Duncan T. Wilson, Glenn I. Hawe, Graham Coates, & Roger S. Crouch. (2013). Scheduling response operations under transport network disruptions. In J. Geldermann and T. Müller S. Fortier F. F. T. Comes (Ed.), ISCRAM 2013 Conference Proceedings – 10th International Conference on Information Systems for Crisis Response and Management (pp. 683–687). KIT; Baden-Baden: Karlsruher Institut fur Technologie.
Abstract: Modeling the complex decision problems faced in the coordination of disaster response as a scheduling problem to be solved using an optimization algorithm has the potential to deliver efficient and effective support to decision makers. However, much of the utility of such a model lies in its ability to accurately predict the outcome of any proposed solution. The stochastic nature of the disaster response environment can make such prediction difficult. In this paper we examine the effect of unknown disruptions to the road transport network on the utility of a disaster response scheduling model. The effects of several levels of disruption are measured empirically and the potential of using real-time information to revise model parameters, and thereby improve predictive performance, is evaluated.
|
Duncan T. Wilson, Glenn I. Hawe, Graham Coates, & Roger S. Crouch. (2012). Estimating the value of casualty health information to optimization-based decision support in response to major incidents. In Z.Franco J. R. L. Rothkrantz (Ed.), ISCRAM 2012 Conference Proceedings – 9th International Conference on Information Systems for Crisis Response and Management. Vancouver, BC: Simon Fraser University.
Abstract: In this paper we describe a work-in-progress decision support program designed for use in the response to major incidents in the UK. The proposed program is designed for use in a continuous fashion, where the updating of its model, the search for solutions to the model through an optimization algorithm, and the issuing of these solutions are carried out concurrently. The model facilitates the inclusion of dynamic and uncertain features of emergency response. The potential of such an approach to deliver high-quality response plans through enabling more accurate modeling is evaluated through focusing on the case of casualty health information. Computational experiments show there is significant value in monitoring the dynamic and uncertain health progression of casualties and updating the model accordingly. © 2012 ISCRAM.
|
Glenn I. Hawe, Duncan T. Wilson, Graham Coates, & Roger S. Crouch. (2012). STORMI: An agent-based simulation environment for evaluating responses to major incidents in the UK. In Z.Franco J. R. L. Rothkrantz (Ed.), ISCRAM 2012 Conference Proceedings – 9th International Conference on Information Systems for Crisis Response and Management. Vancouver, BC: Simon Fraser University.
Abstract: This paper describes work-in-progress regarding STORMI, an agent-based simulation environment for evaluating the response by the emergency services to hypothetical major incidents in the UK. At present, STORMI consists of two main components: a Scenario Designer and a Simulator. The Scenario Designer enables the setting up of a hypothetical multi-site mass casualty incident anywhere in the UK, along with the resources which may be considered for responding to it. This provides input to the Simulator, which through its Multiple Program Multiple Data architecture, models the agents and their environment at a higher level of detail inside incident sites than it does outside, thus focusing attention on the areas of most interest. Furthermore, the multiple programs of the Simulator execute concurrently, thus targeting multi-core processors. © 2012 ISCRAM.
|
Glenn I. Hawe, Graham Coates, Duncan T. Wilson, & Roger S. Crouch. (2011). Design decisions in the development of an agent-based simulation for large-scale emergency response. In E. Portela L. S. M.A. Santos (Ed.), 8th International Conference on Information Systems for Crisis Response and Management: From Early-Warning Systems to Preparedness and Training, ISCRAM 2011. Lisbon: Information Systems for Crisis Response and Management, ISCRAM.
Abstract: As part of ongoing research into optimizing the response to large-scale emergencies, an agent-based simulation (ABS) is being developed to evaluate different rescue plans in silico. During the development of this software, decisions regarding its design have been required in order to best satisfy the following specific application requirements: (1) the construction of a sufficiently detailed virtual environment, representing a real geographical area; (2) the programming of a wide variety of agent behaviors using a minimal amount of code; (3) the computational handling of the “large-scale” nature of the emergency; and (4) the presentation of a highly visual user interface, to encourage and facilitate use of the software by practitioners involved in the project. This paper discusses the decisions made in each of these areas, including the novel use of policy-based class design to efficiently program agents. Future developments planned for the software are also outlined.
|
Graham Coates, Glenn I. Hawe, Duncan T. Wilson, & Roger S. Crouch. (2011). Adaptive co-ordinated emergency response to rapidly evolving large-scale unprecedented events (REScUE). In E. Portela L. S. M.A. Santos (Ed.), 8th International Conference on Information Systems for Crisis Response and Management: From Early-Warning Systems to Preparedness and Training, ISCRAM 2011. Lisbon: Information Systems for Crisis Response and Management, ISCRAM.
Abstract: This paper presents an overview of ongoing research into the development of an integrated framework aimed at adaptive co-ordination of emergency response to dynamic, fast evolving and novel events on a large-scale. The framework consists of (i) a decision support system, supported by rapid adaptive search methods, to enable the real time development of tailored response plans including emergency responder team composition and task allocation to these teams, and (ii) an agent-based simulation of emergency response to large-scale events occurring in real geographical locations. The aim of this research is to contribute to understanding how better agent-based simulation coupled with decision support can be used to enable the effective co-ordination of emergency response, involving the collective efforts and actions of multiple agencies (ambulance services, fire brigades, police forces and emergency planning units), to rapidly evolving large-scale unprecedented events.
|