Practical condition deterioration from historic data for asset management

Many asset management software packages require users to define their own deterioration models. The most basic model is a linear approach, more commonly users can input a quadratic equation to compute a more realistic ‘S-Curve’ model. This approach means users must make assumptions as to the likely future performance of their asset. This approach can be applied to groups of assets or hierarchies within one asset; however it does not take into account local effects. This paper describes the calculation of a deterioration model for a given condition parameter using as little as two years of historic data. Assessing deterioration from one year to the next in different parts of the network we show a full deterioration model can be generated. This provides users with a practical approach to creating models that take into account local factors such as traffic loading and material type automatically without construction records required by a slice in time approach. Models derived from real pavement condition data for rutting, ravelling, and roughness are presented together with comparisons against empirical methods currently in use in the UK.

Language

  • English

Media Info

  • Pagination: 10p
  • Serial:
    • Issue Number: 3.2

Subject/Index Terms

Filing Info

  • Accession Number: 01548040
  • Record Type: Publication
  • Source Agency: ARRB
  • Files: ITRD, ATRI
  • Created Date: Dec 17 2014 10:52AM