Preface

Data science, data visualisation, and digital twins are trending in many disciplines. Appropriate data visualisation and analytics, enabled by data science, are required for informed decision-making in a variety of sectors. If expertise in advanced data analytics techniques are available, advanced data analytics approaches such as Artificial Intelligence (AI) and real-time, web-based, and interactive visualisations are used.

Advanced data visualisation methods, such as 3D, 4D, and so on, as well as dashboards, are great tools for better communication with stakeholders, better understanding and modelling of the current situation, forecasting future trends, and digital twinning of buildings, urban neighbourhoods, infrastructure, and cities in smart cities and built environments.

Professionals, academics, managers, planners, and policymakers have discovered that improved analytical methods of data science, such as machine/deep learning or AI, promise to provide superior insights from data, allowing them to make more educated decisions. In organisations, web-based systems that visualise such insights allow rich interactions among team members, clients, project managers, and stakeholders.

This book highlights established and advanced data science and visualisation technologies, given the benefits of data science, visualisation, and digital twinning. This book is divided into three sections based on the overall themes of the chapters.

Section 1 addresses web and dashboard-based visualisations. In the first chapter of this section, Scanostics features are implemented in JavaScript to illustrate their usability in 2D, 3D, and higher dimensions on the Web. In this chapter, Pham and Dang begin by describing the mathematical definitions of these Scanostics features, then provide installation instructions and implementation scripts for using these features on multivariate data via their GitHub page. In the second chapter in this section, Alpalhao, Castro Neto, and Motta discuss the benefits of developing dashboards for better decision-making in smart cities and show off their developed dashboard for monitoring spatiotemporal mobility patterns and indicators during the COVID pandemic.

Section 2 deals with 3D modelling of trees using point cloud data and digital twinning in the mining industry. In their chapter, Egi and Eyceyurt offer a system that uses machine learning algorithms to reconstruct topography from point cloud data and utilizes 3D tree modelling in such an environment for mobile communications. They also highlight the usability of their sensor fusion technology in smart city applications. In the final chapter of this section, Kalinowski, Dlugosz, and Kaminski suggest a digital twin of a mining shaft and hoisting system utilising BIM models for project management process improvement.

Section 3 demonstrates better flood modelling as well as the predicted future growth of visual data science. In their chapter, Hale, Long, Gude, and Corns present a way for exploiting open data and deploying deep learning algorithms in a GIS framework to anticipate flood inundation profiles for planners to employ. In the final chapter, Schmidt outlines the successful application of data visualisation algorithms in various data science workflows, compares data science libraries for various applications, and draws conclusions on the potential directions for future developments of advanced data visualisations.

## **Dr. Sara Shirowzhan**

University of New South Wales, School of Built Environment, Sydney, Australia Section 1
