Exploiting Data Analytics and Deep Learning Systems to Support Pavement Maintenance Decisions

Abstract

Road networks are critical infrastructures within any region and it is imperative to maintain their conditions for safe and effective movement of goods and services. Road Management, therefore, plays a key role to ensure consistent efficient operation. However, significant resources are required to perform necessary maintenance activities to achieve and maintain high levels of service. Pavement maintenance can typically be very expensive and decisions are needed concerning planning and prioritizing interventions. Data are key towards enabling adequate maintenance planning but in many instances, there is limited available information especially in small or under-resourced urban road authorities. This study develops a roadmap to help these authorities by using flexible data analysis and deep learning computational systems to highlight important factors within road networks, which are used to construct models that can help predict future intervention timelines. A case study in Palermo, Italy was successfully developed to demonstrate how the techniques could be applied to perform appropriate feature selection and prediction models based on limited data sources. The workflow provides a pathway towards more effective pavement maintenance management practices using techniques that can be readily adapted based on different environments. This takes another step towards automating these practices within the pavement management system.

Empirical Models Investigation of Pavement Management for Advancing the Road's Planning Using Predictive Maintenance

“…This is a strategy that is used to maintain the downtime tracking software in an advanced manner. In the context of road planning, several types of machinery are used to upgrade and repair the road and therefore, predictive maintenance keeps an eye on such equipment to update frequently so that it can work effectively in pavement management [29]. Pavement management highly requires predictive maintenance as different types of machinery are the important part and must work in a good flow for efficient and potential planning of the roads.…”

Modelling of pavement performance evolution considering uncertainty and interpretability: a machine learning based framework

“…However, attempts to incorporate such importance evaluation results into the selection of model inputs are still limited. Meanwhile, the existing literature differed from this study in that most of them tried to find a possible compact subset of features that minimizes the model error (El-Diraby, 2020;Zeiada et al, 2019;Zeiada et al, 2020) or selected a certain number of features according to the order of feature importance (Roberts et al, 2021;Yao et al, 2019). In contrast, this study aims to capture more features relevant to the output variable with a so-called all-relevant feature selection method, which is beneficial to understanding the mechanisms of the problems instead of merely building a black box model (Kursa and Rudnicki, 2010).…”

Técnicas de aprendizaje automático para la predicción de intervenciones de mantenimiento en infraestructuras lineales de carretera

“…En el ámbito de las infraestructuras de carreteras, desde la década de los 90s, la tendencia se ha enfocado en la predicción de la evolución del estado de determinados componentes usando técnicas ML y analítica de datos; tal es el caso de la predicción de los índices asociados al pavimento, (Attoh-Okine, 1994;Bosurgi & Trifiro, 2005;Gong et al, 2018), o al estado global de éste (Eldin & Senouci, 1995;Salini et al, 2015;Sollazzo at al., 2017). En relación con la estimación de las intervenciones de mantenimiento ha habido importantes avances (Hanna et al, 1993;Taha & Hanna, 1995;Alsugair & Al-Qudrah, 1998;Domitrovic et al, 2018;Roberts et al, 2021); referenciadas por diversas publicaciones sobre el estado del arte en esta temática (TRB, 1999;Flintsch & Chen, 2004;Ismail et al, 2009;Ceylan et al, 2014;Abambres & Ferreira;Karimzadeh & Shoghli, 2020).…”