Help Forum

Help requested posted on 3rd April 2020:

Annual safety studies

We are looking to identify current best practice for the criteria used to trigger annual road safety studies. We use the WDM software to assist us with this.

We intend to use:

Urban areas Data Based Cluster of 60m radius, > 5 collisions - interval 120

Rural areas Road Based Moving Cursor of 200m radius, > 5 collisions - interval 400

Is this similar to what you use please? Can you help ASAP.


Reply to this request

Response posted on 6th April 2020 by:
David Horsley

T: 01823423485

Annual safety studies

The current criteria Somerset CC use for locating collision cluster sites to trigger road safety studies is as follows:

Urban - minimum of 7 personal injury collisions within a radius of 50 metres during a 60 month period

Rural - minimum of 7 personal injury collisions within a radius of 100 metres during a 60 month period

Buchanan AccsMap software is used to locate these sites.

Response posted on 6th April 2020 by:
Rebecca James


Annual safety studies

Hi I've chatted with my colleagues in our Accident Studies team and they have said that they can supply the criteria we use for including Sites and Lengths in our annual Leeds reports.

The contact is

Response posted on 7th April 2020 by:
Richard Owen

T: 01295731811

Annual safety studies

We use a number of approaches:

Route analysis (classified roads) collision density and risk rate (per 100m vehicles).

Rolling clusters (matched to road number) at 50m intervals but then aggregated to a flexible 'treatment site' length which can be almost any value.

There are other analyses in urban areas but quite often you just end up flagging the same busy junctions. We use either a 5 or 6 year period.

We then often assess RTM / Trend using RAPTOR on the highest priority sites by Rank to understand if these are long-term issues or just statistical 'blips'.

Response posted on 16th April 2020 by:
Adrian Roberts


Annual safety studies

Cornwall uses base criteria of 3 or more injury accidents over 100m over three years, the list generated by Accsmap clusters and then individually moderated. Sites which overlap or extend beyond this distance are 'merged' into larger clusters.

The consequent list is ordered progressively by accident number, severity, and how recent the collisions were, giving a priority list.

We also periodically assess the classified road network by a combination of accident rate and number (trying to avoid the pitfalls of either), giving a basic route priority list.

Response posted on 7th May 2020 by:
Derek Williamson


Annual safety studies

Transport Scotland use the MCP method which involves a 100 radius cluster in both rural and urban in which 3 or more injury accidents are recorded in a 3 year period. We also use a link based analysis in which links with an injury accident rate which is 1.25 times the national average highlights the link for further investigation.

Post a response to the above help request by completing the form below:

Your name
Your email
Your telephone (optional)
My response is: