scholarly journals Multi-variate flood damage assessment: a tree-based data-mining approach

2013 ◽  
Vol 13 (1) ◽  
pp. 53-64 ◽  
Author(s):  
B. Merz ◽  
H. Kreibich ◽  
U. Lall
Keyword(s):  
Data Mining ◽  
Damage Assessment ◽  
Economic Status ◽  
Flood Damage ◽  
Damage Functions ◽  
Data Set ◽  
Promising Alternative ◽  
Damage Models ◽  
Inundation Depth ◽  
Flood Damage Assessment

Abstract. The usual approach for flood damage assessment consists of stage-damage functions which relate the relative or absolute damage for a certain class of objects to the inundation depth. Other characteristics of the flooding situation and of the flooded object are rarely taken into account, although flood damage is influenced by a variety of factors. We apply a group of data-mining techniques, known as tree-structured models, to flood damage assessment. A very comprehensive data set of more than 1000 records of direct building damage of private households in Germany is used. Each record contains details about a large variety of potential damage-influencing characteristics, such as hydrological and hydraulic aspects of the flooding situation, early warning and emergency measures undertaken, state of precaution of the household, building characteristics and socio-economic status of the household. Regression trees and bagging decision trees are used to select the more important damage-influencing variables and to derive multi-variate flood damage models. It is shown that these models outperform existing models, and that tree-structured models are a promising alternative to traditional damage models.

scholarly journals Data-mining for multi-variable flood damage modelling with limited data

2017 ◽  
Author(s):  
Dennis Wagenaar ◽  
Jurjen de Jong ◽  
Laurens M. Bouwer
Keyword(s):  
Data Mining ◽  
Bayesian Network ◽  
Water Depth ◽  
Damage Assessment ◽  
Damage Model ◽  
Residential Building ◽  
Regression Trees ◽  
Flood Damage ◽  
Data Mining Techniques ◽  
Flood Damage Assessment

Abstract. Flood damage assessment is usually done with damage curves only dependent on the water depth. Recent studies have shown that data-mining techniques applied to a multi-dimensional dataset can produce significantly better flood damage estimates. However, creating and applying a multi-variable flood damage model requires an extensive dataset, which is rarely available and this can limit the application of these new techniques. In this paper we enrich a dataset of residential building and content damages from the Meuse flood of 1993 in the Netherlands, to make it suitable for multi-variable flood damage assessment. Results from 2D flood simulations are used to add information on flow velocity, flood duration and the return period to the dataset, and cadastre data is used to add information on building characteristics. Next, several statistical approaches are used to create multi-variable flood damage models, including regression trees, bagging regression trees, random forest, and a Bayesian network. Validation on data points from a test set shows that the enriched dataset in combination with the data-mining techniques delivers a significant improvement over a simple model only based on the water depth. We find that with our dataset, the trees based methods perform better than the Bayesian Network.

scholarly journals Estimation uncertainty of direct monetary flood damage to buildings

2004 ◽  
Vol 4 (1) ◽  
pp. 153-163 ◽  
Author(s):  
B. Merz ◽  
H. Kreibich ◽  
A. Thieken ◽  
R. Schmidtke
Keyword(s):  
Water Depth ◽  
Flood Damage ◽  
Economic Losses ◽  
Damage Functions ◽  
Data Set ◽  
Large Variability ◽  
Flood Impacts ◽  
Inundation Depth ◽  
Damage Data ◽  
Damage Estimates

