Auckland’s New Network West is being implemented on Sunday, 11th of June. This constitutes the most significant single improvement to West Auckland’s bus network in living memory, putting aside the fact that the planned Northwestern Busway will undoubtedly be the next big step change.
With this in mind, it would be good to rewind 25 years to 1992 and reflect on how far we have come since then. Back then I wrote a piece in Passenger Transport Magazine (a desktop-published print magazine in the age before social media, the internet, smart phone et al) called “Interpret the Ranui Bus Timetable and Win Absolutely Nothing.” Thispiecewas not particularly kind to the bus service Ranui had at the time as outlined below.
Focussing in on Ranui, you can easily see how the public transport customer and service experience has significantly evolved since then.
One thing to note is that while there were a fairly skimpy 15 buses per day, there were eight route variants within these 15 trips (077, 085, 086, 097, 98F, 099, 135 and 137). Put simply buses started in 0- travelled via the North Western Motorway with 07s going via View Road, 08s via Te Atatu Road and 09s via Edmonton Road. But there were exceptions – the 98F flyer ran via Lincoln Road to the Motorway and the 099 ran via Te Pai Ave. The 13s ran via Great North Road all of the way to the city but even here there were exceptions. For example, the 10.45pm final bus to Ranui would throw in the Sunnyvale tiki tour at no extra charge (but a solid extra 10 minutes travel time on what was already a very long trip). Simple, eh!
And this is just a subset of the complexity of bus services serving the Te Atatu South, Henderson, Lincoln Road, Larnoch, Ranui and Swanson areas.
If I counted correctly, there were 27 different route variants in services to these areas. As if that weren’t complex enough, there was the added complexity of other services. For example, back in the day Ritchies Greenhalgh ran a couple of peak buses from Waitakere to the city centre via Swanson Road serving parts of Ranui but of course this was a different operator and hence a different timetable.
Fortunately the weekends thinned out the service complexity as it commensurately thinned out the service.
Ranui was serviced by just one route, the 135, except for the aforementioned 145 Saturday late evening Sunnyvale tiki tour. There were 11 trips on Saturday and five on Sunday, all via the entire length of Great North Road with the last Sunday bus leaving the city centre at 4.40pm.
And it gets even more complicated in Te Atatu South. To understand all of the service you had in Te Atatu South, you needed the following bus timetables:
The aforementioned Larnoch/ Ranui/ Swanson timetable.
The Sunnyvale timetable as there are 079 trips not in this timetable at peak times that ran the length of Te Atatu Road.
The Whenuapai timetable for buses that ran the length of Edmonton Road.
The Glendene timetable as late evening and weekend service to Te Atatu South (apart from Whenuapai buses) started in Henderson, ran via Lincoln Road and Te Pai Ave to Te Atatu South before getting to nearly the start of the Glendene bus route at the corner of Te Atatu Road and Tirimoana Ave from whence they would spend another 35 minutes touring Glendene, Kelston and Brains Park (looping back on themselves several times in the process) to get to New Lynn, then another 30 minutes or so to get to the city centre.
Of course, Ranui is on the Western Rail Line but back in 1992 was the absolute low point of urban rail services in Auckland.
In that entire year, rail carried just slightly over 1 million passengers – while in March 2017, rail carried over 2 million passengers in just that month. Quite the change!
The rail timetable from 1983 was in force in 1994 and would not be significantly improved until 2004 when the ex-Perth Diesel Multiple Units had been in place for a few months.
Ranui was served by eight inbound and nine outbound trains on weekdays only. Saturday train service were reintroduced in 2004 and Sunday train services only extended west of Henderson early this decade.
The last inbound train left Ranui at 3.38pm and the last outbound train from the city was catching the 5.40pm Papakura train from Auckland Station and transferring at Newmarket to the 5.50pm Waitakere train. To get there from the city, centre you could catch the ARA platform 4 bus leaving K Road at 5.18pm and hope it made it to the train on time (given that it was the last train!).
The New Network west, beginning on 11th June 2017, is quite different to the situation in 1992. There will be four routes from Ranui to Henderson, with three of these running half-hourly (141. 142, & 143) and one running hourly (146). This is considerably simpler than the 27 route variants serving this part of the west in 1992. And while there will no longer be direct bus service to the city centre, trains to Britomart now run every 10 minutes at peak, every 20 minutes during the day and every 30 minutes late evenings and weekends. While this isn’t the ultimate train frequency, it is certainly a rather substantial improvement on the eight inbound and nine outbound trains that existed in 1992.
On December 13 2016, Toronto City Council gave the green light to Mayor John Tory’s proposal to implement road tolls on the Gardiner Expressway and the Don Valley Parkway. And in September in Auckland, New Zealand the Auckland Strategic Alignment Project (ATAP), agreed a pathway between central Government and Auckland to develop the city’s transport networks over the next 30 years. In the words of the Minister of Transport, Simon Bridges in the New Zealand Government’s media release: “ATAP has concluded that to achieve a step change in the performance of Auckland’s transport system we need to move towards smarter network pricing alongside significant and ongoing investment to support growth and a much greater focus on getting more throughput on the existing network.” [My emphasis]
Given these developments, I thought it would be opportune to reprise my guest blog post, published last year in Toronto-based Brandon Donnelly’s excellent daily city-building blog http://brandondonnelly.com. Subscribe if you don’t already!
