From calibrated risk to actionable response: A prescriptive analytics pipeline for urban flash floods

Abstract

Urban flash-flood management requires turning probabilistic forecasts into concrete, resource-constrained actions. We develop a forecast-to-decision pipeline that links calibrated incident-level harm prediction with prescriptive optimization. Using a processed DesignSafe-CI Texas flash-flood incident dataset covering 2005–2019 (6,137 incidents; 2.17% harm prevalence), we construct GIS-derived predictors of hazard intensity, exposure, and transportation infrastructure, plus physically motivated interactions (e.g., precipitation × bridge density). An 𝓁 ₁-regularized logistic model (LASSO) with out-of-fold isotonic calibration provides decision-grade probabilities and automatic feature selection. Temporal generalization is assessed with a strict year holdout (train 2005-2018; test 2019), leave-one-year-out (LOYO), and rolling-origin schemes. On the 2019 holdout, the calibrated model attains ROC-AUC 0.783, PR-AUC 0.127 (about six times prevalence), and Brier 0.021, although the holdout contains only five positive cases and therefore yields wide uncertainty intervals; complementary LOYO validation gives AUC values spanning 0.66–0.99, reflecting interannual variability that random splits obscure. We embed the learned calibrator as a piecewise-linear mapping in a mixed-integer program that selects intervention sets meeting risk-coverage targets at minimum effort. Results show strong spatial concentration of expected harm within the candidate set of historically active locations: intervening at 61 of 220 candidate assets (28%) captures 80% of baseline risk; 98 candidate assets (45%) achieve 90%, with certified optimality gaps < 0.1%. Leave-one-year-out frontier analysis confirms this concentration pattern is stable across years (median: 27% of assets for 80% coverage; IQR < 2 percentage points), indicating that the prescriptive conclusions are not artifacts of the single holdout year. The framework is reproducible, interpretable, and operationally tractable, providing a transparent pathway from geomatics-informed risk to prioritized preparedness and response portfolios for known high-risk locations.

Keywords

Urban flooding / Flood risk prediction / Prescriptive optimization / Rare-event modeling / Probability calibration

References

[1] Hirabayashi Y, Mahendran R, Koirala S, Konoshima L, Yamazaki D, Watanabe S, et al. Global flood risk under climate change. Nat Clim Chang 2013; 3(9): 816-21.

[2] Winsemius HC, Aerts JC, Van Beek L PH, Bierkens M FP, Bouwman A, Jongman B, et al. Global drivers of future river flood risk. Nat Clim Chang 2016; 6(4): 381-5.

[3] Capshaw KM, Padgett JE. A global analysis of coastal flood risk to the petrochemical distribution network in a changing climate. Resilient Citi Struct 2022; 1(3): 52-60.

[4] Singh P, Amekudzi-Kennedy A, Ashuri B, Parrillo T, Rizzi D, Clark R, et al. Case study of flood risk and vulnerability in the city of atlanta-a social, economic, technical, and institutional perspective. Resilient Citi Struct 2025; 4(2): 1-13.

[5] Jonkman SN, Kelman I. An analysis of the causes and circumstances of flood disaster deaths. Disasters 2005; 29(1): 75-97.

[6] Ashley ST, Ashley WS. Flood fatalities in the united states. J Appl Meteorol Climatol 2008; 47(3): 805-18.

[7] Service NW. Weather fatalities 2019. Tech. Rep.; NOAA National Weather Service; 2019. https://www.weather.gov/hazstat/.

[8] Terti G, Ruin I, Anquetin S, Gourley JJ. A situation-based analysis of flash flood fatalities in the united states. Bull Am Meteorol Soc 2017; 98(2): 333-45.

[9] Diakakis M, Boufidis N, Grau J MS, Andreadakis E, Stamos I. A systematic assessment of the effects of extreme flash floods on transportation infrastructure and circulation: the example of the 2017 Mandra flood. Int J Disast Risk Reduct 2020; 47: 101542.

[10] Zhao B, Tang Y, Wang C, Zhang S, Soga K. Evaluating the flooding level impacts on urban metro networks and travel demand: behavioral analyses, agent-based simulation, and large-scale case study. Resilient Citi Struct 2022; 1(3): 12-23.

[11] Berghuijs WR, Aalbers EE, Larsen JR, Trancoso R, Woods RA. Recent changes in extreme floods across multiple continents. Environ Res Lett 2017; 12(11): 114035.