Abstract. Traditional flood design methods are increasingly supplemented or replaced by risk-oriented methods which are based on comprehensive risk analyses. Besides meteorological, hydrological and hydraulic investigations such analyses require the estimation of flood impacts. Flood impact assessments mainly focus on direct economic losses using damage functions which relate property damage to damage-causing factors. Although the flood damage of a building is influenced by many factors, usually only inundation depth and building use are considered as damage-causing factors. In this paper a data set of approximately 4000 damage records is analysed. Each record represents the direct monetary damage to an inundated building. The data set covers nine flood events in Germany from 1978 to 1994. It is shown that the damage data follow a Lognormal distribution with a large variability, even when stratified according to the building use and to water depth categories. Absolute depth-damage functions which relate the total damage to the water depth are not very helpful in explaining the variability of the damage data, because damage is determined by various parameters besides the water depth. Because of this limitation it has to be expected that flood damage assessments are associated with large uncertainties. It is shown that the uncertainty of damage estimates depends on the number of flooded buildings and on the distribution of building use within the flooded area. The results are exemplified by a damage assessment for a rural area in southwest Germany, for which damage estimates and uncertainty bounds are quantified for a 100-year flood event. The estimates are compared to reported flood damages of a severe flood in 1993. Given the enormous uncertainty of flood damage estimates the refinement of flood damage data collection and modelling are major issues for further empirical and methodological improvements.

scholarly journals Multi-variable flood damage modelling with limited data using supervised learning approaches

2017 ◽  
Vol 17 (9) ◽  
pp. 1683-1696 ◽  
Author(s):  
Dennis Wagenaar ◽  
Jurjen de Jong ◽  
Laurens M. Bouwer
Keyword(s):  
Bayesian Network ◽  
Supervised Learning ◽  
Water Depth ◽  
Damage Assessment ◽  
Regression Trees ◽  
Flood Damage ◽  
Data Set ◽  
Widespread Application ◽  
Learning Techniques ◽  
Flood Damage Assessment

Abstract. Flood damage assessment is usually done with damage curves only dependent on the water depth. Several recent studies have shown that supervised learning techniques applied to a multi-variable data set can produce significantly better flood damage estimates. However, creating and applying a multi-variable flood damage model requires an extensive data set, which is rarely available, and this is currently holding back the widespread application of these techniques. In this paper we enrich a data set of residential building and contents damage from the Meuse flood of 1993 in the Netherlands, to make it suitable for multi-variable flood damage assessment. Results from 2-D flood simulations are used to add information on flow velocity, flood duration and the return period to the data set, and cadastre data are used to add information on building characteristics. Next, several statistical approaches are used to create multi-variable flood damage models, including regression trees, bagging regression trees, random forest, and a Bayesian network. Validation on data points from a test set shows that the enriched data set in combination with the supervised learning techniques delivers a 20 % reduction in the mean absolute error, compared to a simple model only based on the water depth, despite several limitations of the enriched data set. We find that with our data set, the tree-based methods perform better than the Bayesian network.

scholarly journals INSYDE: a synthetic, probabilistic flood damage model based on explicit cost analysis

2016 ◽  
Author(s):  
Francesco Dottori ◽  
Rui Figueiredo ◽  
Mario Martina ◽  
Daniela Molinari ◽  
Anna Rita Scorzini
Keyword(s):  
Damage Assessment ◽  
Damage Model ◽  
Flood Damage ◽  
Physical Damage ◽  
Damage Functions ◽  
Technical Literature ◽  
Actual Knowledge ◽  
Damage Models ◽  
Model Based ◽  
Risk Assessment And Management

Abstract. Methodologies to estimate economic flood damages are increasingly important for flood risk assessment and management. In this work, we present a new synthetic flood damage model based on a component-by-component analysis of physical damage to buildings. The damage functions are designed using an expert-based approach with the support of existing scientific and technical literature, and have been calibrated with loss adjustment studies and damage surveys carried out for past flood events in Italy. The model structure is designed to be transparent and flexible, and therefore it can be applied in different geographical contexts and adapted to the actual knowledge of hazard and vulnerability variables. The model has been tested in a recent flood event in Northern Italy. Validation results provided good estimates of post-event damages, with better performances than most damage models available in the literature. In addition, a local sensitivity analysis has been performed, in order to identify the hazard variables that have more influence on damage assessment results.

scholarly journals INSYDE: a synthetic, probabilistic flood damage model based on explicit cost analysis

2016 ◽  
Vol 16 (12) ◽  
pp. 2577-2591 ◽  
Author(s):  
Francesco Dottori ◽  
Rui Figueiredo ◽  
Mario L. V. Martina ◽  
Daniela Molinari ◽  
Anna Rita Scorzini
Keyword(s):  
Damage Assessment ◽  
Damage Model ◽  
Flood Damage ◽  
Physical Damage ◽  
Damage Functions ◽  
Technical Literature ◽  
Actual Knowledge ◽  
Damage Models ◽  
Model Based ◽  
Risk Assessment And Management