A post last year in Architect This City, The Tragedy of the Commons, raised a fundamental but all too often forgotten point about transportation: That networks where the price of use doesn’t change when demand changes, there is no effective mechanism to manage that demand. Because there is no incentive to act in the public good, we often act in what we perceive to be our own personal interest, which is equally often the antithesis of the public interest. And remember that if we are driving, we are traffic. So often people will sit fuming in their cars in the midst of congestion with thoughts like in this cartoon.
In a world where time is money, we are constantly berated about the economic costs of congestion. In 2011, the Toronto Board of Trade estimated that congestion in the Toronto region alone cost the regional economy $6 billion a year, rising to an estimated $15 billion in 2031 should no action be taken. More recent research by the CD Howe Institute, pegs this figure at up to $11 billion.
Given these sorts of eye-watering figures, one might be tempted to think that car drivers, and in particular the goods industry, would be flinging their wallets open at the chance to buy your way out of congestion. And in fact Toronto has the 407 Express Toll Route which has elements of variable road pricing. However, while the 407 ETR carries around 350,000 vehicles per day, price increases have been matters of controversy. It provides some ability for those who can afford it to bypass Toronto’s notorious traffic congestion, but its fundamental weakness is that it’s just one road in one of North America’s largest city-regions.
Similar stand-alone efforts to address congestion in Metro Vancouver, such as the Port Mann Bridge on the Trans-Canada Highway and the Golden Ears Bridge, have fallen well short of their projected traffic volumes, while nearby untolled bridges such as the Patullo Bridge are heavily congested. We have similar experience in New Zealand where our three tolls roads, with car tolls of $1.80, $2.00 and $2.30 respectively, experience diversion rates of up to 30% over the alternative but substantially longer and slower free routes.
This brings up a fundamental paradox: Congestion costs the economy a fortune; congestion is a top-of-mind frustration, yet people seem reluctant to pay even comparatively small amounts to bypass congestion.
For example, the City of Toronto’s Roundtable on Gridlock & Traffic Congestion in February 2014 came up with the usual shopping list of “transportation systems management” responses – improved management of curbside space and construction projects; synchronized traffic signal phasing; better traveler information and improved incident response. While these are all worthwhile responses, they only improve system operation at the margins. Encouraging greater use of public transit was the very last recommendation and there was not a single mention of charging or pricing as a tool to address congestion. And the feverish activity continued with a hackathon called TrafficJam on October 2 – 4, 2015 with the goal of fixing Toronto’s traffic troubles.
The very few cities that have actually had significant success at reducing traffic congestion – notably Singapore, London and Stockholm – have done this through cordon-based congestion pricing wherein if you pass the cordon, you pay the congestion charge. Entering central London on a weekday between 7am and 6pm will set you back a cool £11.50 ($C19.16; $NZ20.63). From 2003 to 2013, about £1.2 billion ($C2 billion; $NZ2.15 billion) of congestion charge revenue has been invested in public transport, road and bridge improvement and walking and cycling, of which £960 million ($C1.6 billion; $NZ1.72 billion) was bus improvements. This has included significant road space reallocation to improve conditions for pedestrians, cyclists, public transport and the urban realm.
The latest Travel in London report states that “Over the 10-year period from 2003, total trips have increased by 11.4 per cent, with particularly notable increases of 52.3 per cent in rail trips and 32.0 per cent in Underground and DLR trips, with cycle trips (as main mode) increasing by 53.9 per cent. Car driver trips decreased by 12.7 per cent over the same period.”
One interesting insight is that Stockholm trialed congestion charging and then reverted to business as usual of unpriced roads in advance of a referendum on congestion pricing. This gave Stockholmers a clear sense of the difference in traffic congestion and was crucial in supporting a yes vote in the referendum.
Stockholm has experienced a permanent reduction in traffic of about 20% across the toll cordon and congestion decreased by 30 – 50% – which demonstrates that traffic volume reductions have a disproportionately positive impact on congestion. About half of the “disappearing” drivers changed to transit, the rest to other alternatives such as different departure times and destinations and taking fewer trips.
While this sounds very promising, congestion charging has significant equity implications and requires upfront investment to provide people who either choose to or can no longer afford to drive with transportation alternatives. Both Stockholm and London invested very heavily in public transit in advance of implementing congestion charging.
And this brings up a big issue for Toronto (or Auckland or any city considering implementing congestion charging). For congestion charging to have a meaningful impact on congestion without stifling economic activity or impeding people’s ability to move around, the core capacity of a city’s transit system would need to be addressed first. In the case of Toronto, this would include having the Yonge Line capacity enhancements, Metrolinx’s Regional Express Rail and most likely the Downtown Relief Line in place first to provide both capacity and choice for people who either need or want a travel alternative to any congestion charge. This would mean that Metrolinx’s Big Move might need to get even bigger.
