Tuesday, May 12, 2020

Timetable Tuesday #74: Flexible route buses - cure or curse?

Local bus services rarely rate a mention in debates about transport in Melbourne. That is even though their potential to cheaply move more people for more types of trips far exceeds elite hobbyhorses like airport rail and (especially) very high speed interstate rail. 

When they do get discussed it's often in disparaging terms. Eg large noisy fume-belching buses meandering through small streets with no one on them. A solution sometimes proferred is their replacement with smaller flexible route vehicles that takes people nearer to their homes. Add an app and it becomes a fashionable part of discussions about 'smart cities' or 'mobility as a service'.

The types of people who further these discussions rarely see the inside of a public route bus. They are even less likely to talk to people who do. Their knowledge of existing networks can be abysmal, even to the extent that some are unaware that 'demand responsive' buses already run and their strengths and weaknesses are established facts. Melbourne has been running flexible route buses for as long as  many people have been alive.


The longest running flexible route bus has been the Telebus series in Melbourne's outer eastern suburbs. There's one cluster around Croydon/Lilydale and, later, another around Rowville. They were established by Croydon Bus Service in 1978 to serve then new subdivisions difficult to serve with conventional bus routes. Read the founder's ATRF paper here. Some historical timetables are on Krustylink (see the 2001 timetable book under Route 672).

The extent of the Telebus network can be seen from the shading on the maps below. Telebus is the only option in some areas while in others (notably Rowville) there are inefficient overlaps with regular bus routes because Telebuses weren't removed when new fixed routes went in about 20 years ago.

What are the differences between regular bus routes and Telebus? Catching a regular bus requires you to find the route you want on a map, look up a timetable and get to the stop a few minutes before it is due to arrive. The timetable will also show when you will arrive, handy for an appointment or connection to another service. If you know where the bus goes and it's a frequent service then you can skip all these steps, rock up at the stop and be on your way in a few minutes. Failing that, if on a main road, it should have a clockface 'memory timetable' with departures every 15 or 20 minutes as per the Useful Network

A Telebus requires you to look up the time the bus leaves its start point. Then you must telephone at least five minutes before the bus leaves. You mention where you want to be picked up from. The dispatcher will then give you an approximate time to be there. The time you arrive at your destination depends on how many other people have requested diversions. If it's many then your trip is slow. This is why this style of service works best on quiet routes, during off-peak times and when arrival times are not critical (eg you are not seeking to connect with trains or other buses). A surcharge, over and above your myki fare, applies if you use the flexible route option.

Telebuses operate between fixed origins and destinations with the route between them subject to the calls they get. Some stops are fixed. You don't need to call ahead or pay the Telebus surcharge if you board or alight from them. However times will vary more than for a normal fixed route bus. Some telebuses operate bidirectionally between two termini through a residential area while others are effectively a loop serving a single key destination or interchange only. 

PTV lists Telebus routes on their website page here. The founding operator started them with love but the CBD-centric Department of Transport treats them like if it's their stepmother, with an obligation to feed but not to love. Evidence of this is plain in the aforementioned link where maps are grudgingly provided (as pdfs) but links to the all-important timetables are not. Finding these requires a search that doesn't always work. Information at stops is limited too.

Links to Telebus timetables are here if you wish to browse them:

Telebus Area 1 Lilydale - Chirnside Park

Telebus Area 2 Mooroolbark - Chirnside Park

Telebus Area 3 Mooroolbark - Chirnside Park

Telebus Area 4 Mooroolbark - Croydon

Telebus Area 7 Ferntree Gully - Stud Park

Telebus Area 8 Ferntree Gully - Stud Park

Telebus Area 9 Stud Park loop

In addition Route 672 between Croydon and Chirnside Park is a regular bus route with a Telebus service operating on a section of it at off-peak times.

What are Telebuses like as a service and are they useful? Their operating  hours are like a typical 1990s Melbourne bus route, ie no service much after 7pm or on Sundays. Rowville area routes also lack Saturday service. Frequencies are roughly hourly and operating patterns vary during peak periods on some. In other words many of the criticisms against local buses also applies to flexible route buses as they are implemented here. So much for the panacea that promoters of flexible route services claim!

The Department of Transport's stern stepmother policy towards buses (ie food but not nurture) again applies. They paid to keep Telebuses running but did not upgrade them, even when many fixed routes gained minimum service standards from 2006. But, with only one or two exceptions, they also did not delete them nor convert them to conventional fixed route services. Few local bus review recommendations ever got implemented in the outer east. These would have substantially changed some Telebus routes. 