[12] Teng J, Jakeman AJ, Vaze J, Croke B FW, Dutta D, Kim S. Flood inundation modelling: a review of methods, recent advances and uncertainty analysis. Environ Modell Softw 2017; 90: 201-16.

[13] Jamali B, Löwe R, Bach PM, Urich C, Arnbjerg-Nielsen K, Deletic A. A rapid urban flood inundation and damage assessment model. J Hydrol (Amst) 2018; 564: 1085-98.

[14] Nandi A, Mandal A, Wilson M, Smith D. Flood hazard mapping in Jamaica using principal component analysis and logistic regression. Environ Earth Sci 2016; 75(6): 465.

[15] Jonkman SN, Vrijling JK, Vrouwenvelder A. Methods for the estimation of loss of life due to floods: a literature review and a proposal for a new method. Nat Hazard 2008; 46(3): 353-89.

[16] Sabeti R, Stamataki I, Kjeldsen TR. Reconstructing the 1968 river chew flash flood: merging a HEC-RAS 2d hydraulic modelling approach with historical evidence. Geomat Nat Hazard Risk 2024; 15(1): 2377655.

[17] Hu H, Yu M, Zhang X, Wang Y. Urban hydrological model (UHM) developed for an urban flash flood simulation and analysis of the flood intensity sensitivity to urbanization. Geomat Nat Hazard Risk 2024; 15(1): 2302561.

[18] Dwivedi SK, Thakur PK, Dhote PR, Kruczkiewicz A, Upadhyay M, Moothedan AJ, et al. Unravelling flash flood dynamics of song watershed, doon valley: key insights for floodplain management. Geomat Nat Hazard Risk 2024; 15(1): 2378979.

[19] Omidian P, Khaji N, Aghakouchak AA. An integrated decision-making approach to resilience-LCC bridge network retrofitting using a genetic algorithm-based framework. Resilient Citi Struct 2025; 4(1): 16-40.

[20] Tehrany MS, Pradhan B, Jebur MN. Flood susceptibility mapping using a novel ensemble weights-of-evidence and support vector machine models in GIS. J Hydrol (Amst) 2014; 512: 332-43.

[21] Khosravi K, Nohani E, Maroufinia E, Pourghasemi HR. A GIS-based flood susceptibility assessment and its mapping in Iran: a comparison between frequency ratio and weights-of-evidence bivariate statistical models with multi-criteria decision-making technique. Nat Hazard 2016; 83(2): 947-87.

[22] Mosavi A, Ozturk P, Chau K-w. Flood prediction using machine learning models: literature review. Water (Basel) 2018; 10(11): 1536.

[23] Fang Z, Wang Y, Peng L, Hong H. Integration of convolutional neural network and conventional machine learning classifiers for landslide susceptibility mapping. Comput Geosci 2020; 139: 104470.

[24] Zhao G, Pang B, Xu Z, Peng D, Zuo D. Urban flood susceptibility assessment based on convolutional neural networks. J Hydrol (Amst) 2020; 590: 125235.

[25] Göçer M, Erdoğan H, Öztürk B, Co¸skun SB. Seismic damage assessment of RC structures after the 2015 Gorkha, Nepal, earthquake using gradient boosting classifiers. Buildings 2025; 15(19): 3577.

[26] Baltacıoğlu AK, Öztürk B, Civalek Ö, Akgöz B. Is artificial neural network suitable for damage level determination of RC-structures? Int J Eng Appl Sci 2010; 2(3): 71-81.

[27] Henderson K, Chakrabortty RK. A machine learning predictive model for bushfire ignition and severity: the study of Australian black summer bushfires. Decisi Analyt J 2025; 14: 100529.

[28] Islam MA, Rashid SI, Hossain N UI, Fleming R, Sokolov A. An integrated convolutional neural network and sorting algorithm for image classification for efficient flood disaster management. Decisi Analyt J 2023; 7: 100225.

[29] Bates PD, Quinn N, Sampson C, Smith A, Wing O, Sosa J, et al. Combined modeling of US fluvial, pluvial, and coastal flood hazard under current and future climates. Water Resour Res 2021; 57(2): e2020WR028673.

[30] ¸ Salap-Ayça S, Jankowski P, Clarke KC, Nara A. Is less more? Experimenting with visual stacking of coincident maps for spatial global sensitivity analysis in urban land-use change modeling. Environ Modell Softw 2021; 145: 105181.

