Estimating flood losses plays a crucial role in aiding stakeholders to make informed decisions on flood risk management. While several flood damage models have been developed over the years to comprehensively estimate flood losses to residential structures, house contents have often been overlooked, despite their general importance on the overall damage figure. This thesis presents the development and implementation of the INSYDE-Content model for estimating flood damage to house contents. Modelling flood damage to house contents is a real challenge due to its heterogeneous nature as compared to the more standard geometry and features of the structure. Numerous factors come into play, including flood characteristics, building and contents vulnerability, and other social and economic factors. Eleven house contents comprising (beds, sofa, wardrobe, dining table setup, kitchen setup, washing machine, refrigerator, oven, TV, dishwasher, and microwave) have been considered as standard household components. Virtual data collection was employed to gather information on house characteristics and contents from various real estate websites. To ensure the data authenticity, the surveyed houses were identified and were cross-referenced with Google Maps and Street View. Out of about 500 houses checked sixty houses in Northern Italy were selected for a detailed analysis. Quantitative estimation of damage components was achieved through assumptions based on statistical inferences, literature review and expert considerations. All the necessary computations were done in R and Matlab. Flood depth and duration were considered as the driving factors in the damage mechanisms, and damage estimations were made on a component-by-component basis, incorporating thresholds derived from the literature and practical considerations. A sensitivity analysis was conducted to ascertain the influence of various factors on the total flood damage estimation. The model was validated using the 2010 flood loss data obtained from the municipality of Caldogno in the Veneto region of Northern Italy. The INSYDE-Content model showed promising performance by predicting approximately 67% of the observed damage, with RMS error of EUR 23,494.
La stima delle perdite da alluvione riveste un ruolo cruciale nell'aiutare le parti interessate a prendere decisioni informate sulla gestione del rischio di alluvione. Mentre diversi modelli di danno da alluvione sono stati sviluppati nel corso degli anni per stimare in modo esaustivo le perdite da alluvione alle strutture residenziali, i contenuti delle abitazioni sono stati spesso trascurati, nonostante la loro importanza generale sulla cifra complessiva dei danni. Questa tesi presenta lo sviluppo e l'implementazione del modello INSYDE-Content per la stima dei danni da alluvione al contenuto delle abitazioni. La modellazione dei danni da alluvione al contenuto di una casa è una vera sfida a causa della sua natura eterogenea rispetto alla geometria e alle caratteristiche più standard della struttura. Entrano in gioco numerosi fattori, tra cui le caratteristiche dell'alluvione, la vulnerabilità dell'edificio e del contenuto e altri fattori sociali ed economici. Sono stati considerati come componenti standard della casa undici oggetti (letti, divano, armadio, tavolo da pranzo, cucina, lavatrice, frigorifero, forno, TV, lavastoviglie e microonde). La raccolta virtuale dei dati è stata utilizzata per raccogliere informazioni sulle caratteristiche e sui contenuti della casa da vari siti web immobiliari. Per garantire l'autenticità dei dati, sono state identificate le case oggetto dell'indagine e sono state incrociate con Google Maps e Street View. Su circa 500 case controllate, sono state selezionate sessanta case nel Nord Italia per un'analisi dettagliata. La stima quantitativa delle componenti del danno è stata ottenuta attraverso ipotesi basate su inferenze statistiche, revisione della letteratura e considerazioni di esperti. Tutti i calcoli necessari sono stati eseguiti in R e Matlab. La profondità e la durata dell'inondazione sono state considerate i fattori trainanti dei meccanismi di danno e le stime dei danni sono state effettuate componente per componente, incorporando soglie derivate dalla letteratura e da considerazioni pratiche. È stata condotta un'analisi di sensibilità per verificare l'influenza di vari fattori sulla stima del danno totale da alluvione. Il modello è stato convalidato utilizzando i dati relativi alle perdite da alluvione del 2010 ottenuti dal comune di Caldogno, nella regione Veneto dell'Italia settentrionale. Il modello INSYDE-Content ha mostrato prestazioni promettenti, prevedendo circa il 67% dei danni osservati, con un errore RMS di 23.494 euro.
INSYDE-Content : synthetic flood damage model to house-content
Acharya, Pradeep
2022/2023
Abstract
Estimating flood losses plays a crucial role in aiding stakeholders to make informed decisions on flood risk management. While several flood damage models have been developed over the years to comprehensively estimate flood losses to residential structures, house contents have often been overlooked, despite their general importance on the overall damage figure. This thesis presents the development and implementation of the INSYDE-Content model for estimating flood damage to house contents. Modelling flood damage to house contents is a real challenge due to its heterogeneous nature as compared to the more standard geometry and features of the structure. Numerous factors come into play, including flood characteristics, building and contents vulnerability, and other social and economic factors. Eleven house contents comprising (beds, sofa, wardrobe, dining table setup, kitchen setup, washing machine, refrigerator, oven, TV, dishwasher, and microwave) have been considered as standard household components. Virtual data collection was employed to gather information on house characteristics and contents from various real estate websites. To ensure the data authenticity, the surveyed houses were identified and were cross-referenced with Google Maps and Street View. Out of about 500 houses checked sixty houses in Northern Italy were selected for a detailed analysis. Quantitative estimation of damage components was achieved through assumptions based on statistical inferences, literature review and expert considerations. All the necessary computations were done in R and Matlab. Flood depth and duration were considered as the driving factors in the damage mechanisms, and damage estimations were made on a component-by-component basis, incorporating thresholds derived from the literature and practical considerations. A sensitivity analysis was conducted to ascertain the influence of various factors on the total flood damage estimation. The model was validated using the 2010 flood loss data obtained from the municipality of Caldogno in the Veneto region of Northern Italy. The INSYDE-Content model showed promising performance by predicting approximately 67% of the observed damage, with RMS error of EUR 23,494.| File | Dimensione | Formato | |
|---|---|---|---|
|
INSYDE-CONTENT REPORT.pdf
accessibile in internet solo dagli utenti autorizzati
Dimensione
4.8 MB
Formato
Adobe PDF
|
4.8 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/208621