Another criticism of regular route buses is they can be underused. This looks particularly obvious when large buses, despite being necessary during peak times, are used off-peak. Are demand responsive Telebuses particularly productive?

The answer is they are not. Infrastructure Victoria regards 20 passenger boardings per bus service hour (ie roughly 1 passenger boarding per km given normal bus speeds) as a minimum threshold for a bus route to be considered productive. The average for Melbourne buses is somewhere over 20. Most Telebuses are half or less of this average. Weekday average passenger boardings per hour are as follows:

TeleBus 1 Lilydale - Chirnside Park 8
TeleBus 2 Mooroolbark - Chirnside Park 7
TeleBus 3 Mooroolbark - Chirnside Park 13
TeleBus 4 Mooroolbark - Croydon Station 10
TeleBus 7 Stud Park SC - Ferntree Gully Station 7
TeleBus 8 Stud Park SC - Ferntree Gully Station 5
TeleBus 9 Stud Park SC - Stud Park SC 8
Route 672       Chirnside Park - Croydon 15

Melbourne's outer east is not good territory for buses, with higher car ownership and lower residential densities than many other areas. Even so fixed route routes in similar outer eastern catchments generally have higher productivity than the above Telebuses. Examples include Routes 675 (33 weekday passenger boardings/bus hour), 671 (22), 677 (21) and 680 (18). Lower productivity fixed routes eg 663 (15), 679 (14) and 695 (13) are similar to the better performing Telebuses. However they are still below average for Melbourne buses and serve large areas with low population density with frequency that is arguably excessive.

681 and 682 (11-12 bph) are also poor performers, likely due to catchment overlap with the even quieter Stud Park area Telebuses (5 - 8 bph). It would seem that if you do have both fixed and flexible routes in an area more people will opt for the fixed route, even if its frequency isn't very good.

Demand responsive routes never seem to be associated with more than mediocre patronage. This is particularly in areas, such as Rowville, where there is a catchment overlap with fixed route services.

You may accept such low patronage if you deem the service essential on social equity grounds. For example to provide coverage in pockets with narrow and/or disconnected streets without usable through routes. However the first priority should be to see if a fixed route providing reasonable coverage is possible. This is because even in high car-ownership areas with unfavourable demographics for buses fixed routes seem to be more popular.

Such fixed routes might still record a very ordinary 15 to 20 passenger boardings per hour. However that's still better than flexible route buses. Also it is unlikely that, contrary what some have suggested, that there would be much of a saving if fixed route buses were converted to flexible route, unless you were also cutting service standards like hours and frequency or drivers' wages. Be wary of accepting  cost saving claims from exponents of flexible route buses until you are sure that you're comparing like for like service. 

Demand responsive 490

A bit over ten years ago sixteen bus reviews, covering all local areas of Melbourne, were done. Occasionally recommendations included demand responsive buses in hard to serve pockets. The only Melbourne one that got implemented was Route 490 in Gowanbrae. 

From a transport access point of view the suburb should never been developed since it is not geometrically possible to provide an efficient direct route, or, cheaper still, have rerouted or extended an existing service. A Glenroy to Airport West route, for example, would have provided a handy link across the north and given some useful connections for Gowanbrae. However politics overcame history, geometry and geography. A successful local campaign eventually saw a demand responsive bus to Airport West put in (more here). 

Although a similar concept to the Telebuses, the 490 is slightly different. For a start it didn't get the name 'Telebus' - this was associated with the Invicta routes in the east. And there was no surcharge for going nearer your home. Instead you could just go to designated stops (explained on the old PTV website's map, but not explained on their new site's map).

Street layout restricts where the 490 can go. It runs to the nearest larger shopping centre and tram terminus at Airport West. Despite being introduced in 2008 service levels are again like a 1990s regular bus route with only daytime Monday to Saturday service. Services are hourly with a half-hourly weekday peak service. Route 490's patronage productivity is 13 passenger boardings per bus service hour on weekdays and 9 on Saturdays, again well below average for buses in Melbourne. 

Mornington Peninsula community buses

The Mornington Peninsula has large suburban areas that never received the regular bus network coverage that other fringe suburbs have received. Elsewhere on the peninsula are low density peri-urban areas. This has created larger service gaps than other areas have. To fill some of these gaps for the most needy of its population (which it defines as seniors and people with disabilities) the Mornington Peninsula Council runs a 'dial a bus' community transport network. The network is extensive but the service is sparse, with different routes operating on different days of the week. This means that the network works for shopping and certain appointments but cannot perform the school and work travel functions that you'd expect from a regular route bus system. 

