Melbourne-based readers might have noticed the new bus stops for route 903 being erected all over the suburbs, being for the new ‘Red Orbital’ SmartBus service.

This route runs from Altona to Mordialloc, doing a large orbital circuit of suburbs mostly around 10-15km from the CBD in each direction, is 85km long and will take about 3-4 hours to travel end-to-end and will be the first complete route of the ‘orbital’ SmartBus services to commence operation.

Initially the SmartBus program was to only consist of electronic signs (to tell you the next bus isn’t until tomorrow) but was extended to include service improvements on two suburban pilot corridors, Springvale Rd (routes 888/889) and Blackburn Rd (route 703) thanks to lobbying from the PTUA which commenced in 2002.

The basic SmartBus service level is for a bus every 15 minutes during the day on weekdays, and every 30 minutes at night (until 12am) and on weekends, although the two initial pilot routes fail to meet that with both having hourly evening intervals & finishing early on weekends, and the Springvale Rd corridor only receives one bus every 40 minutes on weekends.

My view is that, although a welcome improvement, these kind of service levels still aren’t good enough, particularly on weekends when the 30 minute intervals are not frequent enough to use without checking a timetable, although most stops have electronic signs to tell you the next bus is up to 55 minutes off!

What’s wrong with the red 903

The eventual SmartBus network is to consist of a series of orbital routes through suburbia, as shown in the slightly outdated below map (a couple of routes were cut back in the VTP). New routes are being rolled out at the rate of about one per year.

SmartBus network map, showing present and proposed routes (Larger version – 600kb PDF)

I do not see the point in these long ‘orbital’ routes and would much rather a set of decent main road routes throughout the suburbs. It is very hard to maintain reliable on-time performance over an 85km route through often-congested suburban arterial roads; the existing 700 Box Hill to Mordialloc service (which is being swallowed up by the red orbital) is already on the boundary of the longest service that can be provided while retaining decent performance.

But what benefit is there from a passengers point-of-view of this extremely long route? The majority of journeys on existing services are of a relatively short (<10km) distance, such as Oakleigh to Mentone, Box Hill to Doncaster or Northland to Coburg. So what benefit comes out of the bus you might catch from Chadstone to Oakleigh South having travelled for three hours from Altona before picking you up?

The answer is that there isn’t one. I also don’t see why only a small number of routes are worthy of high frequency services – and why does the planning of a route from Altona to Sunshine have to be combined with one between Mentone and Mordialloc?

Marketing material for this route states that it will run to a 15 minute weekday and 30 minute weekend/evening frequency. As stated above, the weekday frequency is OK but as this route will connect with no less than eight railway lines, many other bus services and several tram routes, a decent frequency is needed if you want to encourage connections. And two buses per hour ain’t that.

Technology

SmartBus liveried 901 bus at Frankston

The electronic signs are little more than a luxury and are only a nice-to-have; I would rather the money be spent on more bus priority measures, such as segregated bus lanes rather than traffic light priority that often doesn’t work. The PIDs and automated voice inside buses is again a luxury that regularly fails.

New low floor buses have been purchased for most routes, a move which I wouldn’t oppose as far more vehicles are needed to deliver much-needed service improvements. But in my view the mass-purchasing of new buses in Melbourne has mainly been used for one-on-one replacement of older buses (usually cascaded to school runs) rather than adding to the fleet.

No doubt some life-expired vehicles need replacement, many older vehicles IMO should be kept for longer to provide better frequencies, even if they are unable to take wheelchairs.

And if we want to go a bit further, why not consider having less stops along major corridors? Surely the gaps between stops should be different between a local route running down back streets to on a major corridor, yet they are usually similar.

Final thoughts

The latest marketing material states the new 903 will operate from 7am-9pm (midnight other days). Why 9pm? None of the railway lines shut down at 9pm. Why can’t these supposedly “premium” bus corridors at least receive the same 18/7 operating hours and frequencies as the rail network (often criticised for poor service levels) does?

Why is it that the I can turn up at Wattle Glen at 1:07am on a Sunday night/Monday morning to catch the last Hurstbridge service, yet services on all of the “SmartBus” corridors have ceased 3-4 hours earlier?