After a lengthy gestation period, Edmonton, Alberta is finally putting its Transit Strategy Strategic Direction Report in front of the city’s Urban Planning Committee on December 7, 2016. This report aims to give direction to Edmonton Transit System in its review of the Edmonton bus system. This report is available at this link which includes background reports on public engagement and a summary of background information.
It has been quite a long road for the Transit Strategy, with work kicking off in early 2013 with Stantec and Jarrett Walker working on a comprehensive review of the Edmonton Transit System. Now, nearly four years later, there is a Strategic Direction report which is intended to set the course for an actual review of the ETS bus system.
With 106,654,119 bus boardings in 2013 (CUTA Transit Factbook 2013) making up 76.3% of Edmonton’s 139,782,594 annual transit boardings, getting the Edmonton bus system right is clearly critical to getting Edmonton right.
While Edmonton’s boardings per capita are excellent by western Canadian standards, besting even much-publicised Vancouver, it is overshadowed by the performance of the large eastern Canadian cities. And success of course is no grounds for complacency, as complacency has been known to lead to success slipping away.
Part of the Transit Strategy process has been engagement, more engagement and yet more engagement. Edmonton does a lot of engagement and this process was one of the largest ever carried out by the City of Edmonton. There were three rounds of such engagement from November 2015 through to June 2016 for the Transit Strategy, designed initially to collect ideas, then to explore trade-offs and later to resolve unresolved issues in prioritising trade-offs. In total over 40,000 ideas were collected during this process.
One thing to understand about trade-offs are that they are inherently difficult and potentially contradictory. For example, you can ask transit customers a question about preferred walking distance and the answer is likely to be short which would imply close stop spacing to meet this need. If you ask the same transit customers if they would prefer faster or slower transit service, inevitably the response would be a desire for faster service. However, shorter walking distance and faster transit service are in counterpoint to one another. Being as clear as about what you are trading off in a trade-off is a crucial element in this conversation. Working through and prioritising the trade-offs has been a notable strength of Edmonton’s engagement process.
The first pillar “Integrate Transit with Community Planning and Design” appears to be virtually superfluous given Edmonton’s proud tradition of good high-level strategic documents and implementation strategies. Such concepts as sustainable compact growth and Transit Oriented Development are well entrenched in the city’s strategic thinking in the higher order documents referenced above along with numerous excellent plans such as the Walkability Strategy to the Edmonton Main Street Guidelines. Edmonton is very good at planning.
The second pillar “Establish a Balanced Approach to Operating Funding and Fare Policy” borders on stating the self-evident given that the level of funding subsidy funding is a public choice that needs to be debated and made explicit and transparent. But there is something missing here. Many cities in the United States very heavily subsidise public transport, regrettably too often seen as a mode of last resort, but perversely manage to achieve very poor transport outcomes for the very people that they are seeking to help because their networks often fail to make crucial choices about what they are trying to do. The level of public funding is important but much more important that that money is spent wisely on the smartest transit network that is making clear choices about what it is trying to achieve, while recognising that by its very nature public transit has a key role in providing accessibility for people who cannot or choose not to drive.
The report only really addresses that the proportion of funding coming from fares can be shifted but it neglects the point that funding can be affected by the level of ridership and does not set clear targets for increasing ridership. [Of note is that the city has a target of 105 boardings per capita by 2018]. Additionally it doesn’t mention the fact that operationally by looking at the peak to base ratio the funding ratio can be improved, given the disproportionate expense of running peak-only versus all-day service. It can cost up to three times as much to run peak service over the marginal cost of non-peak service and ridership elasticities are actually inversely related to current level of service, meaning that frequency improvements outside of the peak achieve bigger percentage gains in ridership than peak-period frequency improvements. Another key issue is affordability to customers. Fare policy has to navigate a tricky balance between not underpricing or overpricing travel for those for whom affordability is less of an issue while not pricing lower income groups out of the mobility that is so vital for access to all of life’s opportunities. One method that can help is to encourage less time-sensitive but price -sensitive customers is through making non-peak travel for all cheaper than peak travel. This can encourage non-peak ridership, take a bit of pressure off the peak where scaling capacity to demand is expensive, and can help cover a significant gaps in fare concessions which are, as in the case of Edmonton, often not available to working people on low incomes or those in receipt of Employment Insurance or social assistance.
The thinking about fare policy appears laudable and defensible on the surface but misses a key element of transit productivity which is bus operating speed. Buses can spend much of their journey time stopped at bus stops or traffic signals. Speeding up bus boarding times through big discounts for pre-payment via smart card and bus operations through bus priority measures are two key measures to make the best of any city’s transit resources. In addition, other techniques such as allowing all-door boarding (with appropriate revenue protection measures), and well-thought out stop spacing can also be of significant assistance in keeping buses moving, not stuck at bus stops. The point here is that fare policy and operational funding needs to be connected to a smart, productive transit network that is clear about what it is trying to achieve and needs to be considered as key elements of optimising bus operation.
The third pillar “Develop of Market-Responsive Approach to Transit Network Design” appears to be proposing two quite different transit systems for Edmonton: A frequent network of corridor services for the urban core within the Inner Ring Road and a more peak-focused one-seat ride network for outer neighbourhoods. This is interesting as it makes explicit a point that many cities know implicitly but rarely express overtly: That the communities in the urban core that were built around transit are much more transit-friendly than the communities that were built around the car. It suggests that a one-size-fits-all model doesn’t work for all urban contexts.