[31] Syifa M, Park SJ, Achmad AR, Lee C-W, Eom J. Flood mapping using remote sensing imagery and artificial intelligence techniques: a case study in Brumadinho, Brazil. J Coast Res 2019; 90(SI): 197-204.

[32] Halder K, Ghosh A, Srivastava AK, Pal SC, Chatterjee U, Bisai D, et al. Sar-driven flood inventory and multi-factor ensemble susceptibility modelling using machine learning frameworks. Geomat Nat Hazard Risk 2024; 15(1): 2409202.

[33] Costache R, Arabameri A, Elkhrachy I, Ghorbanzadeh O, Pham QB. Detection of areas prone to flood risk using state-of-the-art machine learning models. Geomat Nat Hazard Risk 2021; 12(1): 1488-507.

[34] Rahman ZU, Zhang M, Chen F, Ullah S, Ahmad M, Feroz A, et al. Flood susceptibility mapping using supervised machine learning models: insights into predictors’ significance and models’ performance. Geomat Nat Hazard Risk 2025; 16(1): 2516728.

[35] Radhwan AS, Al-Areeq AM, Al Aghbari AA, Ghaleb M, Benaafi M, Al-Areeq NM, et al. A novel voting ensemble model empowered by metaheuristic feature selection for accurate flash flood susceptibility mapping. Geomat Nat Hazard Risk 2024; 15(1): 2360000.

[36] Zhong M, Jiang T, Hong Y, Yang X. Performance of multi-level association rule mining for the relationship between causal factor patterns and flash flood magnitudes in a humid area. Geomat Nat Hazard Risk 2019; 10(1): 1967-87.

[37] Merz B, Kuhlicke C, Kunz M, Pittore M, Babeyko A, Bresch DN, et al. Impact forecasting to support emergency management of natural hazards. Rev Geophys 2020; 58(4): e2020RG000704.

[38] Arrighi C, Pregnolato M, Castelli F. Indirect flood impacts and cascade risk across interdependent linear infrastructures. Nat Hazard Earth Syst Sci 2021; 21(6): 1955-69.

[39] He J, Zhang L, Xiao T, Chen C. Emergency risk management for landslide dam breaks in 2018 on the Yangtze river, China. Resilient Citi Struct 2022; 1(3): 1-11.

[40] Bermúdez M, Cea L, Puertas J. A rapid flood inundation model for hazard mapping based on least squares support vector machine regression. J Flood Risk Manage 2019; 12: e12522.

[41] Pregnolato M, Ford A, Wilkinson SM, Dawson RJ. The impact of flooding on road transport: a depth-disruption function. Transport Res Part D: Transp Environ 2017; 55: 67-81.

[42] Chang S, Wilkho RS, Gharaibeh N, Sansom G, Meyer M, Olivera F, et al. Environmental, climatic, and situational factors influencing the probability of fatality or injury occurrence in flash flooding: a rare event logistic regression predictive model. Nat Hazard 2023; 116(3): 3957-78.

[43] Koks EE, Jongman B, Husby TG, Botzen W JW. Combining hazard, exposure and social vulnerability to provide lessons for flood risk management. Environ Sci Policy 2015; 47: 42-52.

[44] Wing O EJ, Lehman W, Bates PD, Sampson CC, Quinn N, Smith AM, et al. Inequitable patterns of US flood risk in the anthropocene. Nat Clim Chang 2022; 12(2): 156-62.

[45] Roohi M, Ghasemi S, Sediek O, Jeon H, van de Lindt JW, Shields M, et al. Multi-disciplinary seismic resilience modeling for developing mitigation policies and recovery planning. Resilient Citi Struct 2024; 3(2): 66-84.

[46] Boeing G. Osmnx: new methods for acquiring, constructing, analyzing, and visualizing complex street networks. Comput Environ Urban Syst 2017; 65: 126-39.

[47] Biljecki F, Chow YS. Global building morphology indicators. Comput Environ Urban Syst 2022; 95: 101809.

[48] Fleischmann M, Feliciotti A, Romice O, Porta S. Morphological tessellation as a way of partitioning space: improving consistency in urban morphology at the plot scale. Comput Environ Urban Syst 2020; 80: 101441.

[49] Ito K, Zhu Y, Abdelrahman M, Liang X, Fan Z, Hou Y, et al. ZenSVI: an open-source software for the integrated acquisition, processing and analysis of street view imagery towards scalable urban science. Comput Environ Urban Syst 2025; 119: 102283.

