Friday, September 06, 2019

Building Melbourne's Useful Network: Part 19 - Mornington Peninsula buses

The Mornington Peninsula Shire Council recently launched a campaign calling for better buses on the Mornington Peninsula. They've got a point. Bus service per capita there is way lower than almost anywhere else, and many residential areas are without coverage. 

It's not as if the buses there are poorly used. The area's main and busiest route is the 788 (covered in detail here). This runs every 40 minutes on weekdays and 60 - 80 minutes on weekends. This level of service is more consistent with a neighbourhood back-blocks bus than a trunk route that is the peninsula's only connection to the outside world. It gets so crowded that kids can't reliably take it to school as they might not be able to fit on. Similar crowding also happens on weekends. The 788 serves the marginal seat of Nepean, currently held by Chris Brayne MP - click to email. This is why an upgrade of this route features on my list of marginal seat transport upgrades

Then there's the route 787 around Rosebud. It tries to do too much so is very very complex. It runs via more streets than any other bus route in the state and has a different service on Saturdays compared to Monday to Friday. And there are only a few trips timetabled, limiting its usefulness. 

Especially in the years before affordable air travel the Mornington Peninsula was a holiday destination. The upper-middle class often had holiday homes there. Rich retirees and those freed from the constraints of commuting made places like Sorrento their home. And those who have acreages in the hinterland for their vines or horses, aren't usually short of a buck either. 

However that's not all who live in these areas. Outer coastal areas like NSW's Central Coast, Brisbane's Deception Bay, WA's Rockingham or Adelaide's far south have some distinctive class and ethnic demographics. These include high numbers of age pensioners (often postwar UK migrants) and a younger set that is more likely to be unemployed and less likely to be tertiary educated than the metropolitan average. As gentrification and immigration changed the face of inner and middle suburbs, fringe coastal and regional areas are the last areas where those who might identify as 'working class white Aussies' are numerically dominant.

Race doesn't matter but other socio-economic factors probably do. What this means for transport planning is that areas of the Mornington Peninsula, such as Capel Sound, Hastings and parts of Mornington have high social needs and are more likely to use buses.

While pretty much every adult in these parts aims to own a working car, and ownership is indeed high, vehicles may be in for repairs, on blocks in front of the house or awaiting the next time the owners can afford to refuel. Owners may have been disqualified from driving. They may have an injury or disability that lessens their mobility. Or, more prosaically, there are the one and (less commonly) no car families, people too young to drive and those whose circumstances would make owning a car hazardous for their personal finances.

These factors (while sometimes negative for the people concerned) can encourage bus usage. Such higher propensity should be rewarded with better service locally and to key hubs such as Frankston (including its health and education facilities). On the other hand those who are working tend to be retail and tradespeople who drive to local jobs. While their extreme commutes sometimes make media stories about poor transport, few who live in Rosebud would commute to the CBD each day. And speed isn't everything; frequency and fares are also important as demonstrated by the failure of a premium fare express coach from Rosebud West tried in 2008

People often mention that Peninsula Link and East Link have brought population growth to the peninsula, with a new demographic moving there. People with jobs in Carrum Downs or Dandenong South who wouldn't dream to live there now can thanks to better road access. This is an example of the 'city shaping' impact of road projects. They change the travel not only for existing drivers, but encourage more people to make life decisions that involve them driving longer and further. Hence induced traffic and the building of 'bypasses to bypasses' when promised congestion relief proves only short-lived. The proposed Mordialloc Freeway will likely have a similar effect as it encourages those who work in the Moorabbin or Clayton areas to consider peninsula living. As most workers have free parking and public transport won't be time competitive these commuters will be difficult to attract to public transport. Whatever transport mode you improve will attract trips to it until it reaches capacity. So we need to think about whether encouraging more people to drive further is the most sustainable approach. 