However, the waters appear to be a bit muddied by multifarious terminology about transit service typologies from premium bus, BRT, express bus, local service and crosstown service. One of the strengths of Edmonton’s LRT network is the ability to hub around major stations such as Century Park. This enables lower-frequency services to pulse to minimise bus-to-bus and bus-to-LRT transfer times or to have LRT to frequent service connections where the need for a timetable is obviated by the frequency of the service on offer.
In Edmonton’s case, one great way to do this would be to follow the Vancouver model and use premium service with bus rapid transit elements (e.g. off-board ticketing; all-door boarding; traffic signal priority and queue jumps) as a precursor service on future (or even under construction in the case of the Valley Line) light rail lines. Vancouver used this “B-Line” approach as a precursor to the Canada Line, the Millennium Line and its just-opened Evergreen Extension and is doing so for the future Broadway subway and light rail in Surrey.
Another Vancouver approach that could be of use is the use of smaller vehicles running as community shuttles in more suburban areas to connect to local centres and higher-order transit. These vehicles have the advantages of some reduction in operating costs, are better matched to demand and more able to navigate the “loops and lollipops” streets pattern often seen in suburban areas.
By their nature, one-seat-ride or express bus networks are choosing to do one thing well – which is get peak commuters to key destinations without a transfer. But generally this requires a peak bus to perform a single inbound trip in the morning peak and outbound trip in the afternoon peak as there is not enough time for a single bus to do two peak-direction trips. This is extremely hungry in bus operating cost and ties up the capital cost of a bus for very limited use. And by clear implication, the trade-off of the one-seat-ride or express bus networks is fewer resources for other key transit functions, including local connectivity. In order to maintain affordability of providing the service in a one-seat ride network, it is likely that the trade-off would be limited mobility outside of the peak in the outer areas.
A better model is a connective network model to use the same bus resource to provide more frequent local service that hubs with LRT or a higher-order frequent service at a well-designed and customer friendly connection points (and Edmonton does these much better than most Canadian cities). In particularly strong corridors, this can be a combination of all-stops and limited-stop services with the caveat that the requirement to travel does not suddenly stop inbound at 9am in the morning and outbound at 6pm in the evening. All-day networks are by their very nature much more useful than peak-only networks and help balance operating costs by encouraging non-peak trips and reducing the need for additional peak buses.
Auckland, New Zealand is strongly moving to a connective network model as illustrated below which maximises the number of destinations that be accessed in an acceptable travel time by transit by treating connections as an opportunity to add connectivity and to increase frequency for all without needing significant increases in public funding.
The strategy suggests branding for the different bus typologies but this is an area that needs to be approached with caution to avoid unintentionally “dissing” the lower order typologies. Transit functions best as a network, not as a collection of branded typologies. This is not to say there shouldn’t be a hierarchy of services or that there are mechanisms to communicate the higher order services through depiction on network maps and using lower digit numbers (e.g. Auckland uses two-digit numbers for frequent service) and more subtle vehicle branding. But, to paraphrase George Orwell’s Animal Farm, it’s good to avoid the perception that “all buses are equal but some buses are more equal than others.”
The fourth pillar “Improve the Customer Experience” is the mantra of public and private sector organisations the world over. But the best way to transform the transit customer experience is to transform the transit system itself into something that is useful for a wide variety of people to carry out all sorts of different trips at all sorts of different times. A transit system that does only one thing well – typically delivering customers to key destinations such as downtowns and universities in the peak – is not achieving this. The rest of the pillar is really a list of actions that any transit agency worth its salt should already have entrenched in its business as usual processes. This is the case in Edmonton.
The fifth pillar “Develop Transit Organizational Capacity” is really about the back-of-house operations that support transit operation. As such, these should be seamless, largely invisible to the customer but at the same time key enablers of safe, connective, reliable journeys. That a city-owned public transit operator should tightly integrate with broader city strategies and desired outcomes appears to be somewhat of a truism.
Another thing to note is that there is no mention in the Transit Strategy of how the system should be maintained after the network is redesigned. It doesn’t mention how often the network should be evaluated or overhauled to rematch the resources to the demand.
My overall impression of the Transit Strategy is somewhat underwhelming. A strategy should give a clear sense of direction and a road map with clear guidance to ETS about expected outcomes based on the clear results of the thorough and in-depth engagement process that has taken place in Edmonton. Such a strategy should have a clear sense of what success would look like: What would a transformed bus system in Edmonton look like? How would it make a greater contribution to the mobility of Edmontonians? And how would it deliver better value for money for the city’s investment in transit operations and capital?
Frankly I don’t believe that the Transit Strategy does justice to Edmonton’s proud tradition of strong strategic thinking, reflected in The Ways documents and the various plans that underpin them.
Edmonton deserves a bus system that works for all of Edmonton. Here’s hoping that Edmonton gets such a system at the end of this quite lengthy process. I will be watching with great interest the journey of Edmonton’s Transit Strategy.
