A time series analysis framework for transportation infrastructure management

Pablo L. Durango-Cohen*

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

47 Scopus citations


We present an integrated framework to address performance prediction and maintenance optimization for transportation infrastructure facilities. The framework is based on formulating the underlying resource allocation problem as discrete-time, stochastic optimal control problem with linear dynamics and a quadratic criterion. Facility deterioration is represented as a time series which provides an attractive and rigorous approach to specify and estimate performance models. The state and decision variables in the framework are continuous which allows the framework to overcome important computational and statistical limitations that do not allow existing optimization models to address various problems that arise the management of transportation infrastructure. To illustrate the advantages of the proposed approach, we conduct a numerical study where we examine the case of multiple technologies being used simultaneously to collect condition data. Specifically, we illustrate how the framework can be used to quantify the effect of the capabilities of inspection technologies, i.e., precision, accuracy and relationships, on life-cycle costs. This information can be used to compute the operational value of combining technologies, and thus, to guide in their selection based on economic criteria.

Original languageEnglish (US)
Pages (from-to)493-505
Number of pages13
JournalTransportation Research Part B: Methodological
Issue number5
StatePublished - Jun 2007


  • Inspection technologies
  • Kalman filter
  • Maintenance optimization
  • Stochastic optimal control
  • Time series
  • Transportation infrastructure

ASJC Scopus subject areas

  • Civil and Structural Engineering
  • Transportation


Dive into the research topics of 'A time series analysis framework for transportation infrastructure management'. Together they form a unique fingerprint.

Cite this