S. M. Dassanayake, I. Mahakalanda, D. M. R. Sanjula, B. Dissanayake, R. M. Pasan, I. Gunathunga, et al. (2023). Geospatial Impact Analytics of Hydrometeorological Hazards: A Study on Urban and Suburban Floods in Sri Lanka using Online Textual Data. In V. L. Thomas J. Huggins (Ed.), Proceedings of the ISCRAM Asia Pacific Conference 2022 (pp. 156–163). Palmerston North, New Zealand: Massey Unversity.
Abstract: Urban and suburban communities in tropical countries like Sri Lanka typically experience hydrometeorological hazards that substantially damage property and lives. Although accurate forecasts of weather events are available, the decision-makers often fail to mitigate the actual impact of these forecasts alone. The adverse impacts experienced by the community and reported by news and online media complement this fact. The forecast-impact disparity underpins the scope for holistically linking the forecast data with actual impact. This paper presents a work-in-progress study that develops a geospatial analytics framework using online textual data for assessing the spatiotemporal impact of the hydrometeorological hazards in disaster hot spots. The preliminary findings show prospects for extending the study to impact-focused visualization and forecasting that capture the community's and decision makers' attention for better interventions. For example, these include the degree of disaster response, planning and scheduling critical infrastructure and estimating damages, compensations and insurance claims.
|
|
Vihan C.N. Weeraratne, Raymond C.Z. Cohen, Mahesh Prakash, Lalitha Ramachandran, Nikhil Garg, & Valentijn Pauwels. (2023). Assessing Climate Vulnerability Under Future Changes to Climate, Demographics and Infrastructure: A Case Study for the Chapel Street Precinct, Melbourne. In V. L. Thomas J. Huggins (Ed.), Proceedings of the ISCRAM Asia Pacific Conference 2022 (pp. 35–44). Palmerston North, New Zealand: Massey Unversity.
Abstract: The Chapel Street Precinct is a busy commercial and residential corridor in the City of Stonnington Local Government Area (LGA) located in metropolitan Melbourne, Australia. Authorities and planners in the LGA are interested in understanding how the changing climate affects the socioeconomic environment of the region. By considering existing climate hazards (such as extreme heat, flood and water availability), infrastructure, and demographic information in the region together with future projections of climate change and demographic changes, a Socioeconomic Vulnerability Index (SVI) was created at a Mesh Block scale to better identify relatively high-risk Mesh Blocks in the region. The climate projections under medium and high future emission scenarios (i.e., representative concentration pathways (RCP)) as per IPCC (Intergovernmental Panel on Climate Change) fifth assessment report (AR5), RCP4.5 and RCP8.5 respectively for 30-year epochs around 2030, 2050 and 2070 were used in the SVI development. The current-day scenario is considered under Baseline conditions for demographic and asset information representing present-day conditions, whereas the baseline climate dataset considers the climate for the 30 year period 1991-2020 to best represent the present-day climate. The multi-model mean of the future climate projections from 6 different climate models were obtained from the Victoria’s Future Climate tool (https://vicfutureclimatetool.indraweb.io), developed by CSIRO (Commonwealth Scientific and Industrial Research Organisation) Data61 together with the Department of Environment, Land, Water and Planning (DELWP) under Data61’s INDRA framework (https://research.csiro.au/indra/). A version of INDRA is currently under development to allow map-based interactivity, experimentation and scrutiny of the vulnerability indices and their subcomponents across the study region. The SVI was created using a weighted indicator approach utilising a range of indicators belonging to 3 categories, exposure, susceptibility, and baseline adaptive capacity. The indicators were first normalised and the final SVI was given a score between 0-1 for each Mesh Block. The worst levels of vulnerability were observed to be for the RCP8.5 2070 scenario. In general, the RCP8.5 scenarios indicated a worse outcome compared to the RCP4.5 scenario. The area along Chapel Street within the precinct which is a densely built-up area high in population was found to be the most vulnerable area in the study region. It is foreseen that decision makers will be able to use the holistic data-driven outcomes of this study to make better informed decisions whilst adapting to climate change.
|
|
Florent Dubois, Paul Renaud-Goud, & Patricia Stolf. (2022). Dynamic Capacitated Vehicle Routing Problem for Flash Flood Victim’s Relief Operations. In Rob Grace, & Hossein Baharmand (Eds.), ISCRAM 2022 Conference Proceedings – 19th International Conference on Information Systems for Crisis Response and Management (pp. 68–86). Tarbes, France.
Abstract: Flooding relief operations are Dynamic Vehicle Routing Problems (DVRPs). The problem of people evacuation is addressed and formalized in this paper. Characteristics of this DVRP problem applied to the crisis management context and to the requirements of the rescue teams are explained. In this paper, several heuristics are developed and assessed in terms of performance. Two heuristics are presented and adapted to the dynamic problem in a re-optimization approach. An insertion heuristic that inserts demands in the existing plan is also proposed. The evaluation is conducted on various dynamic scenarios with characteristics based on a study case. It reveals better performances for the heuristics with a re-optimization approach.
|
|
Diego Fabian Pajarito Grajales, Livia Castro Degrossi, Daniel Barros, Mohammed Rizwan Khan, Fernanda Lima E Silva, Maria Alexandra Cunha, et al. (2022). Enabling Participatory Flood Monitoring Through Cloud Services. In Rob Grace, & Hossein Baharmand (Eds.), ISCRAM 2022 Conference Proceedings – 19th International Conference on Information Systems for Crisis Response and Management (pp. 213–223). Tarbes, France.
Abstract: Flooding events are more impactful due to climate change, while traditional top-down approaches to flood management give way to new initiatives that consider citizens and communities as active strategic actors. Researchers and practitioners have started to place communities in the centre of creation processes or invite them to co-design digital platforms. However, many citizen science projects re-use well-known technological components without reflecting about how the technology is able to effectively support citizen participation in data generation, including the provision of flexible data storage and exchange. This paper describes a novel digital platform design which adopts cloud services to integrate official and citizen-generated data about urban flooding. It summarises the results of a participatory design process of a digital platform to collect, store and exchange flood-related data, which includes components such as data lakes, Application Programming Interfaces (APIs), and web and mobile interfaces. This work in progress paper presents insights and lessons learned from using cloud services to enable citizen participation and engage communities with flood monitoring.
|
|
Ahmed Laatabi, Benoit Gaudou, Chihab Hanachi, Patricia Stolf, & Sébastien Truptil. (2022). Coupling Agent-based Simulation with Optimization to Enhance Population Sheltering. In Rob Grace, & Hossein Baharmand (Eds.), ISCRAM 2022 Conference Proceedings – 19th International Conference on Information Systems for Crisis Response and Management (pp. 116–132). Tarbes, France.
Abstract: Population sheltering is a recurrent problem in crisis management that requires addressing two aspects: evacuating vulnerable people using emergency vehicles and regulating movements of pedestrians and individual vehicles towards shelters. While these aspects have received considerable attention in modeling and simulation literature, very few approaches consider them simultaneously. In this paper, we argue that Agent-Based Modeling and Simulation (ABMS) and Optimization are two complementary approaches that can address the problem of sheltering globally and efficiently and be the basis of coherent frameworks for decision- and policy-making. Optimization can build efficient sheltering plans, and ABMS can explore what-if scenarios and use geospatial data to display results within a realistic environment. To illustrate the benefits of a framework based on this coupling approach, we simulate actual flash flood scenarios using real-world data from the city of Trèbes in South France. Local authorities may use the developed tools to plan and decide on sheltering strategies, notably, when and how to evacuate depending on available time and resources.
|
|