There are several approaches to this updating for sections with a recent count since the last estimate. (Note the estimate maybe quite dated if the annual update process has not been undertaken for a long time.)

• Use the latest valid count as the estimate for that section/link for the year the count was undertaken.
  • This method suits links that do not have a regular consistent count history. It is recommended the count should it meet auditable criteria such as being within the maximum of say 30% of the previous estimate or 50 vpd or more sophisticated to vary by ONRC category. These should be flagged and assessed whether the count or the estimate is a more valid basis for providing a new estimate. For example, the previous count could be 10 years old or there may not be one, hence the latest estimate maybe based on a fair amount of supposition. The latest count would then provide the best guidance as to the AADT for that link. Alternatively, the count maybe an outlier in comparison to previous history which may suggest the latest estimate is a better basis for adjustment to the estimate for the current year.
  • It is recommended that this method is not used if the latest count is greater than four years old. Then it should be assumed that the link does not have a current valid count from which to base the estimate on. This timeframe maybe relaxed if the count record is felt to reflect the AADT volume of traffic for the link e.g. on a lower volume road less subject to change
• Use the latest valid count adjusted by the seasonal/week type profiles (if being used) as the estimate for that section/link for the year the count was undertaken
  • Similar audit criteria as discussed above should be adopted.
• Use a linear regression analysis to produce the estimate for that section/link. It is recommended at least 3 counts in the last 6 years are required to give a satisfactory outcome. The latest count should be within say greater of 20% or 50 vpd of each other. This method will suit annual and bi-annual sites with a regular count history. Even then, the output should be checked to ensure the trendline produced is sensible.
• A similar approach would use the average of the count data for say the last three years, averaged to give an average value and an “average” year” for that data. This would form an estimate dated to that average year and then the traffic group growth rate is applied to bring the estimate up to the current year.

Where the estimate is old and the latest count is not from the current year, it is recommended to update the estimate from the count year to the current year. This is done by using the traffic group growth percentage. For example if the count is three years old and traffic growth rate is 2%, the update count volume by 2% times 3 year = 6% adjustment. One can also use an exponential growth rate such as 2% ^3 = a 6.1% increase. As you can see, the difference is minor if the count date is not too old or the growth rate too high.

There will be occasions where the link will have more than one count site with enough valid data to create an estimate. However there should not be more than one active site within a carriageway section. Each count site within the link should have an estimate generated via the methods in this section. These should be assigned to each respective carriageway section. If the estimates vary by more than10%, then the link should be split. If the estimates are consistent, the relevant estimate can be applied to the adjacent sections or averaged if the section is equidistant. Alternatively, an average of the estimates can just be applied to the remaining sections in the link.

There are several methods that can be adopted depending on the count data available directly from the link. If it is felt the data is insufficient to determine an estimate, then the section or link should be treated as having no valid count data for the link and the methodologies outlined in the next two sections be adopted.

The estimate is dated to the year of the last count. If the estimates have not been updated for a while, the last count may be some years ago. The estimate can then be updated to the current year using the growth group annual prediction. It is mentioned above as a guideline of say greater than four years since the last count may be somewhat old to consider as a valid reflection from which to assess the current AADT. Again, the choice will set with the Asset Manager. In many circumstances, this will still be the best information from which to base the estimate value from.

There should always be the ability to manually update any estimates based on local knowledge. This may include checking consistency along the road corridor with other links.