Existing Useful Network

Getting back to public transport, I explain the Useful Network concept here. It's those routes that are frequent enough and run over long enough hours to be useful for many trips. I've specified a 20 minute frequency on weekdays and 7 day service until 9pm. In other words the coloured lines on the Melbourne Public Transport Frequent Network map with the 20 minute frequency selected.

No individual route in Mornington Peninsula Shire has a Useful Network level of service. Buses are typically hourly with only the 788 more frequent (every 40 minutes on weekdays). However a Useful Network level of service operates on one corridor from Frankston through Mt Eliza to Mornington due to the coming together of three related routes (781, 784, 785) that each operate hourly.

An expansion of the Useful Network to cover more densely populated parts of the peninsula will likely require more bus resources, such as being requested by Council. 

What the council wants

The council's wish list is below: 

You can see them in pictorial form below: 

To summarise, the council wants an upgrade of Route 788 (to an unspecified frequency) and some express peak trips on it. Some routes around Mornington are made longer but less direct, although they would get some  unspecified frequency increases.

Apart from minor extensions there is little substantial network reform even though $10m is being asked for. Network reform is necessary to provide the best value for money, especially if routes overlap or there are large areas without coverage. At a minimum this could involve the confusing 787, the highly duplicative 887 and possibly also the 788. Routes around central Mornington are also quite confusing with significant backtracking. The maps above don't appear to include substantial simplification.

The shire does however place faith in 'on-demand' bus services (with buzzwords like 'smart technology' or 'microtransit'). These have generally proved to be of limited usefulness and less efficient than fixed route services. However because of bus-hostile local street layouts and a lack of through roads there may be a case for them in parts of Rosebud and Mt Eliza to provide basic coverage. More on that later.

Speaking of 'microtransit', any network review should consider the continuing role (if any) of paratransit-style routes in the area. Examples include the former Penbus (now the Route 887), occasional youth-based services or the dial-up community routes that the council itself runs.

Then there is the role of school buses, especially from the more sparsely populated areas. Could some midday trips be run to provide basic access from parts of the shire (eg Red Hill) that don't justify a full urban bus service? I think this is something that the council's list could have explored.

Mornington Shire's array of niche bus services appears quite large. It is likely a consequence of it not being considered fully metropolitan and (like Knox) never receiving a full regular bus service even in its populated areas. Instead public transport in the Mornington Shire resembles that in many smaller US cities with sparse 'catch-all' public transport networks but significant 'para transit' service for niche client groups provided as a welfare service. Moving to a more versatile and inclusive network would result in a more useful service that serves more trips and (likely) involves less subsidy per passenger carried. 

An expanded Useful Network (the big picture) 

What if we were to be more radical than council and rejig the whole network rather than upgrading  or slightly extending existing routes?

We'll start with a big picture - ie the more frequent Useful Network corridors that we normally talk about on Fridays. These are routes with 7 day service including every 20 minutes or better from morning to night on weekdays.

Three corridors are suggested below. Frankston to Mornington already exists (781/784/785). The Frankston - Rosebud/Rye and Frankston - Baxter corridors require service upgrades. Existing weekday frequencies on these are 40 and 60 minutes respectively (Route 788 and 782/3).

You'll notice that the ends of the Useful Network corridors split. This is because all Useful Network corridors comprise multiple routes. Population density and patronage tends to drop near their ends. However everyone wants a single seat ride to key centres such as Rosebud, Mornington and Frankston.

A reasonable service response then is to operate all three trunks as carefully timed two or three route corridors with them diverging at Mornington, Rosebud/Rye and Baxter (or Somerville) respectively. I have not included the Stony Point train in the latter as it's insufficiently frequent to be useful for a lot of travel, particularly that which is local in nature.

The map is deliberately vague on the alignment of these routes, particularly those to Rye.  Potential options for Route 788 include the current alignment (best for Mt Eliza), a Moorooduc Hwy alignment (good for the university and hospitals) or a (fast) Peninsula Freeway alignment. Alternatives within these broad options are also possible. However all affect what you can do with other routes, particularly around Mornington.