And it is unfortunate that other routes aren’t receiving similar treatment – the ‘minimum standards’ upgrades applied to many routes only mandates than a minimum hourly service until 9pm each day is provided. If the services to where people live aren’t good enough, this impacts on the patronage on trunk corridors even if they get good service levels.

One good example of this would be the route 900 SmartBus to Rowville which often gets poor loadings east of Monash University. Some pro-rail campaigners may use this as an example against buses to say trains would attract better loadings on the same route, but if we look at it closely the area along Wellington Rd is mostly industrial and has little housing nearby.

And most of the suburb of Rowville is beyond walking distance of the terminus, with the incredibly indirect and infrequent route 681/682 providing the only link to most of these areas. No wonder it doesn’t get much usage!

My general view is that SmartBus is a welcome upgrade to Melbourne’s generally poor bus system, but doesn’t go far enough. And the concept of extremely long orbital routes is doomed to fail.

I welcome readers comments.

In South East Queensland, a large bulk of Brisbane’s metropolitan area comes under the one city council, stretching to about 20km from the CBD. Due to it’s size, BCC provides some services that would be managed by state government elsewhere such as Brisbane Transport which operates bus services within it’s boundaries.

The other major public transport provider is QR Citytrain, the operator of rail passenger services over a vast area stretching from Robina on the Gold Coast, to Gympie North beyond the Sunshine Coast hinterland.

Within the BCC area there are very few instances where bus services feed into railway stations, and in some cases it seems as if the bus system had been designed by someone ignorant to the railway’s existence. The focus of the system is undoubtedly the Brisbane CBD, with the majority of bus services continuing there, often duplicating parallel train services.

But it is not as if there are frequent rail services. With the exception of the inner spine between South Bank and Northgate, and the inner half of the Ipswich line to Corinda on weekdays, most stations only receive a train every 30 minutes except during peak hours, when the frequencies can baloon to as much as 10tph at locations that receive 2tph at other times.

This can be compared to services like Brisbane Transport’s BUZ routes that provide a service at least every 15 minutes between 6am and 11pm every day

QR’s system within the BCC area is seen as a service for people who live within walking distance of (or park & ride at) stations as the comprehensive bus network mostly either continues to, or connects with other buses to the CBD. Like with the buses most rail patronage is to or from the City, and where a more frequent buses co-exists alongside the normally half hourly trains, the former wins most users.

Bus vs Rail patronage comparisons

The Citytrain network extends from the centre of Brisbane south to Beenleigh and Robina on the Gold Coast, north to Ferny Grove, Shorncliffe, Caboolture and Gympie, east to Cleveland and west to Ipswich and Rosewood.

The Citytrain network covers approximately 300 kilometres of track and includes 143 stations. Queenslanders make an estimated 49 million passenger journeys per year with Citytrain, with a daily average of 150,000 journeys.
http://www.citytrain.com.au/about/overview/overview.asp

As this example shows, the vast railway network is only carrying comparable numbers to one 16.5km long busway on Brisbane’s southside! I don’t have anything against this busway (although the corridor could potentially be used by heavy rail), but this highlights the poor patronage of the rail system.

The South East Busway opened in 2001 and was a first for South East Queensland. Patronage on the busway has increased by more than 180% on major bus routes since opening. The busway continues to carry more than 150,000 passengers per day (and up to 18,000 passengers every hour during peak times into the city).
South East Busway Extension – Newsletter 1

What can be done

Competition between modes of public transport is purely a waste of resources and benefits nobody – like in this instance there is a rail system that suffers from poor frequencies but otherwise good infrastructure, alongside a bus system taking up too much of the long-haul task.

There has been move toward change in the last few years, with the establishment of Translink and the introduction of integrated ticketing between operators in 2004. Much of the planning responsibility has been taken off individual operators, but no major changes to the bus network has been implemented.

Before any significant changes to the bus system can be implemented, the abysmal frequencies offered by QR need to be correct. Ideally the 30 minute intervals should be boosted to about three times that, which may not be possible in some locations. This is unlikely to happen until QR converts to Driver Only Operation (DOO) as is used interstate to reduce crewing costs.