I was honoured to be invited to share some perspectives from Auckland’s New Network transformation of its public transport network when I was recently in Halifax, Nova Scotia, Canada. Here are some thoughts after my visit from the good folks at the It’s More than Buses transit advocacy group from that city.
Much ink was spilt in the lead-up to the Summer Olympics in Rio de Janeiro on a range of organisational deficiencies around the games. While this is an Olympic staple – bored sports journalists drumming up stories in advance of the kick-off about whether the host city is in fact ready for the event – the coverage has been particularly virulent around the 2016 Summer Olympics.
One central tenet of this coverage has been the purported focus on public transport investment in the relatively affluent ZonaSul (South Zone) of the city to the detriment of the much poorer Zona Norte (North Zone) and Zona Oeste (West Zone). This narrative focuses on the Olympic legacy being conflated with Line 4 of the Metrô do Rio (Rio Metro) from Ipanema to Barra de Tijuca, a fast growing affluent part of the city.
But does this hype about the undoubted inequalities in Rio de Janeiro being mirrored in the transport investment for the games and the 2014 World Cup match the reality? Unlike most non-Brazilian sports journalists, I speak Portuguese so I thought I would dig a little deeper into this question.
But to kick off, here’s a quick primer on Rio de Janeiro’s public transport system. The ridership backbone of the system, as in most cities, is the humble bus, operated under the umbrella of Rio Ônibus, whose four constituent private consortia are contracted by the city to provide bus service. And the backbone of this backbone is the so-called Quentão (big hottie) non-air-conditioned bus – pictured below.
The 440 municipal bus lines carry 6,671,000 passengers daily or roughly two billion trips annually. 29.52% of all main mode trips in Rio de Janeiro are on municipal buses, with an additional 7.88% (1,781,000 passengers) on intermunicipal buses.
The Metrô do Rio (Rio Metro) carries 625,205 passengers daily or 228.2 million annual trips on a 58 kilometre network. The Metro is the main mode for 2.94% of trips.
Supervia is the commuter rail network connecting Rio de Janeiro to the outlying communities of the Baixada Fluminense. It carries 620,000 passengers daily or 152 million annual trips on a 252 kilometre network. 2.51% of main mode trips are by commuter rail.
Now back to the reason for this post: What was promised in public transport investment for the 2014 World Cup and 2016 Summer Olympics? What was actually delivered? And what difference did it make to the mobility of the average Carioca (resident of Rio)?
There were six major public transport projects intended to be delivered as part of the legacy public transport infrastructure for the 2014 World Cup and 2016 Summer Olympics with an extensive Bus Rapid Transit network, known as the High Performance Transportation Ring, made up of four corridors at its core, covering 150 route-kilometres and able to transport 1.7 million daily passengers. These projects are:
The four Bus Rapid Transit (BRT) corridors below each either run or are planned to run in dedicated right-of-ways separated from traffic with a pattern of all-stops and express services fed by a extensive network of feeder buses. Fares are pre-paid off board through the use of smart cards.
TransCarioca BRT runs between Terminal Alvorada in Barra da Tijuca and the international airport in Galeão. Its route covers 39 kilometres with 45 stations. It opened just before the World Cup in 2014 and carries 230,000 passenger per day. End-to-end public transport travel times dropped by 60%. It cost R$ 1.833 billion ($US574 million) to construct.
TransBrasil BRT began construction in 2015 and is planned to connect Deodoro via Avenida Brasil with the city centre and onwards to Santos Dumont Airport with 28 stations. Its estimated budget is R$1.3 billion ($US407 million).
TransOlímpica BRT links Barra da Tijuca and Recreio dos Bandeirantes to Magalhães Bastos and Deodoro covering 26km with 18 stations. It opened on the 9th of July 2016 for Olympic-related travel only and is scheduled to open to the public after the Olympics on the 22nd of August 2016. It cost R$2.2 billion ($US689 million) to construct.
TransOeste BRT links estação Jardim Oceânico in Barra da Tijuca (where it connects to Line 4 of the Metro) to Santa Cruz and Campo Grande. It opened in 2012 and carries 200,000 passengers per day. End-to-end public transport travel times were reduced by 50 per cent – and a whopping 65 per cent for the express service pattern – with an average trip reduction of 40 minutes per passenger. At peak, it is carrying 17,000 passengers per hour, above its design maximum of 15,000 passengers per hour and is struggling with capacity issues at peak times. It cost R$900 million ($US282 million) to construct.
The Light Rail Network‘s first stage opened on the 5th of June 2016. Its primary function is internal circulation in the city centre, linking the Zona Portuária (Port Zone), financial district and cultural corridor, and to connect the various transport terminals – Santos Dumont Airport, the ferry terminal at Praça XV, the Central do Brasil commuter rail station and the Rodovíaria Novo Rio intermunicipal and interstate bus terminal – to one another and to the Rio Metro. It cost R$ 1.167 billion ($US365 million) to construct.
Complexo do Alemão gondola
While not directly a World Cup or Olympics legacy project, the Complexo do Alemão gondola, modelled on Metrocable in Medellín, is noteworthy as an attempt to integrate a favela into the wider city.
