Updating the latest Counts

Before starting the process, all available count data needs to have been loaded into the database. In RAMM, this can be achieved through either loading the results manually, using the MetroCount load facility or the bulk importing of count records. A number of audit checks can be done at this stage also.

Auditing the latest count data

It is important to audit the count data prior to using this data to update the estimates. Ideally this would be done as the data is received as this is when the information about the data and the site is fresh. It also breaks up the auditing process into more manageable segments.

Auditing serves two main objectives. Firstly, to check that the count data is accurate and correct. Secondly, and just as importantly, the count data maybe accurate but we may not want to use it as the basis for determining our estimates. For example we may schedule a count during a major event to see its impact such as the Agricultural Field Days held at Mystery Creek in June. The data maybe well and good, but we do not wish to base our annual average daily traffic estimation from those volumes.

Some suggested checks are as follows:

• Check for significant changes from the previous count or latest AADT estimate
  • Suggest the maximum of 20% change or 100 vpd
    • If outside this limit, consider if data is valid for inclusion in the estimation process.
  • Alternatively a more correct, but more complex, approach is to assess the difference between the count and the projected AADT
• Check count side against carriageway travel direction
  • This checks that you are not assessing a single directional lane instead of both directions
  • Be aware that in split carriageways where there are split carriageway sections in RAMM, both directions assume that both lanes (in the same direction) are being counted.
• Check count is on an inactive site or has no count site logged
  • If this is to be a regular count, then you may wish to create or activate the count site
  • If this is to be a regular count, enter in the proposed frequency into the count site table, else can enter is as an “ad hoc” site.
• Check count for invalid carriageway or location
  • Checking the raw data that the locations are valid. This may check that all counts have loaded and are in the correct carriageway location. Often during the year there will be network changes meaning a count site is no longer aligned to a valid carriageway section.
• Check that the count history is consistent
  • Check that where there are three or more counts in the last 6 years, that they are within 20% or 50 vpd of each other to give a valid estimate from their history.
    • This is an opportunity to perhaps assess whether one or more of the counts should be excluded from future estimation calculations
  • An alternative approach, again more complex, is to use a least squares approach with a regression factor to show a reasonable trend. Unrepresentative outliers can be discarded from the calculation process.
• Check that the latest count (if post the last estimate) aligns with the last estimate
  • Where the count varies by more than the maximum of 30% or 50 vpd, check the count is valid to be included in the estimation process
  • Check that the latest count aligns with the expected pavement use or ONRC/ONF classification
• Check count for valid traffic mix
  • Check the traffic mix has no anomalies such as more trucks than cars or a high percentage of heavies, values add to 100% and there are no null values
• Check count is valid for use in calculating estimate
  • As discussed above, count data maybe accurate but it does not automatically follow that we want to use it for calculating estimates. Therefore, if this the case, flag the count record in the “exclude from estimate” field.

Once these checks are complete, the asset manager is confident that the count data is ready to be used to update the estimates. But prior to updating, the asset manager should check for any sections that are missing estimates. This is the next step.