There maybe many count records for any section or link, even within the one year. These should all be recorded as they all serve a purpose. Some count data flagged not to be used for estimation purposes. For example, a count over the Christmas holiday period, Easter or a major event such as the Field Days. This data maybe accurate but would distort an AADT calculation.

So while we should record all our count data, best practice suggests that the estimate, which equates to the AADT, should have a single value for any one year. This maybe a calendar year or a financial year, depending on the preferences of the asset manager. It is not to say our estimate can not be updated. But we should not store endless iterations of the estimate for each section in any one year if it can be avoided.

One option would be to take the most recent count per link, at whatever date it happens to be. Assuming an even spread of counts through the year, then the average date would be 31 December for the July-June reporting year. Another option would be to use your favourite growth calculation mechanism and project the AADT through to 30 June. The benefit of this option would be that all links could be projected up to the same date, even if there were no counts in the current year, or if there were multiple. The key thing is to be using a projection method that is justifiable.