What led you to be interested in this study?

In line with the United Nations Sustainable Development Goal (SDG) 11, which aims to make cities inclusive, safe, resilient, and sustainable, cities require the ability to better respond to and be prepared for unforeseen events. As incidents resulting from natural disasters such as floods or traffic accidents can become quite complex and span several administrative boundaries and multiple different systems, the demand for data to assess potential impacts increases. 

To better support stakeholders involved in incident response, we need to combine different datasets, such as static data that exists prior to the incident, including topography, administrative borders, infrastructure and population data, with dynamic data that may change throughout the incident.

In this study, we develop an incident response prototype using real-time data and cloud computing technology to store, manage, and analyse frequent incoming big data streams in (near) real-time. Ultimately, this prototype can lead to a digital twin providing a digital representation of real-world objects, systems and processes that can help to monitor the city, capture any changing conditions and test new inventions in a safe environment. 

Who can benefit from this study?

Stakeholders involved in operational incident response and incident management, who work with different (heterogeneous) data, systems and infrastructure can benefit from this research. We also anticipate our study to provide a foundation for the future design of a data ontology supporting multi-agency incident response in smart cities.

What was the process of research like?

Based on the concept that a smart city consists of numerous interconnected entities that together form a system of systems, we adapt the systems engineering approach following steps 1-6: (1) Designing the research problem; (2) Identifying stakeholder needs; (3) Analysing functional requirements; (4) Developing system design; (5) Feedback system design; (6) Developing system prototype.

Using the systems engineering approach, we identify current stakeholders' challenges and formulate requirements for an improved system. Based on these findings, we develop a use case using Microsoft Azure cloud computing technology that provides initial operational support and assists collaboration across agencies through data information sharing and analysis.

What was the process of collaboration like in this study?

Throughout the study, we benefited from numerous conversations with stakeholders involved in operational incident response, such as blue-light services, organisations responsible for data collection and management, and parties conducting data analysis at the local, regional, and national levels. 

We covered questions in the following areas: Processes: What are the steps in current incident response? People: Who are the different stakeholders involved in incident response? Data: What data do you require? How is that data effectively shared? Technology: What technology and systems do you use? and Analytics: What analytical capabilities can support future incident response?

Since feedback is an integral part of the applied methodology, stakeholder engagement will continue throughout the research to provide further insights for my overall PhD research, e.g., user interface.

Why do you think this study is important?

Despite the advances in internet-enabled technology, cloud-based computing resources and practical multi-agency incident response, common challenges identified are silo working across multiple agencies, heterogeneous systems and a lack of integrated real-time data on current weather, traffic, and hazard conditions. In this study, we bring together a wide range of stakeholders from different fields related to emergency and disaster response and ensure that different perspectives are integrated in the prototype development.

While the focus was on integrating disparate data sources for a specific incident response problem, the underlying technology can be applied to other areas. A key criterion is that the technology is tailored to and co-developed with the end users in mind. In this sense, this methodology can serve as an enabler that supports stakeholders in interdisciplinary and complex project environments.

How do the study results build on existing knowledge and improve our understanding of future incident response?

Currently, many existing smart city applications are independent systems that run isolated and do not exchange data across their systems with other domains. Historical incident examples also demonstrate challenges in interconnected multi-agency relationships, such as the lack of incident-related information and collaboration.

Developing the incident response prototype collaboratively with end users from multiple agencies is critical to limit the risks of failed collaboration that can result if the system is designed for only one end user. Further, this study contributes to broader research on: 

• how to integrate heterogeneous data between distributed multi-agency emergency systems;
• how to apply processing steps to continuous real-time data streams; and
• how to analyse and visualise location-based incident data to enhance the understanding of the current incident environment.

Using the incident response prototype, stakeholders can view available resources, send automatic updates and integrate location-based real-time weather and traffic data. The figure below visualises how the current system prototype supports route navigation from closest responder to incident site using real-time traffic and weather data:

Integrating real-time data such as traffic, weather and flood information can help coordinate multiple services, such as navigating emergency vehicles to a site where an accident has occurred somewhere on route. Through future development, real-time IoT data on pedestrian and traffic movement can further improve contextual incident response in a smart, sustainable, and resilient city.

How might this impact the future?

This study aims to improve future multi-agency collaboration by showing how we can best achieve meaningful interpretation of multi-agency incident-related data through an integrated response workflow. What we show in this study using the incident response example can be applied to many other multi-agency environments. By connecting interdependent data across existing systems and stakeholder boundaries, we can provide stakeholders with a more comprehensive operational picture and enable more efficient and informed decision support - demonstrating the real opportunity of what a smart city can deliver.

**********************************************************************************

This study was conducted as part of the United Kingdom Research and Innovation (UKRI) Centre for Doctoral Training (CDT) in Geospatial Systems at Newcastle University (UK) under the PhD title “Multi-scale multi domain geospatial data modelling”, funded by Engineering and Physical Sciences Research Council (EPSRC) and Ordnance Survey, Great Britain’s national mapping agency.

The results of this study inform the overarching PhD research that focusses on real-time data integration for supporting multi-agencies in incident response and impact assessment.