Public transport 'network effect' and buses
One of the issues in the transport unit we have covered was the 20/2020 target. The Victorian Government has committed to the goal of achieving a 20% modal share for public transport by the year 2020 - i.e. of all motorised journeys undertaken in Melbourne 20% of them will be done by public transport.
Public transport patronage rates will have to grow much faster than both population growth and car useage. In terms of bums on seats this represents roughly a tripling of current public transport users.
There has been much discussion on whether it can be done, what are the concequences of not achieving it and whether the State Government has a genuine commitment to this goal.
Personally I'm more interested in what it would take to make Melbourne Public Transport a viable means to get across the city. As a non-car user much of zone 2 land is a mystery to me. The challenge of using PT outside of the section of the city with trams seems to be the biggest limiting factor for greater adoption.
People in the 'burbs use cars because it is the only sensible choice, the alternative is too infrequent, slow and unreliable to be a viable option.
The 'Network Effect' is the idea that once public transport achieves a threshold quality of service people will be prepared to undertake multi-stage journeys (Bus to the train station, Train to the city, Tram down St Kilda Rd and walk to the office), and this reduces the requirements for a single mode to do the whole journey (with the associated costs having many low patronage routes). Further Reading: Mees, P. (2000) _A_Very_Public_Solution_, MUP, Melbourne. (esp the Squaresville section pp. 138-142)
Jan Scheurer & Rolf Bergmaier (2006) in _Keeping_People_Moving_in_Melbourne's_North_East_ (available from http://www.mtf.org.au/n/resources/northeast.html) develop a 'wishlist' of public transport improvements in the Darebin, Banyule, Boroondara & Manningham areas (Northcote, Rezza, Ivanhoe, Hurstbridge, Templestow, Doncaster way).
The general thrust of the report is that to improve public transport we need to upgrade key train stations into multi modal public transport terminals, and then co-ordinate the time tables so that the buses and trains arrive at the terminal together, making transfer between transport modes comfortable and minimising wait times during transfers.
To prioritise the improvements, they used this network effect idea abstracting existing (and proposed) public transport into a series of nodes, linked by public transport paths. To identify the most important nodes on the network they used the following connectivity index:
Ci=(ki-2)(fr+0.3ft+0.1fb)
where
Ci = connectivity index for node i
ki = number of links emanating from node i
fr = train departures per hour
ft = tram departures per hour
fb = bus departures per hour.
The constants appear to be a 'discounting factor' for trams and trains. For unspecified reasons a is bus worth one tenth of a train and one third of a tram. (There is no quality of service data which supports these weightings. It is possible these values represent the author's bias toward certain transport modes).
I think buses get a bad rap in this approach. Some criticism of buses are valid. Bus emissions per passenger kilometre can be as high as cars when driving to route starting points and low patronage rates are considered. They are just as vunerable as cars to oil price spikes**, and in the absence of dedicated road space (bus lanes) they get stuck in the same traffic jams as cars. But on the plus side the buses themselves are cheap (relative to a train or tram) and can be readily rolled out to areas of greatest need (i.e. the outer suburbs see Petrol Price Vunerbility studies of Dodson & Sipe 2005,2007) new areas because they use existing road space (assuming the subdivision pattern of the suburb isn't an impenetable maze of cul-de-sacs and roundabouts). Public transport enthusiasts are skeptical about bus investment. I suspect its because the flexibility of buses mean its just as easy to remove the service later on. Its harder to remove a train or tram service once you've sunk half a billion into buying land, laying track and signalling equipment. Or maybe its just that buses aren't sexy.
Curitiba, Brazil (Rabinovitch and Hoehn 1995; Rabinovitch 1996) is a great case study for the 'buses can be sexy' counter argument. They have a public transport system based on dedicated bus lanes, with co-ordinated time tables and high quality bus stops that provide weather protection and high speed boarding (The passenger pays to get into the bus stop, and then steps straight onto the bus when it arrives - passengers exiting the bus use the other door).
** If oil security is a major motivating factor we could explore adopting trolleybuses (looks like a bus, runs on electricity from overhead wires or from batteries where there are no wires) as means to reduce dependence of public transport on diesel, but remember in the current Victorian context most of our electricity is currently generated using brown coal - the energy equivalent of burning wet socks. (Note that under current circumstances, CO2 emissions per passenger for trams on par with driving a car).