The Complexo do Alemão favela – in reality a compilation of many favelas – had a population of 69,143 at the 2010 Census and a Human Development Index that was the lowest of any neighbourhood in Rio de Janeiro. In late 2010, control was wrested from drug dealers when the police and army invaded the favela and “pacified” it.
The gondola opened for service on 8th of June 2011. It is 3.5 kilometres long, has six stations and links at its Bonsucesso terminus to the Supervia commuter rail network. Favela residents are entitled to two free daily trips on the gondola. It cost R$210 million ($US 66 million) to construct.
In terms of overall mobility, the High Performance Transportation Ring and other World Cup and Olympics legacy public transport investments will, when all projects including the under construction TransBrasil BRT project are complete, increase the proportion of Rio’s residents with access to rapid transit from 18% to 63%.
World Cup and Olympics legacy public transport investments will …increase the proportion of Rio’s residents with access to rapid transit from 18% to 63%.
While much of the BRT network hubs around affluent Barra de Tijuca, its predominant service area are in the poorer Zona Norte (North Zone) and Zona Oeste (West Zone) of the city, where they have led to very significant reductions in public transport travel times over pre-existing on-street bus services operating in mixed traffic.
The minimum wage in the state of Rio de Janeiro starts at R$ 1,052.34 ($US330) per month for domestic and unskilled workers. An integrated fare within Rio de Janeiro city limits is R$3.80 ($US1.19) with one transfer and for services crossing city limits, the fare is R$6.50 ($US2.04), also with one transfer. For 21 working days per month, commuting to and from work within city limits costs R$159.60 (15.1% of minimum wage) and R$273 for trips crossing city limits or 26% of the minimum wage. Fortunately, under Brazilian labour law, it is the employer’s responsibility – with some exceptions – to meet the costs of their employees’ public transport travel. This is known as the Vale-transporte. However, public transport affordability remains a significant challenge for people not in employment. While Rio’s complex system of public transport smart cards offer discounts for journeys involving connections, they are not true integrated fares.
Public transport operations are not subsidised in Rio de Janeiro although fares are regulated by the public sector. For example, when the Flumitrens commuter rail was privatised in 1998, the rights to operate the service were auctioned. The successful bidder, Supervia, paid the Rio de Janeiro state government R$30 million and undertook to invest a further R$250 million of its own funds in service and infrastructure improvements. Similarly, bus services are unsubsidised and in fact the four consortia making up Rio Ônibus collectively made R$70 million in profit in 2013.
While there is public sector capital expenditure on public transport, this principally benefits rail and bus rapid transit projects, with some relatively limited investment in bus lanes on major arterial bus routes, for example in Copacabana, Ipanema and the city centre.
Of significance is the complete dominance of the bus mode making up 42.82 per cent of all main mode journeys with rail modes – Metro and commuter rail – making up just 5.45 per cent of such journeys. Public transport in total, including ferries, makes up 48.76 per cent of all trips, dwarfing the private motorised mode share of 19.47 per cent.
Of the six key public transport projects, two – TransCarioca and TransOeste – have been in service since at least 2014. The light rail network has only been in service since June 2016 and two projects – Metro Line 4 and TransOlímpica – will only enter public service after the Olympic games. One project – TransBrasil – is still under construction. Clearly delivering and opening all six projects to passenger service prior to the Olympics did not happen.
Early evidence from the BRT projects is that they have delivered considerable travel time benefits to bus customers. In the case of TransOeste, this equals 14 full days in travel time savings for an annual commuter. However, they have been less successful at achieving mode shift. A survey of TransOeste customers carried out showed that 84.6 per cent of customers formerly used conventional buses; 6.8 per cent used jitneys and just 2.4 per cent were former car drivers.
While the conventional postcard image of Rio de Janeiro is of high-rise apartments in Copacabana and Ipanema in the Zona Sul (South Zone), this density and that of the Centro (City centre) is confined to a relatively narrow coastal strip, This area is served by the Metrô do Rio, the highest capacity and most frequent public transport mode, including the just-opened Metrô Line 4 to Barra de Tijuca. Meanwhile, high levels of density, albeit at lower heights, is widespread across the poorer Zona Norte (North Zone) and Zona Oeste (West Zone) which are only served by buses and commuter rail, as shown on the map above.
At the risk of over-simplification, the four BRT projects serve the poorer Zona Norte (North Zone) and Zona Oeste (West Zone) while the two rail projects – light rail and Metrô Line 4 – serve the more affluent Centro (City centre) and Zona Sul (South Zone).
Of the six big Olympic legacy projects, the two rail projects serving better off parts of the city absorbed the lion’s share of the public expenditure at 61% while the four BRT projects serving poorer parts of the city make up the remaining 39%. However, what is missing from this is the lack of focus on the 8 million daily bus customers not on the BRT network. A small proportion of these will be will served by the two remaining BRT projects, TransOlímpica and TransBrasil, but the vast majority will not.
While the BRT projects have been useful, a key characteristic of Rio de Janeiro is intense tidal commuter flows from the Baixada Fluminense and Zona Norte (North Zone) and Zona Oeste (West Zone) to the City Centre. Only the under-construction TransBrasil BRT will directly connect to the city centre with the other projects providing connections to commuter rail and the Metro in order to access the city centre.