Flexibility - on whose terms? 

We already run flexible route / demand responsive buses in Melbourne. As well as regional area 'Flexiride' services in Yarrawonga and Woodend. Some routes have been going for years. Not that longevity is any guarantee of usefulness since the Department of Transport funds but rarely deletes underused routes. The apps aspect may appear novel but existing fixed route buses can also be tracked and real-time information provided. Their patronage has not set the world on fire, and on important service level features such as operating hours and frequency they often inferior to regular bus routes. 

Except for Gowanbrae the metropolitan areas that our flexible route buses operate in have not had wholesale service reviews implemented for years if not decades. If they did then it is possible that some flexible routes would have been converted to fixed routes operating over more hours of the day to provide an overall more direct and useful network. They may be more productive and thus more cost-effective than the low numbers seen for 'flexible' routes. 

As for flexibility, it's important to ask whose terms it is on. That's a familiar debate for those who follow industrial relations. Those who wield the most power have the greatest flexibility, leaving those who don't to accept what's offered. When there's a big power imbalance the stronger side (naturally) prefers direct negotiation while the weaker side generally prefers external arbitration and/or government-regulated conditions to stop the powerful side getting away with too much.

Hence the wage breakouts in 1974 and 1981 (when unions were strong) and hostility towards government moves to pause or regulate wages. Unemployment, casualisation and international competition have since weakened unions to such an extent that it's the (now) more powerful employers who, since 1985, have advocated direct negotiation and the winding back of regulated pay and conditions. The side that initially wanted flexibility to go above and beyond a previous agreement may find that their opponents use that same flexibility to go below and pay less if market conditions change.

To summarise, flexibility is good if you've got power and can dictate its terms. If you don't have power you'd rather inflexibility to safeguard against things getting worse. This concept is less familiar but equally important for transport and land use. Possibly even more so because a benefit of inflexibility (for something like a popular railway line) is you can make long-term decisions like a house purchase with a reasonable belief that transport will continue to be available long-term.

The main power passengers have is to withdraw their patronage. Which they did during several decades where service levels declined, strikes took over the system, the suburbs spread and more bought cars. As public transport became increasingly subsidised even this withdrawal of patronage mattered less. That is until you start to value public transport for its community benefits (eg congestion relief, enabling connections to opportunity, reduced carbon emissions, more efficient land uses) which tend to be highest when it is well used.

If you want that then you have to think about flexibility on the passengers' terms. One type of flexibility that may seem passenger-friendly (eg to door service) might be less so when you consider offsetting liabilities such as having to call ahead before travelling, slower trips, buses twisting and turning and less predictable arrival times.

The flexibility that counts as far as the passenger is concerned is frequent service, especially one where they can turn up and go and easily change between services with minimal waiting. Passengers also like good operating hours, direct routes, and predictable travel times. Surveys have repeatedly revealed these as priorities for most passengers.  

A good fixed network can deliver these to population catchments measured in millions for a reasonable cost. It does that by concentrating demand along a grid of frequent and well-served corridors everyone can walk to and interchange between. And because service is concentrated you can justify infrastructure upgrades such as bus shelters, good information at stops and bus priority to further speed travel and make buses more tram-like

In contrast flexible routes encourage dispersed demand, slow and meandering travel while tending towards lower frequency. Having to phone ahead or register on an app one's intention to travel is a barrier to use, especially if an hour's or more notice needs to be given. That's flexible for the bus operator but not at all for the passenger.  All these are in the direction of lesser usage, lesser demand, high running costs and even its eventual closure such as for this failed Sydney trial.  

Having to phone ahead or use an app is a barrier to using a flexible route bus. Especially if the argument for providing the service is largely on social inclusion rather than patronage or environment grounds. It is not uncommon for low income people to run out of mobile data. Some seniors, in particular, do not have a Smartphone to begin with. These are important considerations if you were hoping that your demand-responsive bus would be completely app-based. If you had a phone option then that's another management cost unless the service is small enough for the driver to take calls during his downtime (another source of inefficiency if the bus is sitting at a terminus). 

Transport planners, who by possessing a degree-needing full-time CBD job are in themselves elites, need to be particularly aware of the financial circumstances of their passengers. Business consultant types, with a private corporate rather than public service ethos, may be even less aware. The Grattan Institute recently reported that 10% of households had less than $90 cash in the bank. It is common for passengers to maintain low balances on their myki cards and make only small top-ups each time, lending weight to Grattan's findings. 

