Digital Twin for Roads
Lead Research Organisation:
University of Nottingham
Department Name: Faculty of Engineering
Abstract
This project aims to develop digital twin for roads at suitable scale with application of artificial intelligence to deliver predictive analysis tools for highways maintenance
Existing road pavement condition data contained within existing Pavement Management System is time static, with surveys of varying frequency undertaken for the purpose of ascertaining asset condition to determine if maintenance is required, as opposed to tracking pavement deterioration over time and implementing deterioration prevention techniques.
The proposed research focusses on utilising Artificial Intelligence (AI) and Machine Learning to undertake cross linking of various existing historic asset databases, and external information such as weather conditions and traffic, to develop a Big Data model that accurately predicts pavement deterioration for defined road sections based on existing asset characteristics. The model will be applied to the extensive datasets available at Highways England and AECOM. This solution will, without necessarily requiring additional data collection, will allow to predict when pavement defects and deterioration is likely to occur and act to prevent defect occurrence by means of cost-effective treatment (application of preservatives, thin surfacings, inlay). This will result in increased efficiency, reduced maintenance spend, traffic delay and congestio
Existing road pavement condition data contained within existing Pavement Management System is time static, with surveys of varying frequency undertaken for the purpose of ascertaining asset condition to determine if maintenance is required, as opposed to tracking pavement deterioration over time and implementing deterioration prevention techniques.
The proposed research focusses on utilising Artificial Intelligence (AI) and Machine Learning to undertake cross linking of various existing historic asset databases, and external information such as weather conditions and traffic, to develop a Big Data model that accurately predicts pavement deterioration for defined road sections based on existing asset characteristics. The model will be applied to the extensive datasets available at Highways England and AECOM. This solution will, without necessarily requiring additional data collection, will allow to predict when pavement defects and deterioration is likely to occur and act to prevent defect occurrence by means of cost-effective treatment (application of preservatives, thin surfacings, inlay). This will result in increased efficiency, reduced maintenance spend, traffic delay and congestio
Organisations
People |
ORCID iD |
Kun Chen (Student) |
Studentship Projects
Project Reference | Relationship | Related To | Start | End | Student Name |
---|---|---|---|---|---|
EP/T517902/1 | 01/10/2020 | 30/09/2025 | |||
2767121 | Studentship | EP/T517902/1 | 01/10/2020 | 31/03/2024 | Kun Chen |