It has been said that estimate should look like estimates and count should look like counts. This refers to the element of rounding. An AADT estimate of 18,352 implies as degree of accuracy that is impossible to achieve. It is also important to note that the estimate will have varying uncertainty depending on the method or amount of data available. A continuous count will provide an AADT of high accuracy, conversely an estimate based on a 7-day count taken from an adjacent section will be much less so.

Any rounding does water down the estimate in terms of its relativity to the count data from which it is based on. This has implications if we are trying to assess growth trends over several years. If we were rounding to the nearest thousand for example for estimates over 10,000 vpd, we would get steps of 1,000 vpd every few years, making it difficult to assess the growth trends for that sites.

A rule generated by the redevelopment of the RAMM TCE tool, conducted in parallel with these guidelines, adopted was as follows:

• AADT Estimate < 1,000 vpd – round to the nearest 10 vpd
• AADT Estimate between 1,000 and 10,000 vpd – round to the nearest 25 vpd
• AADT Estimate > 10,000 vpd – round to the nearest 100 vpd.

If an estimate would round to zero – set it to one vpd.

There are options to note the confidence of the estimate of note how it was derived. These assist the asset manager in future to determine how much weighting to give the previous estimate against more recent count data. In summary, it is up to the user as to how the estimate is communicated.