Months of progress and after cross checking against many dozens of references we now have measures of risk broken down by transport mode, found in the underlying risk matrix computed from the unbundling model it provides not only the well known fatality rates by transport mode use but also the overlooked fatality rates caused externally by each transport mode.
The fatality rate per billion passenger km is the measure of risk taken by a user of transport, if they were to travel the same distance the risk to themselves of dying is easily compared. But this is the inward looking risk to ones self and ignores the overall impacts on the community, computing the other direction across the risk matrix we can find the risk of causing a fatality a transport mode imposes on other road users. To ease comparisons lines of constant internal/external risk ratio are plotted along the chart, this is where the figure of bicycle riders being 100 times more likely to be killed than kill someone else comes from.
The obvious outlier is commercial aviation, in which Australia has a world leading safety record and the estimated figure is far away from the other transport modes. Contrast this to general aviation (all other aviation than regular services with the large commercial airlines) which is more risky than driving a car. The other inclusion from outside the road toll data is rail transport, similar to trucks they predominantly transport freight and the measure of passenger km overestimates their impact but the reference point is interesting for a completely segregated transport mode. Suicide events are included in the data and will over estimate the external risk of railways, noting that sadly more people commit suicide in Australia than die on the roads each year. With those additions out of the way we can take a closer examination of the road transport modes.
From the top we can start with Busses, the passenger km metric holds strong for this mode with its only function being to move passengers. While the safest of the road going transport modes for passengers it ranks among the worst in its impact on other road users, from the details pedestrians and motorcyclists are particularly negatively impacted in their excess deaths without any obvious exposure link between busses and motorcyclists. The other heavy vehicles pose similar risks to other road users through their lack of manoeuvrability and higher weight, but also through their utilitarian designs omitting details which would make them safer for other vehicles in a collision. For example side underrun protection is still not mandatory in Australia and those seen on articulated trucks are woefully inadequate.
There is excessive clearance both between the ground and the side protection, and between the bars and the wheels, while the open design can entangle an unprotected road user. The Australian design rules allows for much less ground clearance and while there is a need for vehicles capable of negotiating off road situations the vast majority of trucks can operate with much tighter clearances as seen in car carriers where they maximise the load space on the trailer.
The tight gaps of this trailer eliminate the ability of it to trap a vulnerable road user underneath and run them over. 40% of heavy vehicle vs vulnerable road user fatalities involve a collision to the side of the heavy vehicle, but the focus has been on motorised transport so mandatory front protection is sized to protect cars from underrunning but not pedestrians and no mandatory side protection is required. Despite their danger to the public the note should be made that these statistics do not include fault, professional drivers are rarely at fault in these accidents and a heavy vehicle driver having shown nothing but exemplary driving abilities kills others through the unforgiving nature of their vehicle, this can be changed but Australia lags other regions in introducing mandatory requirements for safer trucks.
Articulated trucks have a raised risk for the driver from their reduced stability from higher centres of gravity and articulation, along with use on long distance routes with fatigue problems. So while the drivers are well protected from other vehicles colliding with them as in other heavy vehicles, there remains a substantially increased risk of fatal single vehicle accidents for drivers of articulated trucks.
Continuing down the external risk we finally reach the crossover from those modes which present more danger to the public than they do to the passengers. The ubiquitous car category includes a broad range of vehicles including four wheel drives, light commercial vehicles (trucks under 4.5 tonnes) along with 3 door hatchbacks and everything in-between. The average across all of these is a lower risk to the occupants than those outside the vehicle, but it is close enough that some sub-categories of these vehicles would sit to the other side of public vs private risk (the data exists to extract this but is not public). This puts typical cars in the self balancing position that their use is more risky for the operator than to the community, if we instead plot lines of constant total fatalities (cost to society) we can quickly group the transport modes.
Now the road vehicles that require special licences are all neatly grouped along a line at approximately 10 times higher total fatalities than the average (which is near cars through their vast majority of mode share), heavy vehicles with their professional drivers, and motorcycles. Along with their well known extreme personal risk it can be seen that motorcycles also present more risk to the public than cars, while motorcycles present a similar risk to a car driver as another car, they present approximately twice the risk to a pedestrian than the broad group of “cars”. Motorcycles appear to be a very bad deal for society, and they need not only road safety campaigns targeting their personal safety (which is awful) but also addressing their unexpectedly high impact on others.
Continuing down the chart bicycles sit close to the total internal+external fatalities of cars, at less than twice the rate. But fatalities are instead dominated by external causes, from the “selfishly safe” motorised transport as previously discussed. The external risk posed by cyclists are almost entirely impacting on pedestrians lacking the energy to threaten any drivers wrapped in their armoured boxes. Finally that other non-motorised transport mode pedestrians unsurprisingly have the lowest risk on other road users although the errors in this measure grow quickly from the low number of reported incidences of pedestrians causing fatalities and the small mode share they represent. But they are like bicycle riders taking disproportionally high risks with their own safety, creating shifts to these non motorised modes requires that their overall impact be brought below that of the alternatives and to convince the self interested public would need the internal fatality rate to be lower alone.
While deaths of passengers (and implicitly, drivers) in cars are the majority of road fatalities it has been politically easy to focus on making improvements to car safety for the occupants, and engineering the road environment for their safety. As seen in this blog non motorised transport is left behind as an afterthought or outright nuisance in the relentless pursuit of personal mobility, despite that all drivers complete some of their journey as a pedestrian it remains the most dangerous form of personal transport behind motorcycles. This needs changing to an environment where people want to walk and doing so is safe, we can throw in anecdotes of people driving cars 50m down the road to the shops or just look at these statistics. Either way it paints a sad state of affairs.
Improvements to road safety have been focused too narrowly on the majority driving cars, but looking at the figures more cost effective improvements could be found by improving the outward safety of heavy vehicles. I’ll go so far to try and convince you that car safety has gone too far with it now entrenching the car as the only viable form of personal transport, passengers for trains demanding car parking or at least pickup/drop-off facilities rather than using non motorised transport or requesting viable links to public busses. A radical rethink of transport safety is required that won’t fit with the publics pre-existing ideas, but a future where all forms of transport are viable options that don’t unnecessarily negatively impact on others.