Timetable Tuesday #98: How Melbourne doesn't do summer seasonal transport

When Sydney knows that a busy shopping and beach season is coming up they will put on extra buses, trams and ferries. Lots of them. Like 1200 extra trips per week this summer. The state government recognises that when places attract thousands public transport is key to making them work.  And visitors appreciate the choice because they know that driving to busy places is a nightmare. 

What does Melbourne do? Yawn. We boost our infrequent bus 788 on the Mornington Peninsula from every 80 to every 40 minutes on weekends for a month. One month. Not even the whole school holiday period. Despite the peninsula being a schoolies destination hosting thousands more people over summer. 788's increase is welcome but it's barely 40 extra trips per week. We also add an Oakleigh - Chadstone shuttle bus. And sometimes there's one serving Highpoint. But that's about it. 

Self-appointed 'technology visionaries' are known to bloviate on so called 'demand responsive' or flexible route buses as being the way of the future. Despite forty years of mostly failed trials, transport department executives, who should know better, also make sympathetic noises. But when real transport demand spikes to the point of existing bus services under their stewardship being overloaded, they are nowhere to be seen. Funny that!

The problem is certainly not a shortage of them. The Victorian Department of Transport has 142 executives according to its annual report. With a median pay of $210k that's an executive wages bill north of $30m pa. 

How many of that 142 do you think have (and use) discretion to add buses when demand calls? 

Zero.

Ramping up bus service to respond to demand in the suburbs is just not the done thing here. 

Instead we accept the gridlock that, like clockwork, happens each year. Not that those who should see it often do, with the average exec rarely beyond earshot of a tram ding unless they've driven to Daylesford. Few successfully internally advocate for improved service levels like how Sir Robert Risson did for trams' retention. The recent state budget amply demonstrated that. 

While buses are potentially flexible, especially when the date of their need is known months in advance, the institutional processes that govern their timetables are not. On this the Department is basically paralysed, leaving bus operators to cope the best they can on gridlocked roads. 

Black Friday's bumblings

Take the so-called 'Black Friday' sales last week. These generate massive traffic disruption around major shopping centres. Chadstone, for example, had an estimated 100 000 visitors that day. The problem though was not too many people but too many cars. An issue that well organised buses are uniquely equipped to solve. 

Instead the opposite happened. Transdev, the bus operator, routed the 903 orbital SmartBus away from the very shops people would want to visit during the sales. People going to the sales had to transfer to a shuttle bus. Below is what they said on their website: 

I make no criticism of Transdev; what they did might have been for the best under circumstances of road gridlock, little priority for buses and no apparent leadership from the DoT. 

They were between a rock and a hard place. If they didn't skip stops then the 903, Melbourne's busiest bus route, would have been severely disrupted. And because it's long (four hours) and not always frequent, delays could have caused bunching with half-hour or more gaps between services. On the other hand diverting buses removed their potential as a low stress way of workers and shoppers to travel without the parking hassles. 

Was Transdev's stop skipping officially sanctioned by the Department of Transport?  I don't know. But it wasn't listed as a disruption on the PTV website on the day. Thus passengers were being kept in the dark. So much for DoT's mantra of 'simple, connected journeys'. 

This last Friday isn't the only problem date. Take Boxing Day, which also generates large sales. Buses then operate on either a Saturday timetable (buses mostly every 30 - 60 minutes, with some finishing at midday), or, if it's a Sunday, a Sunday timetable (up to hourly waits, with some routes not operating). Neither service level is suitable for the potential increased patronage or as an alternative to driving. See Daniel Bowen's write-up on Boxing Day 2017 for background. 

Comparison with other major events

Contrary to the impression that might have been gained, the concept of planning transport for major events that thousands flock to is not unknown here. Melbourne even bills itself as Australia's events capital. We are fortunate that most venues are near high capacity train and tram services that generally operate at reasonable daytime frequencies. There are also processes for major event organisers to notify the Department of Transport well in advance so that potential disruptions to public transport can be worked around and (potentially) additional services scheduled. 

Shopping centre sales like Black Friday, Boxing Day, the lead up to Christmas and those in January have many characteristics of major events. Eg they are planned well in advance, occur on the same days each year and are highly promoted. Visitor numbers can be comparable to a major sporting event. They also generate heavy traffic flows that gridlock surrounding streets and disrupt public transport. 

The latter alone justifies classifying them as major events (like sporting matches and concerts) with transport plans drawn up. While that carries a cost (that the event organiser should contribute to), frequent regular public transport given fast passage to and through a centre can contribute to a sale's success. 

Then there's beach transport. Our beaches are mostly better set up with more trains and trams than those elsewhere. However trains typically parallel the shore so are less useful for travel from inland areas. 

Such travel needs buses like the three SmartBus orbitals and local routes. On weekends these typically operate every 30 to 60 minutes, compared to every 10 to 20 minutes for trains. Apart from that the main service gap is on the Mornington Peninsula where the 788 bus (mostly) every 80 minutes valiantly jostles with weekend tourist traffic for its two hour run from Frankston to Portsea. Such infrequent buses stuck in traffic provide a limited and unreliable service ill-equipped to take many cars off the road. 

