Comparison precision of Classical Methods and Analytic Hierarchy Process (AHP) in evaluating parameters affecting flood risk (Case study: Dezful flood, April 2020)

Abstract. This research compares two methods for evaluating parameters affecting flood risk in Dezful area: the Analytic Hierarchy Process (AHP) and the classical method (simple average). The parameters under consideration include soil permeability, drainage density, distance from the river, elevation, slope, and aspect. The AHP method utilize pairwise comparisons to weight the parameters, whereas classical method assigned equal weights to all parameters. The results calculate based on these respective methodologies.
To assess the accuracy of the results, the Modified Normalized Difference Water Index (MNDWI) is extracted from the Landsat8-OLI sensor dated April 8, 2019, as the ground truth map, and Euclidean distance is used as the evaluation criterion. The results indicate that AHP has greater validity, with a Euclidean distance of 21.2, compare to 25.9 to classical method. The Mann-Whitney U test (U=12034.5, p<0.05) statistically confirms this difference. The findings demonstrate that AHP, in comparison to the classical approach, more accurately identifies distinct risk categories (ranging from vulnerable to critical) and achieves a more balanced classification of vulnerability levels. Furthermore, it mitigates the tendency toward extreme assessments observed in classical method. This study confirms that weighting parameters within the AHP framework significantly enhances the precision of flood risk assessment.