So while the public transport investments considered as legacy projects for the 2014 World Cup and 2016 Summer Olympics are useful to a greater or lesser degree, they make only a modest contribution to the challenges of a city where the private vehicle fleet is growing by 5% per year and where public transport capacity, speed, frequency and reliability are key to enabling accessibility in a city marked by particularly pronounced extremes of wealth and poverty. The Plano Diretor de Transportes has a long shopping list of projects needed to keep Rio de Janeiro moving. Here’s hoping that Rio de Janeiro builds on the useful building block of the Olympic legacy projects and focuses its investments where the need is, not where the money is.
[Note: This piece is by necessity based on reading and interpretation of publicly available information in Portuguese and English from a distance. It is a potted summary of a range of very complex transport and urban issues that defy easy solution and digestion into a single blog post. Comments welcome.]
Disclaimer: The author of the above post is an employee of Auckland Transport, however, the views, or opinions expressed in this post are personal to the author and do not necessarily represent the views of Auckland Transport, its management or employees. Auckland Transport is not responsible for, and disclaims any and all liability for the content of the article.
I was recently approached by the good folks at It’s More than Buses, a transit advocacy group in Halifax, Nova Scotia, Canada asking me to offer insights from Auckland’s public transit transformation known as the New Network that might be of benefit to Halifax as it proceeds to implement its Moving Forward Together plan to transform its transit system.
Much official, public and stakeholder attention and money is focused on the shiny baubles of rail transit while the humble bus, often the workhorse or even sole mode of transit systems, gets the crumbs left over. [Why buses and bus network planning deserves much more attention will be the subject of a future post.]
And Halifax just has the humble bus. A commuter rail system was studied a few years back but it was concluded that a system that delivered customers far from their final destinations might not be best, a mistake that took Auckland, New Zealand 73 years to fix once first identified.
Halifax also has many of the things that make transit and overall transport planning in Auckland, New Zealand such a challenge. The MacDonald and MacKay bridges over the Halifax Harbour are obvious pinch points. Others geographic barriers include Nova Scotia’s 100-series expressway network around Halifax and the complete absence of a street grid outside of the Halifax Peninsula and Dartmouth on the Eastern Shore. This means that key arterial roads are often the only viable links between communities and have to provide for all modes.
Auckland, New Zealand, has been through a journey similar to Halifax’s – aiming to transform public transit while recognizing the unlikelihood of significant boosts in funding. This means that more needs to be done to make the best of what is available. In the 1990s, Auckland had a breathtakingly huge 511 different routes and route variants in our system. The sheer number and complexity of the system meant transit riders only understood the particular trips they used, not the routes in their area, let alone the whole network. Through ongoing “routicide” we are down to 360 routes and will end up with 120 routes in our New Network without any significant changes in coverage but very significant improvements in frequency, simplicity, legibility and, most importantly, usability.
We are confident from our past experience that the changes will yield positive results for transit riders . When Auckland implemented the New Network in Green Bay and Titirangi in 2014, as a test and phased implementation, we reduced route numbers from 24 to 9 and achieved a 35% increase in ridership within a year due to the New Network’s simple structure and consistent bus frequency. The network, route structure, and frequency of service meant buses pass each stop at the same times every hour (e.g., 10 minutes past the hour, 25 minutes past, and so on) every day of the week. Put simply, customers will use a network that they can simply grasp, and not one that causes them to scratch their heads.
The way we achieved these successes is in part due to the size of the public transit fleet and how much of that fleet is used all-day and how much is just used in the peak. In industry parlance, this is known as the peak-to-base ratio and the closer you get to one, the more efficient your fleet utilization is. While that might sound dry and technical, the simple impact of a peak-to-base ratio nearer to one is more all-day service, which is useful for a range of trip purposes and not just the journey to work or school. And this means fewer buses spending large portions of their days sleeping at bus garages and more buses delivering service to customers. This has an obvious impact on operating costs – typically in Auckland it costs three times as much to run peak service as it does to run non-peak service. Effectively, non-peak service can be run at marginal cost because the cost of the bus is covered by the requirement to use it in the peak.
Auckland’s transformation to a simple all-day network has been a long one and I have unflatteringly described our legacy bus network as spaghetti thrown at a map as you can see in the figure illustrating our legacy network.
Figure 1: Auckland’s legacy North Shore bus network. Map courtesy Auckland Transport
The legacy network shown is the current North Shore network and is focused on doing one thing and it does it well. It gets commuters to and from Downtown Auckland in the peak with 10,000 bus passengers crossing the Auckland Harbour Bridge in around 200 buses. However, this network sucks at doing everything else such as providing local service within the North Shore and non-peak service to Downtown Auckland. The peak-to-base ratio is up to four to one, meaning that three-quarters of the North Shore’s bus fleet spend their day at their depots, rather than providing service to customers.
