Looking at the big picture: Can we rely on satellite data to inform road maintenance decisions?

Revolutionary approach uses satellite data and machine learning to predict road pavement conditions, offering a cost-effective and extensive alternative to traditional ground-based evaluations, despite the trade-off in accuracy.”

 Dr. Torres-Machi

Please read the abstract for a detailed overview:

Abstract:

Pavement condition is traditionally assessed using ground-based methods that require driving over the road network and result in precise and accurate evaluations that are expensive and time-consuming. Due to the high costs of ground-based inspections, agencies often limit their monitoring to major roads and the condition of some elements of the road network remains unknown. Satellites offer a cost-effective alternative to monitor pavement condition. They are capable of rapidly collecting information over wide areas, but this wide coverage comes at the expense of lower levels of accuracy.
This talk explores the capabilities of satellite data to monitor pavement condition. In particular, this talk quantifies the value of this highly uncertain information to inform maintenance decisions and showcases an application in which satellite information was integrated with machine learning algorithms into a prediction tool that to evaluates the condition of ancillary pavements from publicly available satellite data.

About Dr. Torres-Machi :

Dr. Torres-Machi is an Assistant Professor at the University of Colorado Boulder, where she leads the Innovation for Resilient Infrastructure (IRI) research group. Torres-Machi’s research seeks to enhance the condition and resilience of infrastructure by developing data-driven, risk-based, and cost-effective methodologies to optimize decision-making. In this process, she develops stochastic models to understand how infrastructure deteriorates over time and identify the main drivers affecting this deterioration, she develops simulation models to predict future condition and inform decision-making under uncertainty, and she employs multi-objective optimization algorithms to ensure an optimal allocation of available funding.