Using supervised machine learning algorithms in pavement degradation monitoring

Shtayat, Amir; Moridpour, Sara; Best, Berthold; Abuhassan, Mohammad

doi:10.1016/j.ijtst.2022.10.001

Cited by 10 publications

(8 citation statements)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Figures 7 and 8 show all the dec boundaries found. The accuracy obtained with this procedure is similar to, or greater those obtained in other literature research conducted by machine learning and art intelligence approaches [5,6,26,27]. The two graphs showed that both parabolic and linear decision boundaries allow a good prediction of the severity level of the data couple represented by speed and GPB.…”

Section: Resultssupporting

confidence: 82%

“…Figures 7 and 8 show all the decision boundaries found. The accuracy obtained with this procedure is similar to, or greater than, those obtained in other literature research conducted by machine learning and artificial intelligence approaches [5,6,26,27]. 8 shows a good ability to propose decision boundaries to predict the sev class of the road condition surface given the information about GPB and speed minimum value was observed for the medium severity level, which had an average equal to about 77%.…”

Section: Resultssupporting

confidence: 79%

“…In recent years, this topic has been extensively debated, with the consideration of new data analysis techniques such as Machine Learning and Artificial Intelligence [12,[26][27][28][29]. Martinelli et al processed the real data collected by sensors located on car dashboards and floorboards, employing decoding algorithms machines.…”

Section: State Of the Artmentioning

confidence: 99%

“…Shtayat et al used two multiclassification Learning Model Machines (Random Forest and Decision Tree) to classify and automatically detect different kinds of pavement distresses. Moreover, their research with a Support Vector Machine algorithm developed a binary model for the same classification and detection purposes [26].…”

Section: State Of the Artmentioning

confidence: 99%

See 3 more Smart Citations

A Vibration-Based Methodology to Monitor Road Surface: A Process to Overcome the Speed Effect

Meocci

2024

Sensors

View full text Add to dashboard Cite

Road pavement monitoring represents the starting point for the pavement maintenance process. To quickly fix a damaged road, relevant authorities need a high-efficiency methodology that allows them to obtain data describing the current conditions of a road network. In urban areas, large-scale monitoring campaigns may be more expensive and not fast enough to describe how pavement degradation has evolved over time. Furthermore, at low speeds, many technologies are inadequate for monitoring the streets. In such a context, employing black-box-equipped vehicles to perform a routine inspection could be an excellent starting point. However, the vibration-based methodologies used to detect road anomalies are strongly affected by the speed of the monitoring vehicles. This study uses a statistical method to analyze the effects of speed on road pavement conditions at different severity levels, through data recorded by taxi vehicles. Likewise, the study introduces a process to overcome the speed effect in the measurements. The process relies on a machine learning approach to define the decision boundaries to predict the severity level of the road surface condition based on two recorded parameters only: speed and pavement deterioration index. The methodology has succeeded in predicting the correct damage severity level in more than 80% of the dataset, through a user-friendly real-time method.

show abstract

Section: Resultssupporting

confidence: 82%

Section: Resultssupporting

confidence: 79%

Section: State Of the Artmentioning

confidence: 99%

Section: State Of the Artmentioning

confidence: 99%

See 2 more Smart Citations

A Vibration-Based Methodology to Monitor Road Surface: A Process to Overcome the Speed Effect

Meocci

2024

Sensors

View full text Add to dashboard Cite

show abstract

“…• Advanced Analytics and Artificial Intelligence (AI): BIM combined with advanced analytics and AI will enable more accurate prediction of road degradation and deterioration patterns. AI algorithms can analyse historical data, sensor inputs, and other relevant factors to forecast maintenance needs, optimise resource allocation, and develop more efficient maintenance strategies [11,12].…”

Section: Perspective Of Bim In Road Asset Managementmentioning

confidence: 99%

Innovative Approach to Road Infrastructure Asset Management

Best

2024

Asphalt Materials - Recent Developments and New Perspective [Working Title]

Self Cite

View full text Add to dashboard Cite

Efficient management of municipal road assets demands consistent condition monitoring. Traditional methods, encompassing measurement techniques and visual inspections, fall short in the face of urban environment dynamics, requiring substantial human and financial resources. The periodic assessments currently in use make it challenging to maintain an up-to-date perspective on road conditions and provide continuous road availability. In response to these challenges, one noteworthy outcome of the current research is the introduction of vibration monitoring. Leveraging advanced data acquisition techniques and Artificial Intelligence (AI) in the data evaluation, this approach enables continuous data collection and condition assessment with minimal human intervention. Allowing for real-time updates facilitates early detection of damage and supports the development of optimal maintenance strategies, mitigating the need for costly premature resurfacing or rehabilitation. Its impact extends beyond facilitating faster and more reliable decision making in finance, operations, and maintenance. It is crucial to outline quantifiable benefits resulting from this approach, such as improved road sustainability, increased availability, and potential cost savings. Despite its potential, the adoption of vibration monitoring for road condition assessment may encounter challenges, such as implementation costs, technological barriers, regulatory compliance, and a lack of scalability to varying scenarios. Acknowledging these hurdles is crucial for a comprehensive understanding of the proposed solution. Collaboration with stakeholders, including local communities and relevant authorities, is essential for successful implementation. In conclusion, the introduction of vibration monitoring marks a significant leap forward in the continuous assessment of road conditions. It offers tangible benefits such as improved sustainability, increased road availability, and potential cost savings. Acknowledging challenges, connecting to existing research, and addressing practical implementation considerations are essential steps toward realising the full potential of this innovative approach to municipal road asset management.

show abstract