Abstract. Methodologies to estimate economic flood damages are increasingly important for flood risk assessment and management. In this work, we present a new synthetic flood damage model based on a component-by-component analysis of physical damage to buildings. The damage functions are designed using an expert-based approach with the support of existing scientific and technical literature, loss adjustment studies, and damage surveys carried out for past flood events in Italy. The model structure is designed to be transparent and flexible, and therefore it can be applied in different geographical contexts and adapted to the actual knowledge of hazard and vulnerability variables. The model has been tested in a recent flood event in northern Italy. Validation results provided good estimates of post-event damages, with similar or superior performances when compared with other damage models available in the literature. In addition, a local sensitivity analysis was performed in order to identify the hazard variables that have more influence on damage assessment results.

On the modeling of significance for flood damage assessment

2014 ◽  
Vol 10 ◽  
pp. 381-391 ◽  
Author(s):  
D. Molinari ◽  
F. Ballio ◽  
J. Handmer ◽  
S. Menoni
Keyword(s):  
Damage Assessment ◽  
Flood Damage ◽  
Flood Damage Assessment

scholarly journals Ex post damage assessment: an Italian experience

2014 ◽  
Vol 14 (4) ◽  
pp. 901-916 ◽  
Author(s):  
D. Molinari ◽  
S. Menoni ◽  
G. T. Aronica ◽  
F. Ballio ◽  
N. Berni ◽  
...  
Keyword(s):  
Data Collection ◽  
Damage Assessment ◽  
Local Level ◽  
Data Gathering ◽  
Recovery Phase ◽  
Flood Damage ◽  
Local Authorities ◽  
Damage Models ◽  
Ex Post ◽  
And Storage

Abstract. In recent years, awareness of a need for more effective disaster data collection, storage, and sharing of analyses has developed in many parts of the world. In line with this advance, Italian local authorities have expressed the need for enhanced methods and procedures for post-event damage assessment in order to obtain data that can serve numerous purposes: to create a reliable and consistent database on the basis of which damage models can be defined or validated; and to supply a comprehensive scenario of flooding impacts according to which priorities can be identified during the emergency and recovery phase, and the compensation due to citizens from insurers or local authorities can be established. This paper studies this context, and describes ongoing activities in the Umbria and Sicily regions of Italy intended to identifying new tools and procedures for flood damage data surveys and storage in the aftermath of floods. In the first part of the paper, the current procedures for data gathering in Italy are analysed. The analysis shows that the available knowledge does not enable the definition or validation of damage curves, as information is poor, fragmented, and inconsistent. A new procedure for data collection and storage is therefore proposed. The entire analysis was carried out at a local level for the residential and commercial sectors only. The objective of the next steps for the research in the short term will be (i) to extend the procedure to other types of damage, and (ii) to make the procedure operational with the Italian Civil Protection system. The long-term aim is to develop specific depth–damage curves for Italian contexts.

scholarly journals Brief communication: simple-INSYDE, development of a new tool for flood damage evaluation from an existing synthetic model

2020 ◽  
Author(s):  
Marta Galliani ◽  
Daniela Molinari ◽  
Francesco Ballio
Keyword(s):  
Damage Assessment ◽  
Direct Analysis ◽  
Flood Damage ◽  
Damage Evaluation ◽  
Explanatory Variables ◽  
Synthetic Model ◽  
Flood Damage Assessment ◽  
Correct Implementation

Abstract. INSYDE is a multi-variable, synthetic model for flood damage assessment to dwellings. The analysis and use of this model highlighted some weaknesses, linked to its complexity, that can undermine its usability and correct implementation. This study proposes a simplified version of INSYDE which maintains its multi-variable and synthetic nature, but has simpler mathematical formulations permitting an easier use and a direct analysis of the relation between damage and its explanatory variables.

Sign in / Sign up

Export Citation Format

Share Document