摘要: |
There is a need for SCANNER survey results to be consistent, in order for a reliable quantitative comparison between the condition of roads in different parts of the country to be made, along with the monitoring of the change in condition over time. Therefore, the DfT required an investigation of the consistency, and hence reliability of results produced from accredited SCANNER survey machines, their aim being to quote survey results with statistical confidence limits. The aim is for Local Authorities to obtain a BVPI score for their network and know the confidence limits that can be put on this figure, thus giving them an indication of whether an increase in their BVPI score implies that the condition of their network has deteriorated, or whether it could be just a fluctuation due to machine consistency. The most recent data available, from survey year 2006/7, has been used to calculate consistency values for the parameters used for the SCANNER RCI calculation and the RCI itself. Data taken from the SCANNER Road Routes, surveyed by every machine during the SCANNER Accreditation tests, along with data from the Auditors’ and Contractor’s repeat surveys has been used for this assessment. The method, devised by Chris Britton Consultancy within their work on SCANNER consistency, to calculate bias and random error has been utilised to calculate the consistency of the individual parameters and RCI for each 10m sub section length. In addition to this, a discussion of the effects of the consistency of the RCI on the BVPI has been included. |