Passengers also need to be willing to give up their single-seat journeys to the City and be willing to change modes. The prime example of this would be the routes from the Centenary suburbs which largely overlap the Ipswich line between Indooroopilly and the CBD.

Were a proper interchange between modes built, as well as TUAG frequency for both provided it is hard to see how someone would complain given the train’s faster travel time, although some might argue 7km is too close to the CBD to inflict a change-of-mode on passengers.

Benefits

I see the key benefit of this as being the more efficient utilisation of resources; using trains as the trunk service from suburbs to the CBD and buses as feeder services. There is little sense in designing networks independant of each other.

In the case of buses, surplus vehicles can be used to boost frequencies on the suburban portion (connecting to trains) and other routes. Understandably some passengers would be unhappy with having to change vehicles, but trying to give every back street an express to the CBD as-present is very inefficient transport planning.

Of course, not all middle-outer suburbs to City routes would have to be cut back, many could be justified on the grounds that they also service other non-rail serviced destinations on the way, although if they are no longer carrying long-haul passengers than capacity is made available for inner suburban passengers.

The key benefit to using the rail network more heavily is the high construction & maintenance, but low variable costs that rail has. Although QR are shifting large passenger loads on some lines in peak, to run so few services in the off-peak is a waste. I beleive the bus system should work in a similar matter to other Australian cities.

Conclusion

The present public transport system in the Brisbane area has been planned poorly, with too much emphasis on buses where train services exist. It is in the best interest of the public to re-design the bus network to complement & feed the rail system to provide an efficient integrated system, with a focus on more destinations than funelling all services into Central Brisbane (also see Let’s all go to the City).

QR’s suburban system is in need of an increase of the off-peak frequency on all lines to 10-15 minute intervals if it is to win non-captive ridership, and provide a useful trunk role. It may be worth considering the usefulness of the infrequent Doomben service which is a waste of inner city paths, running to an area with few trip generators.

An articulated 'Phileas' bus travelling along a busway in Eindhoven, Netherlands.

Bus services are often seen by the travelling public, policy makers and planners alike as an inferior third rate public transport service which travellers fall back on when other services aren’t available. When planned and implemented effectively however, buses can provide a fast, frequent and efficient service which is attractive to commuters. The list of criticisms of buses is long, and there is no shortage of lobby groups and advocates who lobby for the laying of rails as the panacea to all their transport woes. While there are certainly cases where replacement of bus services with rail is warranted by passenger demand, this is not always the case. Somebody and MJJA have written recent posts about matching the appropriate modes to the required task. Rail solutions are often put up as solutions to problems where the answer in fact lies not in the vehicle used, but the manner in which the service is delivered. Given the right operating environment, buses can provide an alternative to trunk metro or light rail services with shorter implementation timeframes and lower implementation costs. 

Bus services as we know them

In many Australian cities, bus services involve poor frequencies (often half hour at best), old vehicles, contending with traffic congestion, poor signage and route information, poor passenger amenity (bus stops which consist of little more than a pole at the side of the road), poor interconnectivity and frequent stopping delaying journey times. Journeys are further delayed by the need for the driver to sell tickets to passengers. Buses are therefore said to be slow, irregular, uncomfortable and unpredictable. Furthermore, it is often claimed that passengers don’t catch buses because they “don’t know where bus stops are and they don’t know where the bus is going to go”. Conversely, it is said, trams and trains, give prospective travellers an idea of where the vehicle travels by virtue of the fact that the vehicles follow the rails.

Can buses play a mass transit role?

Let’s contrast this poor standard of services with the other extreme on the bus transport spectrum – the deluxe model in bus transport, Bus Rapid Transit (BRT). Based on the pioneering BRT model from Curitiba, Brazil, BRTs have been employed in other parts of Latin America and increasingly catching on in Europe and North America. Australasian bus systems which contain significant features of BRT include Brisbane’s South East busway, Adelaide’s O-Bahn, Sydney’s Liverpool – Paramatta transitway and Auckland’s Northern Busway. Queensland Transport is considering options for a Cairns Busway and a busway has been planned for Canberra.

How is BRT different?