It's important to ask who is pushing flexible route services and how they will deliver claimed cost savings from an inherently low-productivity mode. Is it app developers who want a percentage cut of referred mobility services? Transit-illiterate tech-heads? Do some see a business opportunity in replacing moderately paid bus drivers with low-paid Uber-type mini-bus drivers on local routes? Or is it those who want to slash  apparently inefficient local bus services but are playing a pea and thimble trick to disguise a service cut in the name of 'flexibility'?

A place may exist for flexible route smaller buses. However I suspect that it will be in areas with very sparse demand or unhelpful street layouts. In other words only ever a small proportion of the network. In other areas it's better to reform local bus networks along Useful Network lines to make conventional fixed routes work better. Flexible route buses may help with a particular local issue but are not a wholesale panacea. Contrary to the hopes of people like Infrastructure Victoria's Michel Masson, their implementation will rarely free resources to boost buses elsewhere. 

A big problem is that flexible routes don't scale up well if more people start using the service. Which is exactly what you want if you want buses to be efficient, useful and economical to run. The more passengers flexible routes get the more deviations they make. That slows travel for everyone else on the bus. Hence we have Route 672 that only operates as a Telebus during off-peak times. Though even that adds complexity to the service.

To some extent it's self-regulating because some passengers will get frustrated by the slow speed and quit using buses, making travel quicker for the remaining passengers once again. However that's not a sound approach to encouraging sustained usage or growing public transport's share of trips. One solution is to reduce the number of places it will deviate to, reducing variability. Eventually it might become a conventional fixed route that is even more efficient to schedule due to consistent running times.

Whenever someone (even if they claim to be a transport expert) mentions the word 'flexible', always counter with 'Flexibility for whom'? And ask if it's necessarily a good thing. It might not be. In many cases it may be that our unloved fixed route bus network is not inherently bad, but it just needs some more attention for it to deliver real flexibility on the passengers' terms. 

An index to all Timetable Tuesday items is here.

You might enjoy these well-regarded books on transport topics

Better Buses, Better Cities: How to Plan, Run, and Win the Fight for Effective Transit Steven Higashide 

The Public City: Essays in honour of Paul Mees Gleeson & Beza

A Political Economy of Access: Infrastructure, Networks, Cities, Institutions (Access Quintet Book 4) David Levinson

Human Transit: How Clearer Thinking about Public Transit Can Enrich Our Communities and Our Lives Jarrett Walker

Transport for Suburbia: Beyond the Automobile Age Paul Mees

(Sales links: I get a small commission if you buy via the above - no extra cost to you)


Anonymous said...

What ever happened to Telebus areas 5 and 6?

Peter Parker said...

I think there were some previous TB workings. Eg around 671 & 680? Can't find any documentation though.

Marcus W said...

How is the Telebus surcharge collected from passengers? It would presumably have to be cash, given myki doesn't support anything beyond 'standard' fares.

Peter Parker said...

Some mention is in the main manual here but not payment methods.
https://www.ptv.vic.gov.au/footer/legal-and-policies/victorian-fares-and-ticketing-manual/ Had a quick look on Ventura's website and cound't find anything. But you might find something in the 2001 timetable book that I linked to in the post.

Heihachi_73 said...

Regarding TeleBus surcharges, back in the 90s when I still lived in Mooroolbark you paid the driver the phone surcharge on top of the fare, I seem to remember it being an extra 20c (it was probably more for full fare, but I was only a kid thus got concession rates).

The 675 bus has extra patronage compared to regular TeleBuses most likely because it's a direct link between Chirnside Park and Mooroolbark without all the meandering the normal TeleBuses have, unfortunately it's still a TeleBus at heart and runs the same buses and the same useless hours and of course nothing on Sundays. 671 used TeleBuses as well, and just like the 675 it was a normal bus route with no deviations.

The 680 always had big buses ever since I can remember (or midibuses as they were back in the 90s/00s, it was actually one of the first routes in Melbourne to get low floor buses, or LoRiders as Invicta called them). It was upgraded when the Lilydale TAFE was built, and for a while it still terminated there even when the TAFE had closed down around a decade ago (before it reopened), making the terminus a bit pointless. The 680 still follows its school heritage and only runs Monday to Friday despite being the only service in the Hull Rd area.