Isolated or hard to serve pockets / flexible or demand responsive routes

These areas are the direct opposite of the major corridors (or even their direct branches) mentioned above. However services to less populated areas need to be considered as part of the design of an integrated network. The normal approach is to operate a low frequency fixed route via the alignment that is within a reasonable walking distance of most homes in a neighbourhood. Examples of currently poorly served areas that suit fixed route services (even if isolated) are below.

Unfortunately poor road layout sometimes makes fixed routes difficult because reasonable coverage is not possible without significant backtracking. The maps below show examples of this in Rosebud and Mt Eliza.

In other cases pockets might be isolated or have only sparse demand. Here one might divert an existing route if people book first. However booking can be a barrier to usage compared to regular fixed routes that don't need booking or advance notice to use. 

These are the sorts of areas where you might run a flexible route service to provide a scattered population with safety net coverage. This is the concept behind Telebus in outer eastern suburbs like Rowville and Lilydale and the 490 in Gowanbrae. These isolated or hemmed in areas have some similarities to those on the map above.

Flexible route services tend to have low boardings/bus hour and therefore high costs per passenger carried. You might accept this to achieve network coverage aims. However many flexible route services have not lasted. They have either been discontinued where poorly used or, if popular, converted to a fixed route. The latter is sensible because the more passengers a flexible route gets the slower the trip due to all the diversions  required (unlike a fixed route which just needs to stop rather than also divert). Whereas fixed routes are more easily scaleable if patronage grows.

There's more on flexible route services here. Leaving aside efficiency concerns, they can work for 'homeward bound' trips as arrival time is not critical. However, because travel time is affected by the number of diversions and pick ups they have, arrival times can vary widely. This variability makes reliable coordination difficult with other services, especially if they only run every 40 to 80 minutes like all of Mornington Peninsula's main routes. Unless you are intending for the flexible routes to be purely local shopper services, a revised network would not feature them unless the trunk routes they're feeding are made more frequent. This is why Telebus routes like the 672 operate to a fixed route during peak periods and a flexible route during off-peak shopper hours.

Regular routes

These services, normally operating hourly, comprise the bulk of bus services in the shire. Key issues include:

Weak termini (eg 781, 784, 785 and many in adjoining Frankston South): Route does not finish anywhere useful so is made less useful for connectivity between services or access to major destinations. Sometimes (eg 781) the route is too short to end where it should, leaving a major coverage gap.

Frequencies and operating hours: Only a few routes (781, 784, 785) operate 7 days per week to minimum standards (ie hourly or better until 9pm). Routes not to comply include the trunk route 788.

Duplication: There is substantial overlap between routes in cases where they don't form a frequent corridor. The most notable example is the 887's overlap with parts of the 781 and 788.

Indirectness: Some routes are indirect and have poor legibility.  Most notable is the 787, which apart from it having more turns than any other bus route in Melbourne, has different arrangements on weekdays versus Saturdays. 784 and 785 are also quite complex near their ends with significant backtracking around Mornington.

A (not exhaustive) summary of some of these network issues is below.

A new network

It is inescapable that more routes, more driver hours and likely more buses will be needed to simplify and improve this area's network. This is because each route tries to do too much, especially around Mornington. Apart from the largely duplicative 887, the Mornington Peninsula does not suffer from an excessive number of bus routes as is the case in some areas (like Greater Dandenong) whose network mainly comprises multiple hourly routes with often overlapping catchments.

Fixes range from easy and obvious to very difficult and not obvious. Solutions depend on what you wish to do. For instance a network optimised for fast commuter travel to Frankston station will have differences to one that prioritises good local access to local destinations.

I'll start with the easier steps, ending in some that are harder but necessary to make buses genuinely useful to larger numbers of people. Provided their order is retained they can be phased in, with one or two steps introduced at a time.

