Smart urban pavement management system

Abstract

A Smart Sustainable City (SSC) uses information and communication technologies (ICTs) to meet the demands of its citizens and workforce both now and in the future. As urbanization accelerates, new problems emerge, such as social inequality, road congestion, urban infrastructure management concerns, and related issues. Due to the proven capabilities of new data acquisition and analysis tools, urban infrastructure management can be improved. One of the critical parameters of an SSC is its road infrastructure management. A Pavement Management System (PMS) is an integral element of an SSC to optimize natural or financial resources. PMS collects, stores, analyzes, and models road condition data and could be enhanced by adding some of the latest technologies. It is in this context that this thesis develops an integrated system for intelligent urban road management. This system is based on automated crack detection (ACD), and automatic maintenance, and rehabilitation need assessment (MRNA). The system also provides a critical supporting visualization tool for the elected decision-makers representing the stakeholders or any citizen to appreciate the present condition and future needs. This thesis adapts a conventional system for ACD based on IoT, notably light detection and ranging (LiDAR) and RGB camera, which integrates artificial intelligence (AI) a cracking prediction model, and a visualization tool to appreciate the consequences of different maintenance and rehabilitation (M&R) strategies on the future condition of the road network. This thesis includes three papers. A novel architecture on urban infrastructure management is presented in the first paper. Using this architecture, all aspects of an innovative urban infrastructure management system (SUIMS) are assessed, and models and projections of the future road conditions are analyzed. The first paper also provides a detailed assessment of the SUIMS's present and future needs based on innovative methodologies, including the digital twin (DT) framework. In the second paper, a 3D mobile LiDAR sensor accompanied by an RGB camera as urban IoT sensors, are used to capture geometrical pavement data and cracks, uploaded on a GIS platform, and analyzed with the mechanistic-empirical pavement design guide (MEPDG) to identify M&R needs. Based on the PMS results, the CityEngine model, which is based on city administration officers, municipalities, and stakeholders' priorities, is provided. The third paper establishes the potential of an additional feature, augmented reality (AR), in an intelligent PMS to illustrate M&R recommendations based on each road section's present and future conditions. The final visualization result is delivered in the paper as an AR experience, which is unique to this thesis. Using the game engine and embedded textures, this paper provides a dynamic data analysis product and develops a Decision Support System (DSS) for real-time analysis for the end users. This thesis concludes that researchers need many different modules to design and implement an efficient PMS to move toward an SSC. The concept of SSC is meaningless without a tight collaboration between all distinctive parts of each urban infrastructure management system. Additionally, this thesis attempts to provide an outlook for future research.

Date	21 Jun 2023
Original language	American English
Awarding Institution	École de technologie supérieure
Supervisor	Gabriel J. Assaf (Supervisor)