Dr Ze Zhou Wang

Research Interests

Dr. Wang’s research interest is at the intersection of infrastructure sensing, computational modelling, data science, artificial intelligence, reliability, and the broad domain of geotechnical engineering. He has several main research thrusts: (i) field monitoring and model updating of geosystems, (ii) geotechnical reliability analysis and design, and (iii) geotechnical site investigation and subsurface characterization. His research vision is to transform the assessment of geo/civil infrastructure over their entire lifespan through a synergistic paradigm that encompasses monitoring, interpretation, modelling, and assessment. This paradigm will collectively inspire the planning, analysis, design, operation, and intervention of next-generation infrastructure, addressing critical issues related to automation, adaptability, and resilience, among others. Ultimately, his paradigm will shift us from a data-driven to a knowledge-centric geo/civil infrastructure asset management, empowering both engineers and policymakers in building intelligent, adaptable, and sustainable cities of tomorrow.

Strategic Themes

Smart Materials

Research Project

Title: An innovative zonation-based machine-learning methodology for studying the interactive impacts of traffic, microclimate and natural hazards on pavement deterioration

Theme: Smart Materials

Abstract: This data-driven project aims to provide new knowledge on the interactive impacts of traffic, microclimate and natural hazards on pavement deterioration. A combined historical pavement image dataset obtained from Google Street View and Cambridge University will be processed using deep-learning algorithms to obtain the characteristics of pavement deterioration. Next, through a close collaboration with partner organizations, a wealth of datasets that include (i) traffic and heavy vehicle volume, (ii) national temperature and precipitation data, (iii) microclimate, (iv) extreme climate events, and (v) pavement material-related data are processed. To effectively handle the data, I propose to classify the data into three levels: micro-scale, meso-scale and macro-scale data. An innovative zonation strategy will be established to divide the country into multiple macroscopic and mesoscopic zones. The concept of hierarchical machine-learning will be utilized to develop a first model for all the macroscopic zones, which will then be used as the foundation model for the second machine-learning model developed for all the mesoscopic zones. In this way, it will be viable to provide refined predictions considering regional/location-specific information, e.g., microclimate. Ultimately, the new knowledge and the developed machine-learning methodology will be output as a standalone software, which will be implemented in practice for both pavement engineers and policy makers in partner organizations. The socially desirable outcomes will be: (i) the frequency of maintenance work can be significantly reduced, contributing to achieving carbon neutrality in U.K., and (ii) the new knowledge can lead to informed decision-makings on future road planning and design.

Biography

Ze Zhou Wang is currently a Marie Skłodowska-Curie fellow at the University of Cambridge. Funded by the European Union and UK Research and Innovation, his research is at the intersection of infrastructure sensing, computational modelling, data science, artificial intelligence, reliability, geotechnical/infrastructure engineering. His current fellowship project at the University of Cambridge aims to utilize both historical performance data and continuous monitoring data to advance the understanding of how climate, microclimate, traffic, and natural hazards interactively impact England's road assets. He will then provide recommendations for maintenance and repair decisions that balance cost, carbon and environmental benefits.

Prior to Cambridge, Dr. Wang holds both a bachelor’s degree in civil engineering and a doctorate in geotechnical engineering from the National University of Singapore and was a visiting research scholar at the Massachusetts Institute of Technology. In his doctoral research, he developed and implemented a range of self-compatible computational techniques to streamline sensor data interpretation for construction safety control and decision-making. In his subsequent independent research, he pioneered the integration of deep-learning algorithms into geotechnical reliability analyses. In his other research projects, he developed innovative machine-learning algorithms to integrate geological features into the numerical reconstruction of geological subsurface stratigraphy. At present, results from his research led to 20 articles in leading international journals, such as ASCE Journal of Geotechnical and Geoenvironmental Engineering and Canadian Geotechnical Journal, 11 conference papers and two software. He was also the recipient of the best paper award from 8^th International Symposium for Geotechnical Safety & Risk in Australia. Notably, the software he developed in his research is now being employed by government agencies for ongoing construction projects in Singapore