STEP 1: Frequency upgrade on Route 788. The most urgent upgrades are weekday peak periods and on weekends. While an upgraded summer weekend timetable operates its month-long duration is too short, especially if the existing 80 minute weekend frequency is retained for the other 11 month of the year. Upgrading weekday peak service from 40 to 20 minutes and off-peak service to 30 minutes (including weekends) would not be excessive. Sunday morning service starts too late and there are very poor connections when coming off trains arriving at Frankston, especially on Sunday mornings. Operating the Route 970 Night Network service as 788 trips could further simplify the local network.

STEP 2: Upgrade of both 782 and 783 to operate hourly 7 days, with 783 operating via West Park to allow 782 to be straightened. Both would be evenly offset to provide a 30 minute combined service. Some weekend Route 782 trips could extend to Flinders to provide a 7 day service.

STEP 3: Southward extension of 781 to Safety Beach to cover coverage gap in Mt Martha and connect with 788.

STEP 4: Reform of Route 787 to form direct route from Sorrento to Rosebud via Melbourne Rd, Rye, Alma St and Rosebud Hospital operating 7 days per week. It could then extend to Frankston via Route 788 alignment. 787 and 788 would form a two route corridor operating every 20 minutes between Rosebud and Frankston. A 40 minute frequency could apply on each leg west of Rosebud.

Because of 787/788's increased frequency and the extension of Route 781, 887 would no longer be needed. Instead its driver resources could be used to provide a new (possibly flexible) 7-day route 797 between Rosebud and Dromana serving isolated areas south of the freeway between Jetty Rd and Rainier Av (now served by the eastern part of the current 787).

STEP 5: Rerouting the upgraded 787/788 between Mornington and Frankston to operate via Bungower Rd and Moorooduc Hwy to serve Monash University and the hospitals. This further obviates the need for the 887 as its original rationale (in the PenBus days) was to provide a university connection. This could be accompanied by reforms in the Mornington area involving 781, 784, 785 and likely a new route 794 to retain coverage and connectivity while increasing directness. Special care should be taken to retain a simple 20 minute frequency corridor between Frankston, Mt Eliza and Mornington. Upgrading 784 and 785 to being every 40 minutes each could allow the 20 minute corridor to be run with just those two routes. This could allow greater flexibility for 781 to go a different way to Frankston, for instance via Mornington - Tyabb Rd and Frankston South.

STEP 6: Resolve coverage gaps in the Mt Eliza area south and west of the shops. Options include extending a Frankston South route (eg 773) to Mt Eliza and/or introducing a new (possibly flexible) local route 793 to serve Kunyung Rd, The Peninsula School, Mt Eliza village and the Eliza Dr area. If 773 is extended opportunity should be taken to tidy the complex network in Frankston South involving routes 772, 774 and their many variations.

There are two things remaining I haven't discussed.

The first is a cross-peninsula service from (say) Rosebud or Mornington to Hastings. Trials have been run in the past. However I have not discussed them here. This is because so many people are currently without service that providing coverage is top priority. Along with resolving overcrowding on existing routes. In any case higher frequency on existing radial routes and (potentially) different alignments will speed travel even without a dedicated route.

Secondly I haven't given much attention to city commuters. Instead I've prioritised access to local destinations such as universities and hospitals (including those at Frankston where people can connect to a city-bound train). However, especially if Baxter electrification proceeds there may be a case for some peak trips on Route 788 to operate there, a change that would also assist cross-peninsula travel (due to potential connections at Baxter). However I'd do this only once local routes operate at a good peak frequency because of the large effect that long waits have on overall travel time for all trips, not just city commutes.


I have described (but not mapped) a potential reformed bus network on the Mornington Peninsula. What do you think? Are they better or worse than the Council's proposals? And if you live in the area, why not also let Council and local MPs your thoughts?

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 NEW!

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)

No comments: