Table of Links Abstract and 1 Introduction 1.1. Spatial Digital Twins (SDTs) 1.2. Applications 1.3. Different Components of SDTs 1.4. Scope of This Work and Contributions 2. Related Work and 2.1. Digital Twins and Variants 2.2. Spatial Digital Twin Case Studies 3. Building Blocks of Spatial Digital Twins and 3.1. Data Acquisition and Processing 3.2. Data Modeling, Storage and Management 3.3. Big Data Analytics System 3.4. Maps and GIS Based Middleware 3.5. Key Functional Components 4. Other Relevant Modern Technologies and 4.1. AI & ML 4.2. Blockchain 4.3. Cloud Computing 5. Challenges and Future Work, and 5.1. Multi-modal and Multi-resolution Data Acquisition 5.2. NLP for Spatial Queries and 5.3. Benchmarking the Databases and Big Data Platform for SDT 5.4. Automated Spatial Insights and 5.5. Multi-modal Analysis 5.6. Building Simulation Environment 5.7. Visualizing Complex and Diverse Interactions 5.8. Mitigating the Security and Privacy Concerns 6. Conclusion and References 5.4. Automated Spatial Insights Spatial digital twins generate vast amounts of data, often sourced from a wide variety of origins. It is critical to be able to automatically identify interesting insights from such data without the need of a human input. Furthermore, the techniques that can predict future behavior, risks, opportunities and trends are also important so that appropriate actions can be taken. While automatically identifying insights has been studied [77, 78, 79], none of these techniques are designed specifically for the spatial data. Therefore, these techniques cannot provide spatial insights which are crucial for the operation and management of spatial digital twins. Spatial insights can take various forms, such as detecting neighborhoods with abnormally high greenhouse gas emissions, uncovering spatial correlations between different attributes such as air quality and traffic accidents in various parts of a city, or mapping and highlighting regions with above-average crime rates and their correlation with seemingly disparate attributes such as electricity usage, waste production, and traffic patterns. Furthermore, temporal aspects of spatial data must also be considered in generating insights. For example, it might be interesting to study how the spatial correlation between two or more attributes evolves over time. Unfortunately, the existing techniques cannot be applied or easily extended for spatial digital twins due to their inability to consider spatial features. Furthermore, it is crucial to design techniques that are efficient so that the insights can be generated in a timely manner, allowing system operators to intervene promptly if necessary. 5.5. Multi-modal Analysis The integration of image and text in multimodal language models, such as CLIP [103], enables them to learn the data space jointly and effectively address challenges related to multimodal data. Furthermore, the remarkable advancements in multimodal large generative models, like GPT-4 [99], have demonstrated significant potential in the field of multimodal learning. We envision that it is possible to answer many queries and to find interesting insights from the captured satellite image/drone image (i.e., raster spatial data) with other spatial and non-spatial data. More specifically, combining different forms of data such as real-time data (e.g., traffic, energy consumption) collected by sensors, images taken by a satellite or a drone, spatial neighborhood features, and train a large multimodal model may be able to generate useful insights. 5.6. Building Simulation Environment As many factors such as social interaction, economic factors, and human factors may not be possible to be captured in the SDTs, it is important to build a simulated environment, where these factors can be simulated so that their association with the captured SDT data can be assessed. Future research should focus on how to build a realistic simulated environment tailored for the SDTs. Different simulation software like AnyLogic, OpenStudio, and Simio, have been developed for applications such as transportation, logistics, and manufacturing. As the scope and scale of an SDT are significantly different from that of DTs, there is a potential avenue for research on how to develop a platform for simulating different factors involving an SDT. 5.7. Visualizing Complex and Diverse Interactions An SDT involves various forms of data such as infrastructure data (e.g., 3D buildings, roads, etc.), sensor data (e.g., energy/gas/water consumption, traffic, etc.), and social media data (e.g., Twitter, Instagram, etc.). There are a number of data visualization challenges involving the interaction of these complex data objects. For example, as the visualization of more than three dimension is not comprehensible for typical users, it is hard to combine different forms of data, e.g., 3D building with time series energy consumption of the building on a map. Also as the data can be of higher dimension and of different types, visualizing correlations among those datasets across spatial and temporal dimension needs more research. Also, how to add and visualize different layers of data attached to a particular location on a map or GIS software to observe interesting insights and interaction of the associated data needs to be investigated. Authors: (1) Mohammed Eunus Ali, Department of Computer Science and Engineering, Bangladesh University of Engineering and Technology, ECE Building, Dhaka, 1000, Bangladesh; (2) Muhammad Aamir Cheema, Faculty of Information Technology, Monash University, 20 Exhibition Walk, Clayton, 3164, VIC, Australia; (3) Tanzima Hashem, Department of Computer Science and Engineering, Bangladesh University of Engineering and Technology, ECE Building, Dhaka, 1000, Bangladesh; (4) Anwaar Ulhaq, School of Computing, Charles Sturt University, Port Macquarie, 2444, NSW, Australia; (5) Muhammad Ali Babar, School of Computer and Mathematical Sciences, The University of Adelaide, Adelaide, 5005, SA, Australia. This paper is available on arxiv under CC BY 4.0 DEED license. Table of Links Abstract and 1 Introduction Abstract and 1 Introduction 1.1. Spatial Digital Twins (SDTs) 1.1. Spatial Digital Twins (SDTs) 1.2. Applications 1.2. Applications 1.3. Different Components of SDTs 1.3. Different Components of SDTs 1.4. Scope of This Work and Contributions 1.4. Scope of This Work and Contributions 2. Related Work and 2.1. Digital Twins and Variants 2. Related Work and 2.1. Digital Twins and Variants 2.2. Spatial Digital Twin Case Studies 2.2. Spatial Digital Twin Case Studies 3. Building Blocks of Spatial Digital Twins and 3.1. Data Acquisition and Processing 3. Building Blocks of Spatial Digital Twins and 3.1. Data Acquisition and Processing 3.2. Data Modeling, Storage and Management 3.2. Data Modeling, Storage and Management 3.3. Big Data Analytics System 3.3. Big Data Analytics System 3.4. Maps and GIS Based Middleware 3.4. Maps and GIS Based Middleware 3.5. Key Functional Components 3.5. Key Functional Components 4. Other Relevant Modern Technologies and 4.1. AI & ML 4. Other Relevant Modern Technologies and 4.1. AI & ML 4.2. Blockchain 4.2. Blockchain 4.3. Cloud Computing 4.3. Cloud Computing 5. Challenges and Future Work, and 5.1. Multi-modal and Multi-resolution Data Acquisition 5. Challenges and Future Work, and 5.1. Multi-modal and Multi-resolution Data Acquisition 5.2. NLP for Spatial Queries and 5.3. Benchmarking the Databases and Big Data Platform for SDT 5.2. NLP for Spatial Queries and 5.3. Benchmarking the Databases and Big Data Platform for SDT 5.4. Automated Spatial Insights and 5.5. Multi-modal Analysis 5.4. Automated Spatial Insights and 5.5. Multi-modal Analysis 5.6. Building Simulation Environment 5.6. Building Simulation Environment 5.7. Visualizing Complex and Diverse Interactions 5.7. Visualizing Complex and Diverse Interactions 5.8. Mitigating the Security and Privacy Concerns 5.8. Mitigating the Security and Privacy Concerns 6. Conclusion and References 6. Conclusion and References 5.4. Automated Spatial Insights Spatial digital twins generate vast amounts of data, often sourced from a wide variety of origins. It is critical to be able to automatically identify interesting insights from such data without the need of a human input. Furthermore, the techniques that can predict future behavior, risks, opportunities and trends are also important so that appropriate actions can be taken. While automatically identifying insights has been studied [77, 78, 79], none of these techniques are designed specifically for the spatial data. Therefore, these techniques cannot provide spatial insights which are crucial for the operation and management of spatial digital twins. Spatial insights can take various forms, such as detecting neighborhoods with abnormally high greenhouse gas emissions, uncovering spatial correlations between different attributes such as air quality and traffic accidents in various parts of a city, or mapping and highlighting regions with above-average crime rates and their correlation with seemingly disparate attributes such as electricity usage, waste production, and traffic patterns. Furthermore, temporal aspects of spatial data must also be considered in generating insights. For example, it might be interesting to study how the spatial correlation between two or more attributes evolves over time. Unfortunately, the existing techniques cannot be applied or easily extended for spatial digital twins due to their inability to consider spatial features. Furthermore, it is crucial to design techniques that are efficient so that the insights can be generated in a timely manner, allowing system operators to intervene promptly if necessary. 5.5. Multi-modal Analysis The integration of image and text in multimodal language models, such as CLIP [103], enables them to learn the data space jointly and effectively address challenges related to multimodal data. Furthermore, the remarkable advancements in multimodal large generative models, like GPT-4 [99], have demonstrated significant potential in the field of multimodal learning. We envision that it is possible to answer many queries and to find interesting insights from the captured satellite image/drone image (i.e., raster spatial data) with other spatial and non-spatial data. More specifically, combining different forms of data such as real-time data (e.g., traffic, energy consumption) collected by sensors, images taken by a satellite or a drone, spatial neighborhood features, and train a large multimodal model may be able to generate useful insights. 5.6. Building Simulation Environment As many factors such as social interaction, economic factors, and human factors may not be possible to be captured in the SDTs, it is important to build a simulated environment, where these factors can be simulated so that their association with the captured SDT data can be assessed. Future research should focus on how to build a realistic simulated environment tailored for the SDTs. Different simulation software like AnyLogic, OpenStudio, and Simio, have been developed for applications such as transportation, logistics, and manufacturing. As the scope and scale of an SDT are significantly different from that of DTs, there is a potential avenue for research on how to develop a platform for simulating different factors involving an SDT. 5.7. Visualizing Complex and Diverse Interactions An SDT involves various forms of data such as infrastructure data (e.g., 3D buildings, roads, etc.), sensor data (e.g., energy/gas/water consumption, traffic, etc.), and social media data (e.g., Twitter, Instagram, etc.). There are a number of data visualization challenges involving the interaction of these complex data objects. For example, as the visualization of more than three dimension is not comprehensible for typical users, it is hard to combine different forms of data, e.g., 3D building with time series energy consumption of the building on a map. Also as the data can be of higher dimension and of different types, visualizing correlations among those datasets across spatial and temporal dimension needs more research. Also, how to add and visualize different layers of data attached to a particular location on a map or GIS software to observe interesting insights and interaction of the associated data needs to be investigated. Authors: (1) Mohammed Eunus Ali, Department of Computer Science and Engineering, Bangladesh University of Engineering and Technology, ECE Building, Dhaka, 1000, Bangladesh; (2) Muhammad Aamir Cheema, Faculty of Information Technology, Monash University, 20 Exhibition Walk, Clayton, 3164, VIC, Australia; (3) Tanzima Hashem, Department of Computer Science and Engineering, Bangladesh University of Engineering and Technology, ECE Building, Dhaka, 1000, Bangladesh; (4) Anwaar Ulhaq, School of Computing, Charles Sturt University, Port Macquarie, 2444, NSW, Australia; (5) Muhammad Ali Babar, School of Computer and Mathematical Sciences, The University of Adelaide, Adelaide, 5005, SA, Australia. Authors: Authors: (1) Mohammed Eunus Ali, Department of Computer Science and Engineering, Bangladesh University of Engineering and Technology, ECE Building, Dhaka, 1000, Bangladesh; (2) Muhammad Aamir Cheema, Faculty of Information Technology, Monash University, 20 Exhibition Walk, Clayton, 3164, VIC, Australia; (3) Tanzima Hashem, Department of Computer Science and Engineering, Bangladesh University of Engineering and Technology, ECE Building, Dhaka, 1000, Bangladesh; (4) Anwaar Ulhaq, School of Computing, Charles Sturt University, Port Macquarie, 2444, NSW, Australia; (5) Muhammad Ali Babar, School of Computer and Mathematical Sciences, The University of Adelaide, Adelaide, 5005, SA, Australia. This paper is available on arxiv under CC BY 4.0 DEED license. This paper is available on arxiv under CC BY 4.0 DEED license. available on arxiv available on arxiv
