Uptake of sensor data in emergency management

Abstract

While disasters are becoming larger, more complex and more frequent, traditional emergency management response capacities are not increasing at the same rate. Sensor capabilities could fill this gap by providing improved situational awareness, or intelligence, for emergency managers. Data from sensors is increasing exponentially in quality and quantity while the cost of capturing and processing these data is decreasing. This creates immense opportunities to bring sensor data into emergency management practices. Unfortunately, not all sensors are created equal. The accuracy, precision, presentation and timeliness of data varies depending on the source, the way the product is structured and who produces it. It is therefore difficult for emergency managers to incorporate sensor data into decision making, particularly when they have not seen the data type before, and do not know where it originated, or how to use. This thesis researches how data product creators can tailor products to increase the likelihood of their product being incorporated in emergency management decision making. It focuses on the issue of data product uptake, which is inclusion of data products in decision making processes. This issue has been poorly covered in the existing literature. This thesis synthesises literature from a range of disciplines then designs and conducts three targeted studies to build upon this knowledge. The first study compares four international data systems which use the same data source but make different choices in the design of their products, this then provides examples of the impacts of these design choices. The second study looks at disaster inquiries in Australia to consider how sensor data has been used in decision making in the past, and what lessons have been learnt from these experiences. The third study surveys Australian emergency managers to collect their views on what products they use, trust and what factors lead to that trust. The results from these studies combine to create a comprehensive collection of design choices available to data product creators. This collection covers not just technical choices like accuracy, but also presentational and data policy choices, to create a more holistic picture of how creators can influence their products. The collection is then presented in a framework which, if applied throughout product development, would be expected to increase uptake of sensor data in emergency management decision making. Design choices and user-oriented design processes are emphasised as a crucially important yet poorly-examined aspect of data uptake in emergency management. This thesis finds that trust is key to whether emergency managers use a product or not, and that trust is created through a series of design choices which can be grouped into quality, reputation, maturity and data policy. Show more