Browsing by Author "Rye, Sara"
Now showing 1 - 3 of 3
Results Per Page
Sort Options
Item Open Access A multi-attribute decision support system for allocation of humanitarian cluster resources based on decision makers’ perspective(MDPI, 2022-10-18) Rye, Sara; Aktas, EmelThe rush of the humanitarian suppliers into the disaster area proved to be counterproductive. To reduce this proliferation problem, the present research is designed to provide a technique for supplier ranking/selection in disaster response using the principles of utility theory. A resource allocation problem is solved using optimisation based on decision maker’s preferences. Due to the lack of real-time data in the first 72 h after the disaster strike, a Decision Support System (DSS) framework called EDIS is introduced to employ secondary historical data from disaster response in four humanitarian clusters (WASH: Water, Sanitation and Hygiene, Nutrition, Health, and Shelter) to estimate the demand of the affected population. A methodology based on multi-attribute decision making (MADM), Analytical Hierarchy processing (AHP) and Multi-attribute utility theory (MAUT) provides the following results. First a need estimation technique is put forward to estimate minimum standard requirements for disaster response. Second, a method for optimization of the humanitarian partners selection is provided based on the resources they have available during the response phase. Third, an estimate of resource allocation is provided based on the preferences of the decision makers. This method does not require real-time data from the aftermath of the disasters and provides the need estimation, partner selection and resource allocation based on historical data before the MIRA report is released.Item Open Access A rule-based predictive model for estimating human impact data in natural onset disasters—the case of a pred model(MDPI, 2023-05-26) Rye, Sara; Aktas, EmelBackground: This paper proposes a framework to cope with the lack of data at the time of a disaster by employing predictive models. The framework can be used for disaster human impact assessment based on the socio-economic characteristics of the affected countries. Methods: A panel data of 4252 natural onset disasters between 1980 to 2020 is processed through concept drift phenomenon and rule-based classifiers, namely the Moving Average (MA). Results: Predictive model for Estimating Data (PRED) is developed as a decision-making platform based on the Disaster Severity Analysis (DSA) Technique. Conclusions: comparison with the real data shows that the platform can predict the human impact of a disaster (fatality, injured, homeless) with up to 3% error; thus, it is able to inform the selection of disaster relief partners for various disaster scenarios.Item Open Access Serious games as a validation tool for PREDIS: a decision support system for disaster management(MDPI, 2022-12-09) Rye, Sara; Aktas, EmelIn this paper, we validate PREDIS, a decision support system for disaster management using serious games to collect experts’ judgments on its performance. PREDIS is a model for DISaster response supplier selection (PREDIS). It has a PREDictive component (PRED) for predicting the disaster human impact and an estimation component to Estimate the DISaster (EDIS) needs to optimise supplier-based resource allocation. A quasi-experiment design embedded in a participatory simulation game is conducted to compare the opinions of equal samples of 22 experts and non-experts. The following questions are put forward. First, “Does PREDIS model assists the decision makers to make the same decisions faster?” Second, “Does the PREDIS model assist the non-experts as simulated decision makers to decide like an expert?” Using AHP weights of decision makers’ preferences as well as Borda counts, the decisions are compared. The result shows that PREDIS helps to reduce the decision-making time by experts and non-experts to 6 h after the disaster strike, instead of the usual 72 h. It also assists 71% of the non-experts to make decisions similar to those made by experts. In summary, the PREDIS model has two major capabilities. It enables the experts and non-experts to predict the disaster results immediately using widely available data. It also enables the non-experts to decide almost the same as the experts; either in predicting the human impact of a disaster and estimating the needs or in selecting suitable suppliers.