Four steps to better seasonal and event transport

What should we do to make our network more robust and reliable so that it's useful on busy days when it should be the default mode of choice? After all many who don't otherwise take public transport will ride it to the football, the tennis, the races or the show. Would it not also be reasonable for it to have a similar role on other occasions where movement fails if too many people drive?  

Here are four possible steps to improvement:

1. Split the SmartBus orbitals for improved network robustness

Our long orbital bus routes permit long trips to be made without changing buses. However that comes at a cost. I’ve often mentioned how orbital SmartBuses can have catchments that vary from apartments to semi-rural. The effect of having a single frequency for the whole route is that dense areas have too little service and sparse areas receive too much. Another issue is that a traffic snarl on one side of town can have knock on reliability effects on the other. 

Shortening very long routes, starting with splitting the SmartBus orbitals into two to four segments, allows optimisation of service levels and isolating delays to only the section affected. Confining delays reduces the effect of traffic problems and speeds the timetable’s recovery from them. And better allocating service kilometres allows improved frequency where the people are. This is beneficial during busy times, even if no extra seasonal trips are added. 

The orbital most affected by traffic around shopping centres is the 903 since it serves Essendon Fields DFO, Northland, Doncaster Shoppingtown and Chadstone to name a few. Two of those centres were skipped by the 903 on Friday as noted before. A 903 orbital split proposal with some wider benefits is discussed here.

2. Upgrade service frequency where justified throughout the year 

I’ve previously looked at bus routes that, because of their high usage, need service upgrades (examples here, here, here and here).  They tend to fall into two categories; those with fairly dense migrant-rich bus-using residential catchments (eg Tarneit, Sunshine, Springvale and Craigieburn) and those serving major shopping centres and universities (notably in the Box Hill – Monash – Chadstone area). The busiest routes tend to have a bit of both. 

Patronage on major shopping centre routes, particularly on weekends, is high even when there are no special sales. Boardings per service hour can be two to four times higher than for regular bus routes. Especially on weekends busy routes such as the 733 and 900 justify frequency upgrades from every 30 – 60 minutes to every 10 – 20 minutes. Chadstone's main highway bus from Dandenong, the 800 past the premier's electorate office, runs only every two hours on Saturday afternoons and not at all on Sunday. Meanwhile Route 468, Highpoint's bus connection to Essendon and the Craigieburn line, is only every 40 minutes on Saturday and nothing on Sunday. 788 on the Mornington Peninsula also needs an upgrade though, in one of its few bus service initiatives, the 2020 state budget will deliver one by 2022. 

High frequencies put more buses on the road. This relieves crowding, lessens long waits and makes services more robust in the event of delays (almost entirely due to car traffic). Costs can be reduced if we simplify networks as discussed most Fridays in Useful Networks. But easy ways for buses to move is also required as you'll see next.  

3. Add bus priority on congested sections 

A bus system becomes operationally unworkable and unattractive for passengers to use if traffic delays buses so they can’t keep to timetables. Extra ‘fat’ could be put into schedules but this costs extra vehicles to maintain a specified frequency and means waiting at time-points when roads are quiet. And buses become incredibly slow, especially when waiting and connection times are added, with overall end-to-end travel speeds often in the 15 to 20 km/h range. 

The gold-standard form of bus priority is a dedicated way that gives buses a free run, uninterrupted by other traffic. Melbourne is big enough for these to be considered around key hubs including major shopping centres. Several ‘bus wormhole’ ideas are outlined here. If they had complete separation, buses could have provided the high capacity transport that centres like Chadstone need rather than be crowded out by cars on days when they are most needed. 

A level below that is dedicated lanes on roads. However they are only justified where bus frequency is very high. Where this is not so and bus lanes appear almost empty to drivers the political pressure to remove them becomes overwhelming (as happened on Stud Rd whose main bus route, the 901, runs only every 15 minutes, even in peak times). 

More modest measures include short queue jump lanes, signal priority at intersections and temporary traffic management during peak times (including events like sales) at pinch points. These can be highly cost-effective in areas where there are many bus movements, such as around shopping centres. However it is again best that the bus network has been made simpler and more frequent. And there needs to be planning and staffing, which gets us onto the next point. 

4. Treat seasonal events that attract large crowds such as large sales and holiday seasons as major events with important transport needs 

Such serious treatment includes events having their own transport plans. These should be drawn up months in advance with stakeholders such as bus operators, shopping centres and local government. These plans could cover matters such as information for passengers, altered access and parking arrangements for drivers, traffic management including active bus priority, mode shift and extra services where required. The emphasis should be on making public transport an enabler of improved access rather than a passive victim of gridlock.  

Conclusion

Because we’ve been weak at network reform and aligning service levels to need, Melbourne’s bus network is especially fragile when it comes under pressure from increased patronage or car traffic. Its low profile and poor reputation also means that its potential role as an enabler of access and movement is either disregarded or ignored by departments and governments who should know better.  

Points 1 to 3 above seek to make the bus network more useful and robust at all times, not just during special events. Point 4 ensures that events that generate surrounding area traffic have a transport plan that embraces rather than stymies public transport access. Following these points would put buses in a better position to contribute to the overall transport effort for everyone’s benefit.  

PS: An index to all Timetable Tuesday items is here.