The BRT model provides high frequency services, with high occupancy articulated buses carring up to 300 passengers. Vehicles can be gas powered, reducing emissions. Multiple doors allow for rapid boarding/alighting in the same manner that passengers enter and exit trams or metro carriages. The bus driver is relieved of the responsibility for selling or checking tickets as these are purchased at stations and various other outlets before boarding. Rather than stopping at frequent bus stops, BRT buses stop at bus stations which resemble light rail stops.

Busways can be created without the need for tracks or overhead cabling and can often be laid in less time than it takes to construct rail networks. Furthermore, because residents along major roads are already used to high volumes of traffic, new busways tend not to be subject to NIMBYism to the same extent that new rail or light rail lines are.

Ideally, BRT involves dedicated ‘bus roads’ or busways which separate buses from cars and regular traffic. Separating bus routes has the obvious advantage of removing buses from congestion – in the case of Brisbane’s South East Busway, buses zip along at 80km/h adjacent to the South Eastern Freeway, which in peak hour resembles a car park (a good depiction of this appears briefly in the video linked below). A journey from Eagle Plains to the city takes 18 minutes, compared to 45 – 60 minutes by car in peak hour. [1]. 

Achieving separation

Adeaide's O-Banh runs on a guided system designed to achieve a smoother ride. Running at top speeds of 100 km/h, it is among the fastest BRTs in the world.

Separation can be achieved by delineating existing lanes of traffic exclusively for bus use (in some cases by using physical barriers such as low curbs or in Chinese examples ‘fences’). More elaborate schemes involve the complete physical separation of busways from roads. This can be done by locating busways in the medians along arterial roads or freeways (eg. Curitiba, Brazil), in tunnels, elevated roads and overpasses (eg. Brisbane), or building new transport corridors (eg. Adelaide). Integrating busways with existing tram/light rail lines is also an option, however this option would be more prone to bunching.

Providing a dedicated, clearly legible bus route also removes the problem of “not knowing where the bus is going to go or where to catch it”. Moreover, it allows for trunk bus routes to be indicated schematically on metropolitan transit maps alongside trains, trams and light rail. This has been done effectively on Brisbane transit maps.

Bus stations

Bus stations are located at intervals similar to those of metro stations. Bus stations can provide ticket sale facilities, high levels of commuter information, real time departure info, and passenger amenities. When designed properly, bus stations enable buses to overtake those ahead of them and avoid the bunching phenomenon sometimes experienced by trams.

Significantly, bus stations can be staffed. Contributors at other forums have derided the suggestion that a ‘bus stop’ might be staffed at all, however staffing has significant potential benefits. Staff can sell tickets, undertake fare enforcement functions, provide customer assistance and information (particularly useful upon opening of new services), and ensure safety and amenity at bus stations. I understand that Brisbane bus stations and interchanges have begun employing staff at peak times to pre-sell tickets in order to expedite boardings and reduce dwell times.

Bus stations in Bogota have glass doors which open when the bus pulls up.

Off-road operations and network integration

Finally, busways enable integration between trunk services and local feeder bus services. This can be done by allowing feeder routes to enter the busway and travel along it to a final destination (CBD or a cross city destination), providing a single seat journey as is done in Brisbane and Adelaide (39% and 80% off system boardings respectively)[2]. Alternatively, feeder routes can be designed to connect with busway trunk services, allowing passengers to transfer to the busway. This is the case in Sydney’s Liverpool – Parramatta Transitway where no on-road bus routes directly access the busway. The latter option, with turn-up-and-go services can avoid the need for real time travel information, but may require high level of coordination with on-road feeder services if the trunk service is less frequent.

With these higher standards of service, one would expect sizable passenger volumes. In peak hour Brisbane’s South East Busway moves 15,000 people per hour. It has achieved a 56% increase in corridor ridership growth resulting carrying 26 million passengers a year. Adelaide’s O-Bahn carries 7 million pax/year, 4,500/hour in the peak, achieving a ridership growth of 25%, of whom 40% previously drove cars. [3].

