Towards multi-agency sensor information integration for disaster management

Highlights

• Studied the process of incorporating multi-agency sensor data for disaster management.

• Grounded on an empirical case study around sensor information transaction across emergency management sector.

• Developed components for standard-based access, harmonization, and connection of multi-agency sensor resources.

• Employed OGC Sensor Web Enablement standards for integrating multi-vendor sensor data sources.

• Implemented a GIS-based tool, IDDSS-Sensor that enables the integration of sensor data in real-time, and aids flood response.

Abstract

Having access to real-time spatial information is central to the functioning of disaster management, and in particular disaster response. Existing spatially-enabled solutions for managing urban disasters provide limited support for time-sensitivity and urgency underlying emergency situations. These approaches mainly suffer from low temporal resolution and inability to source a broad range of required disaster data, together with insufficient support for automated operations. However, disaster management procedures, integrated with in situ sensing, promise an extensive range of real-time data and automated processes to acquire and manage disaster information. In this research, we study the process of integrating multi-agency in situ sensors for supporting disaster management. For this purpose, the research was adopted in Australia as the case study area in disaster management of a flood by emphasizing on the response phase. This paper first identifies the issues and existing requirements in the process of multi-agency sensor information integration and then proposes a standard-based approach to overcoming these integration issues. Afterward, based on the presented approach and identified requirements, a GIS-based software IDDSS-Sensor is implemented to provide the functions of standard-based access, as well as on-the-fly harmonization, integration and usage of multi-agency sensor information. We evaluate the applicability of our developed approach by applying it to the use case of supporting flash flood evacuation response.

Introduction

Disaster is a major challenge in today's world that causes loss of lives and devastating impacts on infrastructures and economies. In 2014, natural disasters caused 7700 fatalities and losses of US$110 billion worldwide (Munich, 2014). Amongst all disasters, flood occurring is the most common (Leskens, Brugnach, Hoekstra, & Schuurmans, 2014). In Australia, flooding counts for an average $377 million in damages annually (Middelmann-Fernandes, 2009).

Currently, the role of spatial information and its exchange between public safety officials is part of the research agenda (Tran et al., 2009, Zlatanova et al., 2007, Mansourian et al., 2006) and is acknowledged in the current practices for multi-agency incident management (VINE, 2013). Recently, attention has been turning towards sourcing and exchanging dynamic disaster information between responding agencies for increased situational awareness (Chen et al., 2014, Farnaghi and Mansourian, 2013).

In line with this demand, in situ sensing has emerged as a spatial data sourcing technology that provides the automated collection of varied information in (near) real-time (Alamdar et al., 2014, Wang and Yuan, 2010). The complications surrounding urgency and time-sensitivity underlying emergency decision-making could be handled by enabling sensor-derived situational awareness to be shared across responder organizations. However, it poses threefold challenges: (1) ensuring interoperability between sensor data providers and disaster management authorities, (2) dealing with existing sources of inconsistencies in sensor data, and (3) derivation of actionable emergency information from raw sensor observations. The goal of this study, therefore, is to tackle these challenges by presenting an approach based on OGC Sensor Web Enablement (SWE) with the following novel contributions:

• Empirical study: The paper is grounded on an empirical case study on current processes for sensor information integration in the emergency management of Victoria, Australia.

• Conceptual development: Definition of components for standard-based sensor data access, harmonization, and connection to realize the real-time integration of multi-agency sensor resources across emergency operation centers.

• System implementation: Development of a GIS-based software tool that enables the integration of sensor information, and aids decision-making in flood response.

This work builds on our earlier article: Alamdar et al., 2014 which set out a thorough survey of existing approaches to the state-of-the-art sensor monitoring research for disaster management.

The paper is structured as follows: Section 2 first provides a review of associated concepts, theories and related work. Next, section 3 outlines the results of the case study. Following this, section 4, presents the new approach for multi-agency sensor information integration and sets out its conceptual framework. On this basis, section 5 presents IDDSS-Sensor by describing the associated architecture, technologies and implementation results. Next, section 6 discusses the lessons learned from this deployment and future research considerations. Finally, the conclusion remarks are described in section 7.