Abstract: Decision making in extreme events presents a difficult challenge to emergency managers who are legally responsible for protecting life, property, and maintaining continuity of operations for their respective organizations or communities. Prior research has identified the benefits of gaining situation awareness in rapidly changing disaster contexts, but situation awareness is not always sufficient. We have investigated “option awareness” and the decision space to provide cognitive support for emergency managers to simulate computationally possible outcomes of different options before they make a decision. Employing a user-centered design process, we developed a computational model that rapidly generates ranges of likely outcomes for different options and displays them visually through a prototype decision-space interface that allows rapid comparison of the options. Feedback from emergency managers suggests that decision spaces may enable emergency managers to consider a wider range of options for decisions and may facilitate more targeted, effective decision making under uncertain conditions.

Abstract: Problems of collaborative sensemaking are evident in major incident response where sharing salient information is key to the shared understanding of the situation. In this paper we propose that differences in sensemaking performance can be captured through quantitative methods derived from consideration of network structure and information diffusion as the group collaborates to achieve consensus in a problem-solving task. We present analysis from a large international study in which groups of people collaborate to solve an intelligence analysis problem. Our initial analysis suggests that 'edge' groups are able to collaborate more efficiently and perform better than those which have a hierarchical control structure.

Abstract: This paper presents a novel, advanced research approach to investigate organizational safety culture as a complex phenomenon, combining agent modeling and organizational ethnography. Safety culture is an emergent property of organizations that largely influences the resilience of organizational responses in crisis situations. However, theory describing the precise ways in which safety culture influences resilience is lacking. Thus the first step is to understand how safety culture gradually emerges from interactions between formal and informal organizational processes. The paper explains the proposed research methodology illustrated by a case of an aircraft maintenance organization. A preliminary analysis is performed from which a conceptual model is derived, and the subsequent simulation and automated analytical techniques that will be used to validate the model and gain new insights are explained.

Abstract: A rapid, holistic view of earthquake disasters begins with earthquake location and magnitude, alerted by seismic networks. The initial source characteristics, along with any available ground-shaking observations, can be used to rapidly estimate the shaking extent, its severity (e.g., ShakeMap), and its likely impact to society, for example, employing the U.S. Geological Survey's Prompt Assessment of Global Earthquakes for Response, or PAGER, system. When serous impacts are likely, PAGER's impact-based alerts can, in turn, begin the process of primary response at the local, national, or international level, and the process of reconnaissance via social media, the mainstream media, scientific analyses, and remotely-sensed and ground-truth observations. In this work-in-progress report, we describe our initial efforts to incorporate event-specific ground-truth observations and model secondary ground-failure hazards back into the loss-modeling domain in order to provide a more holistic view each earthquake disaster.