Safe Speed Forums

There's a much extended new version of the Pedestrian Incursion spreadsheet. The new version has 13 "speed buckets" with a normal distribution of vehicle speeds across them.

You can vary average speed, standard deviation of speed (width of the distribution), and vary reaction time across the speed buckets.

Using the "Minnesota" risk of fatality curves the results are far from realistic. With typical settings 4.4% of pedestrians hit die, whereas in the real world it's more like 1.0%. We don't get anywhere near enough misses either. (and if we assume that some light hits in the real world result in no injury or no report the 1.0% figure drops very considerably - perhaps to 0.25%.

The burning question is: "What's wrong with the model?"

I think the answer can only be that drivers slow down in danger areas.

It's notable that most deaths occur in the 31.5mph speed bucket.

Anyway, here's the 2.3MB spreadsheet: http://www.safespeed.org.uk/pedinc2.xls

_________________
Paul Smith
Our scrap speed cameras petition got over 28,000 sigs
The Safe Speed campaign demands a return to intelligent road safety

Combining this spreadsheet with the speed distribution in the Ashton Mackay Demolishedthread produces something interesting.

The observation is that the average pre crash vehicle speed in the Ashton Mackay sample data had a speed distribution centred on 20 or 21 mph.

This may be evidence that real world speed distributions in the preconditions for pedestrian crashes are also centred on 20mph or so.

If this is right, we actually have something approaching evidence that drivers reduce speed in areas of pedestrian danger on average.

Putting that figure into the spreadsheet in this thread produces something potentially realistic...

_________________
Paul Smith
Our scrap speed cameras petition got over 28,000 sigs
The Safe Speed campaign demands a return to intelligent road safety