At some point I was thinking that there must be some data available out there on car accidents. With some free time on my hands I downloaded few numbers from the Central Bureau of Statistics in Israel, which has basically any road accident statistics you can think of. (in Israel)
I played around a bit with “day of the week” and “hour of the day” effects on road accidents. One can also think about slicing the data into severity level of the accident, number of casualties, and incorporate gender impact and so forth.
The graph you see is number of accidents according to the day of the week and hour of the day. I differentiated between weekdays and weekends for obvious reasons.
Well, no real surprise here. Most dangerous time on the roads in Israel (when you weight a busted door and 3 wounded the same.. ) is Thursday at around 18:00, which is nice way to kick off your weekend. Sunday at 09:00 is also a bitch. Had a nice weekend?!.. pa-pao..
I would also try to avoid traffic Friday at noon.
Again, no real surprise, just an empirical confirmation for what you always knew, when it is more crowded on the road, more accidents occur.
R code for the figures:
|
1 2 3 4 5 6 7 8 9 10 11 12 13 14 |
par(mfrow = c(2,1) , mar = c(5, 4, .6, 2) ) matplot(t0[,2:6], ty = "b", pch = 1:5, col = c(1:4,6) , xaxt = "n", xlab = "Hour", ylab = " Accidents") leg1 = c('Sun','Mon','Tue','Wed','Thu') axis(1, seq(0,24,2)) ; legend(1,3000, pch = (1:5), col = c(1:4,6), legend = leg1 , bty = 'n') matplot(t0[,7:8], ty = "b", pch = 1:2, col = c(1,4) , xaxt = "n", xlab = "Hour", ylab = " Accidents") leg2 = c('Fri','Sat') axis(1, seq(0,24,2)) ; legend(1,3000, pch = c(1,2), col = c(1,4), legend = leg2 , bty = 'n') |