[50] Félix R, Moura F, Lovelace R. Reproducible methods for modeling combined public transport and cycling trips and associated benefits: evidence from the biclar tool. Comput Environ Urban Syst 2025; 117: 102230.

[51] Boonmee C, Arimura M, Asada T. Facility location optimization model for emergency humanitarian logistics. Int J Disaster Risk Reduct 2017; 24: 485-98.

[52] Duhamel C, Santos AC, Brasil D, Châtelet É, Birregah B. Connecting a population dynamic model with a multi-period location-allocation problem for post-disaster relief operations. Ann Oper Res 2016; 247(2): 693-713.

[53] Bayram V, Tansel BÇ, Yaman H. Compromising system and user interests in shelter location and evacuation planning. Transport Res Part B: Methodol 2015; 72: 146-63.

[54] Mazzoleni M, Alfonso L, Chacon-Hurtado J, Solomatine D. Assimilating uncertain, dynamic and intermittent streamflow observations in hydrological models. Adv Water Resour 2015; 83: 323-39.

[55] van de Lindt JW, Kruse J, Cox DT, Gardoni P, Lee JS, Padgett J, et al. The interdependent networked community resilience modeling environment (IN-CORE). Resilient Citi Struct 2023; 2(2): 57-66.

[56] Karatas M, Yakıcı E. A multi-objective location analytics model for temporary emergency service center location decisions in disasters. Decisi Analy J 2021; 1: 100004.

[57] Mishra BK, Dahal K, Pervez Z, Bhattarai S. A multi-objective evolutionary optimisation model for heterogeneous vehicles routing and relief items scheduling in humanitarian crises. Decisi Analy J 2022; 5: 100128.

[58] Zuhanda MK, Anggraini S, Puspita D, et al. A stochastic decision model for emergency logistics using hybrid metaheuristic methods. Decisi Analy J 2025;: 100649.

[59] Roushan A, Das A, Dutta A, Bera UK. A pentagonal type-2 fuzzy variable defuzzification model with application in humanitarian supply chains. Decisi Analy J 2023; 8: 100303.

[60] Sumrit D, Jongprasittiphol O. An integrated multi-stage decision model for upstream supply chain disaster management readiness assessment. Decisi Analy J 2025; 14: 100538.

[61] Yazdi J, Salehi Neyshabouri S, Niksokhan MH, Sheshangosht S, Elmi M. Optimal prioritisation of watershed management measures for flood risk mitigation on a watershed scale. J Flood Risk Manage 2013; 6(4): 372-84.

[62] Maier HR, Kapelan Z, Kasprzyk J, Kollat J, Matott LS, Cunha MC, et al. Evolutionary algorithms and other metaheuristics in water resources: current status, research challenges and future directions. Environ Modell Softw 2014; 62: 271-99.

[63] Fletcher TD, Shuster W, Hunt WF, Ashley R, Butler D, Arthur S, et al. Suds, lid, bmps, wsud and more-the evolution and application of terminology surrounding urban drainage. Urban Water J 2015; 12(7): 525-42.

[64] Mugume SN, Gomez DE, Fu G, Farmani R, Butler D. A global analysis approach for investigating structural resilience in urban drainage systems. Water Res 2015; 81: 15-26.

[65] Boulange J, Hanasaki N, Yamazaki D, Pokhrel Y. Role of dams in reducing global flood exposure under climate change. Nat Commun 2021; 12(1): 417.

[66] Yang L, Jones BF, Yang S-H. A fuzzy multi-objective programming for optimization of fire station locations through genetic algorithms. Eur J Oper Res 2007; 181(2): 903-15.

[67] Roberts DR, Bahn V, Ciuti S, Boyce MS, Elith J, Guillera-Arroita G, et al. Cross-validation strategies for data with temporal, spatial, hierarchical, or phylogenetic structure. Ecography 2017; 40(8): 913-29.

[68] Meyer H, Reudenbach C, Hengl T, Katurji M, Nauss T. Improving performance of spatio-temporal machine learning models using forward feature selection and target-oriented validation. Environ Modell Softw 2018; 101: 1-9.

[69] Bergmeir C, Hyndman RJ, Koo B. A note on the validity of cross-validation for evaluating autoregressive time series prediction. Comput Statist Data Analy 2018; 120: 70-83.

