UKTC Archive

RE: Windshield/Drive-by | Speed Limit

Subject: RE: Windshield/Drive-by | Speed Limit
From: David Evans
Date: Dec 09 2019 10:43:31
<<Back to basics here. risk is severity of harm x probability of target 
presence x likelihood of failure.>>

Hi Julian

Not according to the Risk Professor who built the engine of VALID's App, and 
other risk modelling experts I consulted whilst researching the whole thing.  
I think one of the problems we've had with tree risk assessment methods so 
far is that arborists have built them.  We might like to think we know a lot 
about trees, but we don't know very much about risk modelling.  I say this as 
having been part of the problem in the past.

As for the rest.  I can kind of see what you're trying to do here but I refer 
back to the risk modelling point.  On top of that, the tortuous numberwang is 
not only implying an accuracy and precision that's not credible, it then 
becomes way too complicated to apply by the highway inspector, or the 
arborist.  It's a tangle the Tasmanian Government had gotten themselves into 
when they asked me to help them.

Then there's some glaring errors, like this one for example.

<<Quite sensibly, on roads the probability is not just the relatively short 
time a car is under a tree but also the stopping distance time, which is a 
function of safe driving speed (capped at the speed limit)>>

Two points here.

1) Don’t forget to include the recognition and reaction time.  You need that 
on top of the stopping distance and driving speed.  When you do this it turns 
out that the busiest roads have a likelihood of occupation that's greater 
than 1/1 in numbers, or constant with words.  In other words, on average, 
more than one vehicle is exposed to the risk.

2) As I'm sure you know, you can't have a probability greater than 1/1; ie on 
average, more than one vehicle being exposed to the risk.  So you can't use 
probability as your measurement tool here.

<<Clients seem to like the simplicity of that.>>

This, by the time I'd waded through to the end of your post and made some 
sort of sense to it, made me roar with laughter.


Acer Ventura

PS <<I'll venture to suggest that a worst-case period for trees of various 
sizes, ages, species etc. along a busy and/or fast roadside is a lot less 
than 5 years.>>

One of the odder things in the recent posts around this subject (which simply 
began with a question about speed limit) is how VALID's proposed 5 yearly 
Active Assessment + Passive Assessment in the tree risk-benefit management 
strategies is quoted back as, "5 years", with the Passive Assessment part 
stripped out.

The UK Tree Care mailing list
To unsubscribe send

The UKTC is supported by Bosky Trees arboricultural consultancy