A shopping list of options

BRT is the top of the range in bus services – the top shelf product. Like all transport modes, it isn’t without costs. An obvious one is higher staffing costs arising from higher numbers of vehicles involved. Another is the initial infrastructure cost compared with on-road buses. Installation costs vary depending on busway alignment, station intervals, and infrastructure such as tunnels and bridges. Brisbane’s South East Busway, which involved eight tunnels over 1.6km, a further 1.6km of bridges and viaducts, and 10 bus stations over a distance of 16 km (other sources report 20km) came in at a pricey >$24m/km [4]. There is no question that this was an expensive busway however much of this cost must be attributed to high quality station design, tunnelling and bridges. Adelaide’s O-Bahn cost only around $8m/km (in 1998 prices)[5]. Both of these were new busways built along entirely new alignments (although Brisbane’s follows an existing freeway). BRT costs would be reduced significantly when constructed on existing roadways, or within existing road reserves as has been done in South American and Asian examples. The quality of station design, amenities, and provision of real time travel information adds to the costs, as does the provision of elevators where bus stations are located above or below the ground.

Implementing low cost improvements utilising BRT principles

Even without full blown BRTs running inside the medians of major freeways and arterial roads, there are elements of BRT which can be incorporated into existing bus systems to provide better levels of service. Where opportunities exist, the busway separation concept can be partially applied at no cost, (eg. the use of freeway shoulders by buses in peak hour, as is done on Melbourne’s Eastern Freeway). Other initiatives may involve some cost. It is possible to cherry pick any one or more of these to provide improved services:

Buses share the tracks with trams in Mannheim, Germany.

- bus priority on the road network (eg. bus only lanes and priority signalling at intersections),
- improved traveller information,
- real time departure displays,
- improved cross modal integration,
- improved frequency,
- expedited boarding, and
- route legibility & identification.

In the interests of length I won’t describe each in great detail however I will briefly discuss the last two.

Expediting boarding and reducing dwell times is a no-brainer . This involves enabling passengers to enter/exit through all doors and relieving the driver of the ticket sale and enforcement function. Many European bus systems (particularly those in Eastern Europe) have significantly wider doors than Australian buses and operate much like Melbourne trams do – that is, passengers get on at any door and validate their ticket on board. Ticketing is not the driver’s concern. Some networks provide conductors, but most simply rely on random ticket checks by ticket inspectors.

Secondly – route identification. Without the need to construct busways, it should be possible to convey to (sub)urban dwellers the existence and connectedness of a bus network in the same way that tram tracks do through their presence.  What I have in mind is high visibility bus stops with matching on-road signage that indicates to potential travellers the presence and direction of a bus route. I’m not aware of any existing system, where cars and buses share lanes of traffic, which uses road markings indicate bus routes in a similar fashion to those used for dedicated bus lanes. I’m sure that some bright minds could come up with effective designs (like a continuous wavy line, or regular yellow polka dots) which could be used to indicate the presence of a bus route.

Another tool in the transport planner’s toolkit.

Buses are by no means a panacea. There will still remain a role for heavy and light rial. Buses need not however remain the poor cousins of trains, trams and metros with expectations of lower service standards. They can be an effective means of rapid mass transit and have proven to be effective in this role both in Australia and in other parts of the world. They should be considered as a viable option where the mode suits the purpose.

——-
References
[1] Currie, Graham. 2006. Journal of Public Transportation, 2006 BRT Special Edition. “Bus Rapid Transit in Australiasia: Performance, Lessons Learned and Futures”. p12. (references – Cr Quirk, and Currie, p12. ) .
Cr Quirk, quoted on “”Making Things Happen with Rapid Bus Transit Part II”. Accessed at http://au.youtube.com/watch?v=3LEtf32Bu3Y&NR=1 on 21 Nov 2008.
[2] Currie, Graham. 2006. Journal of Public Transportation, 2006 BRT Special Edition. “Bus Rapid Transit in Australiasia: Performance, Lessons Learned and Futures”. p7.
[3] ibid.
[4] ibid. p5.
[5] ibid.

Useful Resources
1. “Go BRT” A ‘Fact Sheet’ with several good photos (PDF).
2. “Making Things Happen with Rapid Bus Transit“. YouTube video in two parts. (A look at BRT through rose coloured glasses, but several interesting images).
3. National Urban Transit institute “At-Grade Busway Planning Guide” – full of technical design info for all you technical folk.