Public transport patronage rates will have to grow much faster than both population growth and car useage. In terms of bums on seats this represents roughly a tripling of current public transport users.
There has been much discussion on whether it can be done, what are the concequences of not achieving it and whether the State Government has a genuine commitment to this goal.
Personally I'm more interested in what it would take to make Melbourne Public Transport a viable means to get across the city. As a non-car user much of zone 2 land is a mystery to me. The challenge of using PT outside of the section of the city with trams seems to be the biggest limiting factor for greater adoption.
People in the 'burbs use cars because it is the only sensible choice, the alternative is too infrequent, slow and unreliable to be a viable option.
The 'Network Effect' is the idea that once public transport achieves a threshold quality of service people will be prepared to undertake multi-stage journeys (Bus to the train station, Train to the city, Tram down St Kilda Rd and walk to the office), and this reduces the requirements for a single mode to do the whole journey (with the associated costs having many low patronage routes). Further Reading: Mees, P. (2000) _A_Very_Public_Solution_, MUP, Melbourne. (esp the Squaresville section pp. 138-142)
Jan Scheurer & Rolf Bergmaier (2006) in _Keeping_People_Moving_in_Melbourne's_North_East_ (available from http://www.mtf.org.au/n/resources/northeast.html) develop a 'wishlist' of public transport improvements in the Darebin, Banyule, Boroondara & Manningham areas (Northcote, Rezza, Ivanhoe, Hurstbridge, Templestow, Doncaster way).
The general thrust of the report is that to improve public transport we need to upgrade key train stations into multi modal public transport terminals, and then co-ordinate the time tables so that the buses and trains arrive at the terminal together, making transfer between transport modes comfortable and minimising wait times during transfers.
To prioritise the improvements, they used this network effect idea abstracting existing (and proposed) public transport into a series of nodes, linked by public transport paths. To identify the most important nodes on the network they used the following connectivity index:
Ci=(ki-2)(fr+0.3ft+0.1fb)
where
Ci = connectivity index for node i
ki = number of links emanating from node i
fr = train departures per hour
ft = tram departures per hour
fb = bus departures per hour.
The constants appear to be a 'discounting factor' for trams and trains. For unspecified reasons a is bus worth one tenth of a train and one third of a tram. (There is no quality of service data which supports these weightings. It is possible these values represent the author's bias toward certain transport modes).
I think buses get a bad rap in this approach. Some criticism of buses are valid. Bus emissions per passenger kilometre can be as high as cars when driving to route starting points and low patronage rates are considered. They are just as vunerable as cars to oil price spikes**, and in the absence of dedicated road space (bus lanes) they get stuck in the same traffic jams as cars. But on the plus side the buses themselves are cheap (relative to a train or tram) and can be readily rolled out to areas of greatest need (i.e. the outer suburbs see Petrol Price Vunerbility studies of Dodson & Sipe 2005,2007) new areas because they use existing road space (assuming the subdivision pattern of the suburb isn't an impenetable maze of cul-de-sacs and roundabouts). Public transport enthusiasts are skeptical about bus investment. I suspect its because the flexibility of buses mean its just as easy to remove the service later on. Its harder to remove a train or tram service once you've sunk half a billion into buying land, laying track and signalling equipment. Or maybe its just that buses aren't sexy.
Curitiba, Brazil (Rabinovitch and Hoehn 1995; Rabinovitch 1996) is a great case study for the 'buses can be sexy' counter argument. They have a public transport system based on dedicated bus lanes, with co-ordinated time tables and high quality bus stops that provide weather protection and high speed boarding (The passenger pays to get into the bus stop, and then steps straight onto the bus when it arrives - passengers exiting the bus use the other door).
** If oil security is a major motivating factor we could explore adopting trolleybuses (looks like a bus, runs on electricity from overhead wires or from batteries where there are no wires) as means to reduce dependence of public transport on diesel, but remember in the current Victorian context most of our electricity is currently generated using brown coal - the energy equivalent of burning wet socks. (Note that under current circumstances, CO2 emissions per passenger for trams on par with driving a car).