction.


 Three of the four reformed network routes (150, 167 and 495) only run every 40 minutes off-peak and on weekends. There's little or no after 9pm service. They serve growing housing areas where, since two incomes are often required to get a home loan, may have high workforce participation. This, plus the tendency for schools to be bigger and more widely spaced than they used to be, means high peak usage. In recognition of this Routes 167 and 495 recently gained peak trips. 495 is the outstanding example, with an approximately 10 minute frequency during morning and afternoon peaks. The very direct Route 180 is another good performer with a 20 minute, 7 day frequency provided, thus being more useful for non-peak/non-commuting trips.  All these routes have timetables that aim to meet trains with harmonised frequencies and careful coordination (a job made easier because they serve one or at most two stations, unlike the long orbital SmartBuses). 

What about 529 to Craigieburn West? It's productive because its catchment is so good as discussed here. In that regard it's like the 167 that, while not straight, has a large unique and densely populated catchment.

City of Wyndham - the star performer

As we saw before proximity to a university is a major predictor of a bus route being busy. But there's only a few big campuses around the metropolitan area. We can't magically create campuses to boost bus usage. 

However one municipality with no significant university campus has more productive bus routes than any other. Wyndham has an extraordinary 9 out of Melbourne's 11 most productive routes if you exclude those serving major universities (601, 301, 401, 733, 630, 406, 900, 703, 201). The only routes in the top 20 that serve neither the City of Wyndham nor a university are the aforementioned 529 and the 318 - a city commuter route from Deep Creek with a handful of one-way trips.  


Why? There's multiple reasons for Wyndham being successful. Some can't be replicated elsewhere while others can. Here are a few:

* New suburbs with families on small house blocks, leading to high population densities
* Low to middle income skewed labour force, some with pressure on household budgets where other goals like buying a home, financially assisting family overseas or paying for education may be more important than running a second car 
* More widely spaced schools than in older suburbs, making buses more viable than walking for some trips
* A heavily migrant-based demographic, which due to our skilled immigration program, favours younger age groups with high workforce and education participation, generating demand for travel (including on buses)
* Local demographics include low income/low car ownership pockets in Werribee and Tarneit (who may use buses more)
* Limited local employment including a lack of suburban hubs with high-skill jobs (No equivalent to the Monash precinct in the eastern suburbs)
* Local demographics include areas (eg Point Cook) where many commute by train to CBD jobs and may catch buses to the station (especially 494 and 495). Other outer suburbs have a lower proportion of workers going to the CBD, making them much more likely to drive. 
* A radial rail network with two local lines and direct roads between them, allowing efficient bus routes carrying high numbers in both directions
* Widely spaced train stations with just seven serving over 200 000 people; even if you are on a line you probably will be beyond walking distance of a station
* Parking pressures at stations, with buses being the easiest (or least worst) way to get to one
* A grid street layout that facilitates fast, direct and efficient bus routes along main roads to stations
* Transformative bus networks in 2013 (Point Cook) and 2015 (Werribee, Wyndham Vale, Tarneit, Hoppers Crossing, Truganina) that improved coverage, directness, frequency and coordination while lessening inefficient overlaps (with some further service extensions and upgrades since)
* Network coverage lagging development (notwithstanding recent extensions) with some new routes having large catchments (including some who need to walk over 1km to their nearest stop)
* Underservicing. Current service levels maximise productivity but not service quality (in plain language, overcrowding). More on that later. 

The above goes roughly from wider non-transport factors to bus-specific factors. The latter are easiest to change and are frequently discussed here. 

Opportunities for other suburbs

How are other outer suburbs, demographically similar to Wyndham, doing? I'll go through some of them quickly. 

Craigieburn: Demographics are similar to Tarneit. Two of its bus routes are very productive despite the limitation of only one local train line. Newer areas around Craigieburn West have high population densities for an outer suburb. The main route through it (529) already ranks in the top 10 with the nearby route 533 not far behind. Some local bus routes are not harmonised with train frequencies and opportunities exist for a local network revamp along the lines described here. Scope also exists for coverage improvements and connections to neighbouring lines (eg Mernda line at Epping). However the productivity of such connections won't necessarily be high until there is continuous develoment along the intervening route (as there is with Wyndham's 160, 170 and 180 but not parts of our SmartBus orbitals). 

Melton: Suffers from having one infrequent V/Line train line. Its station, unlike Werribee and Sunbury, is remote from the main town centre. The basic structure of the local bus network hasn't changed for years, with the awkward positioning of the old town centre, Woodgrove Shopping Centre and the station making route planning difficult. Poor road layouts and a lack of rail crossings make efficient routing impossible in some areas while in others a plague of roundabouts impede pedestrian access across roads one might run direct buses along. However opportunities exist for Wyndham-style improvements as soon as trains go to an even every 20 minutes, including direct, more frequent routes along main roads and new routes to improve fringe area coverage. 

Narre Warren/Berwick/Pakenham: Another growth area. Again only has one rail corridor but it's electrified with the basic frequency improved from every 30 to every 20 minutes most times. Bus routes have gradually accreted over years but there has been no thorough-going review since the rail upgrades created opportunities for more frequent 20 minute bus corridors on main roads neatly meeting trains.  Some new areas lack service and there are opportunities for routes to connect the Pakenham and Cranbourne lines via new housing areas like Clyde North. 

Cranbourne: This had a major network revamp in 2016. However its routes aren't yet as productive as Wyndham's. Some service levels are quite generous, even on less direct routes. Opportunities exist  to further integrate its local network with routes on adjoining rail lines such as Pakenham and Frankston. 

Inner and middle suburbs: Different demographics to Wyndham but they contain some productive key routes that haven't had real network or timetable reviews for years (often decades). Opportunities exist due to their low car ownership, grid streets and/or major centres (especially shopping centres and universities) that generate patronage. Many are discussed in the Useful Network series

Limitations and conclusion

Boardings per hour is just one measure of a bus route's performance. It has its limitations and biases. While it's good for cost-effectiveness (since bus hours fairly reflects operating costs), it treats all trips equally, whether it is 1km or 100km. A regional coach on a 200km route might be moderately loaded but its usage on a boardings per hour basis might seem poor. Yet passengers might have an average trip length of 100km so in this regard the route is doing a lot of work. Passenger kilometres could be an alternative measure if this distinction is important. On the other hand routes with very short trips (eg university shuttles) rate very highly with measures like boardings per hour (or boardings per km). Bear this in mind to avoid jumping to some misleading conclusions.

Another thing, even more important, is that while very high boardings per hour numbers may appear good it may indicate that buses are crowded and uncomfortable and that both service and patronage could and should be better.

Yes, very high productivity can be bad for a bus route!

More on this next week.

PS: Back in 2011 I favourably reviewed Human Transit by Jarrett Walker. It talks a lot about public transport network design topics like we cover here. You can buy it via the link below. The small commission I receive from purchases helps support Melbourne on Transit