The alternative to the direct service approach is a connective network approach. The graphic below shows how connective networks can deliver up to 15 times the accessibility of direct service networks. This is because the connective network option in the example below gives access to five times as many destinations at three times the frequency (10 minute versus 30 minute frequency) for the same number of buses. On random arrival at stops, trips are nearly always faster on a connective network than on a direct service network, even where connections are involved. This is a powerful argument – for a similar amount of public money, you can get a vast improvement in mobility if people are prepared to connect. It puts the apparently intuitive logic of the one-seat ride to the test. If the only goal of your transit system is to get people to work downtown during the peak, then by all means the one-seat ride system is the way to go. But, if the aim of the transit system is to enable access to the whole of the city for a whole range of different purposes, then a connective network is the way to go.
Figure 2 : Direct Service vs Connective Service Model. Graphic courtesy Auckland Transport
When thinking about the right service model for transit in a community, there are some key questions:
Who are your current customers? It is crucial to understand your current customers and their needs.
Who are your potential customers (if the network worked better)?
What are the key travel patterns of current and potential customers? Often downtowns and universities are the key focus of current networks but travel patterns and destinations are much more diverse than this.
What is the mode share target for transit to reach? Transit is competing in a mobility marketplace for market share and should think like a business. How do we work to transit’s inherent strengths to make it work as best as possible?
What are the key pinch points for transit service and how do bus priority measures relate to these pinch points? Our experience in Auckland is that giving buses priority at pinch points where lots of traffic is stuck self-markets the benefits of public transit. For example, when Auckland implemented 4km of peak period bus lanes in 1998 on Dominion Road, one of our busiest bus routes, we achieved 20% year on year growth each year for four years for the price of paint and signs.
What is the coverage/ridership trade-off? This needs to be made explicit and requires an open conversation with the community. While ridership is a key goal of transit, transit also provides essential mobility for people who cannot or choose not to use other modes to access their homes, jobs, school, businesses, and all the places we travel to.
As a rule of thumb, service frequencies on the core routes in a network need to be high to generate all-day, every day ridership (at least every 15 minutes but preferably 10 minutes or better particularly on key corridors in the urban core) of the region – in the case of Halifax, the Halifax Peninsula and Dartmouth. Dedicating a lot of bus resources to express services to Downtown reduces the ability to focus resources on the high-ridership routes that generate the bulk of all-day ridership and allows better utilization of the bus fleet throughout the day.
In Auckland, we are converting the “spaghetti thrown at a map” North Shore network to a rapid transit model shown below where much more local service feeds the Northern Busway with one-seat rides replaced by the busway rapid transit services doing the heavy lifting for the trip to Downtown Auckland. This strengthens both the busway and local service while using roughly the same amount of buses. Local communities benefit from much more frequent service than would otherwise have been possible and Downtown benefits by more frequent rapid transit service on the busway spine.
Figure 3: 2018 frequent network vs status quo network. Graphic courtesy Auckland Transport
The figure of the New Network shows the difference between the amount of frequent service that Auckland can provide on a connective network (on the left) – with services at least every 15 minutes, 7am-7pm, 7 days a week – and the level of frequent service we would have been able to provide if we had continued the status quo of the one-seat ride network (on the right). Clearly this is a significant increase in the size and geographical reach of the frequent transit network.
One key element of this frequent transit grid is a strong network of cross-town connections to fill it out, particularly to meet key cross-regional travel flows that don’t pass through downtown.
Transit agencies use metrics to assess their performance and should be related to their service goals; however, transit metrics can either drive you the right way or the wrong way. For example, there may be a metric measuring the percentage of people within 500 metres of a bus stop. This bus stop may only have one or two trips a day whereas other bus stops may have frequent service 18 hours a day. Hence, it’s a metric of some extremely basic (& highly variable) level of transit accessibility. A better metric is to measure how much of the city or how many destinations a person has access to within a certain number of minutes travel time (e.g., 15 minutes or 30 minutes including walking and connections).
Clearly, transit planning is not, as might be the case in the popular imagination, drawing lines on maps. It is a exercise of very considerable complexity carried out in a framework of principles that should be as simple as they are intuitive.
There are crucial choices and trade-offs to be made as funding is limited and it is crucial when making these choices to be explicit about what you are trading off and what the benefits of the choices are. And it is important to realize that if you are choosing one thing – for example one-seat rides as a priority – then you are choosing to miss out on the power of connective networks to deliver the freedom of the city to all transit customers, including those on the coverage network.
Halifax has many of the key prerequisites already in place to make transit work even better. Here’s hoping Halifax grasps the opportunity presented by Moving Forward Together to make a transformational shift in its public transit network, delivering much better service to customers and much better value for the money invested in supporting transit service.
Disclaimer: The author of the above post is an employee of Auckland Transport, however, the views, or opinions expressed in this post are personal to the author and do not necessarily represent the views of Auckland Transport, its management or employees. Auckland Transport is not responsible for, and disclaims any and all liability for the content of the article.
It was an incredibly popular post at the time, so I’m thrilled that Darren volunteered to do another one on road tolls. This is a topic that I’m very interested in and have written about a few times. Road pricing, as you’ll see below, puts us in a bit of a chicken-and-egg situation. But sooner or later I think we will need to get our head around it, as will many other cities.