Travel Safety - Distance or Time
#blog by Bobby
The problem
It seems to be common practice to measure travel safety by fatalaties per distance, rather than fatalaties per time. This seems odd to me; In a lot of cases the decision we have to take is mostly about where to go given our limited time and not only the mode of transportation to use for a fixed destination.
The latter would lead to measuring travel safety by fatalaties per kilometer since we would have a fixed distance that will be traveled no matter what by the mode of transportation we see fit.
A possible solution
The former however seems to make sense in quite a lot of cases where we have a fixed amout of time at our dispoasal and need to decide where to go in this given timeframe, be it for leisure or even for business cases. In that situations, the travel time would be fixed and it would be much more appropriate to compare travel safety by fatalaties per hour.
The paradoxa
The source of our data, PhD Charles Higgins, seems to agree in this and also points out, that the former mode of comparison leads to some curious paradoxa. The first being a situation in which we don't really travel at all (like only from bed to sofa), but with a non-zero chance to have a fatal injury, so the fatalaties per km would reach an extremely high value, rendering going from bed to sofa to the most dangerous mode of transportation. Respectively on the other hand of the spectrum we'd have space travel - where even if 9/10 space missions would theoretically end in total loss of crew life (1/10 is found in the data below), if the remaining crew travels let's say to mars and back, this would be by very far the safest mode of transportation, measured by distance. [1]
The third option
For completeness there should be mentioned, that a third unit of measurement comes to mind; fatalaties per journey. This would sit somewhere in between those above and share some of the benefits and pitfalls of both, as extremely short journeys (either by time, distance or both) would count in the same way as extremely long ones. This seems to lead to similar problems, which I cannot find with the fatalaties per hour unit.
Summary of the Options
Deaths per distance
Appropriate if the distance or destination is fixed and the journey has to take place one way or the other. In this case evaluation based on distance seems most useful.
Deaths per journey
Although it sits inbetween both other options, it would only really make sense to use per journey measurements when the number of journeys is fixed, as in "I need to make 3 journeys today no matter the duration nor distance". This does not seem applicable to me. Even though the source[1:1] does provide data for this case, they will be ommited here.
Deaths per hour
Appropriate if the time for a journey is fixed, while the distance is allowed to vary. Examples would be "where to go on vacation this weekend", "Where to go this afternoon" or "where to travel for a customer in a reasonable time of work hours".
The data
Original Source from UK 1990-2000[2][3]
Deaths per billion kilometres
| Mode of Transportation | Deaths |
|---|---|
| Air | 0.05 |
| Bus | 0.4 |
| Rail | 0.6 |
| Van | 1.2 |
| Water | 2.6 |
| Car | 3.1 |
| Space Shuttle | 16.2 |
| Bicycle | 44.6 |
| Foot | 54.2 |
| Motorcycle | 108.9 |
Deaths per billion hours
| Mode of Transportation | Deaths |
|---|---|
| Bus | 11.1 |
| Rail | 30 |
| Air | 30.8 |
| Water | 50 |
| Van | 60 |
| Car | 130 |
| Foot | 220 |
| Bicycle | 550 |
| Motorcycle | 4840 |
| Space Shuttle | 438019 |
Conclusion
As it seems, measuring travel safety by fatalaties per hour is the most appropriate unit of measurement in a lot of cases, when we have a fixed amout of time and need to decide where to go in this given timeframe. It has some advantages over the more common fatalaties per kilometer unit and doesn't seem to lead to paradoxa like the latter. As always, you need to choose the right tool for any task - and all three units of measurements seem to have valid use cases. As mentioned by Higgings, one might suspect, that the very widespread use of the first method despite it's flaws in application could have something to do with economic interest of the aviation industry.
PS.
Climate Change
For those very attentive; you may have noticed that unmotorized travel seems to be quite dangerous by the hour. This might further indicate the need for safe, climate friendly travel options. As a positive mention we can say that Bus and Train - both quite climate friendly mobility options - do also lead in travel safety.
Vehicle Size
Also it might be added that the vehicle size does not matter in this calculations. If i assume a fixed fatalaty rate of 10% and all other variables equal (distance, time) and take 10 airplanes with 100, 100 busses with 10 as well as 1000 cars with 1 traveler each, no matter the mode of transportation I choose, I will always have a 1 in 10 chance of being under the 100 people that will die with each mode of transportation, even though those will be distributed over 1 airplane, 10 buses and 100 cars crashed.