[70] Cerqueira V, Torgo L, Mozetiˇc I. Evaluating time series forecasting models: an empirical study on performance estimation methods. Mach Learn 2020; 109(11): 1997-2028.

[71] Diakakis M, Deligiannakis G. Vehicle-related flood fatalities in Greece. Environ Hazard 2013; 12(3-4): 278-90.

[72] Chawla NV, Japkowicz N, Kotcz A. Special issue on learning from imbalanced data sets. ACM SIGKDD Explorat Newsl 2004; 6(1): 1-6.

[73] He H, Garcia EA. Learning from imbalanced data. IEEE Trans Knowl Data Eng 2009; 21(9): 1263-84.

[74] Saito T, Rehmsmeier M. The precision-recall plot is more informative than the ROC plot when evaluating binary classifiers on imbalanced datasets. PLoS ONE 2015; 10(3): e0118432.

[75] Ozenne B, Subtil F, Maucort-Boulch D. The precision-recall curve overcame the optimism of the receiver operating characteristic curve in rare diseases. J Clin Epidemiol 2015; 68(8): 855-9.

[76] Gneiting T, Balabdaoui F, Raftery AE. Probabilistic forecasts, calibration and sharpness. J Roy Statist Soc Ser B: Statistical Methodology 2007; 69(2): 243-68.

[77] Wilks DS. Statistical methods in the atmospheric sciences; vol. 100. Academic press; 2011.

[78] Nearing GS, Kratzert F, Sampson AK, Pelissier CS, Klotz D, Frame JM, et al. What role does hydrological science play in the age of machine learning? Water Resour Res 2021; 57(3): e2020WR028091.

[79] Kratzert F, Klotz D, Shalev G, Klambauer G, Hochreiter S, Nearing G. Towards learning universal, regional, and local hydrological behaviors via machine learning applied to large-sample datasets. Hydrol Earth Syst Sci 2019; 23(12): 5089-110.

[80] Lundberg SM, Lee S-I. A unified approach to interpreting model predictions. Adv Neural Inf Process Syst 2017; 30.

[81] Friedman JH. Greedy function approximation: a gradient boosting machine. Ann Stat 2001;: 1189-232.

[82] Mukherjee F, Singh D. Detecting flood prone areas in harris county: a GIS based analysis. GeoJournal 2020; 85(3): 647-63.

[83] Zadrozny B, Elkan C. Transforming classifier scores into accurate multiclass probability estimates. In: Proceedings of the eighth ACM SIGKDD international conference on knowledge discovery and data mining. 2002, p. 694-9.

[84] Niculescu-Mizil A, Caruana R. Predicting good probabilities with supervised learning. In: Proceedings of the 22nd international conference on machine learning. 2005, p. 625-32.

[85] Vielma JP, Ahmed S, Nemhauser G. Mixed-integer models for nonseparable piecewise-linear optimization: unifying framework and extensions. Oper Res 2010; 58(2): 303-15.

[86] Gharaibeh N, Meyer M, Zou L, Olivera F, Sansom G, Chang S, et al. Reducing the human impacts of flash floods: development of microdata and causal model to inform mitigation and preparedness. 2023. https://doi.org/10.17603/ds2-e91y-cv92

[87] Watanabe K, Ishiwatari M, Sasaki D, Sakamoto A, Nakayama M. Investing in resilience: a long-term analysis of China’s flood protection strategies. Resilient Citi Struct 2025; 4(1): 115-23.

[88] Braik AM, Gupta HS, Koliou M, González AD. Multi-hazard probabilistic risk assessment and equitable multi-objective optimization of building retrofit strategies in hurricane-vulnerable communities. Comput-Aid Civ Infrastruct Eng 2025; 40(15): 2074-97.

[89] Gupta HS, Adluri T, Sanderson D, González AD, Nicholson CD, Cox D. Multi-objective optimization of mitigation strategies for buildings subject to multiple hazards. Int J Disast Risk Reduct 2024; 100: 104125.

[90] Rudin C. Stop explaining black box machine learning models for high stakes decisions and use interpretable models instead. Nat Mach Intell 2019; 1(5): 206-15.

[91] Gupta HS, Nofal OM, González AD, Nicholson CD, van de Lindt JW. Optimal selection of short-and long-term mitigation strategies for buildings within communities under flooding hazard. Sustainability 2022; 14(16): 9812.

[92] Nofal OM, Amini K, Padgett JE, van de Lindt JW, Rosenheim N, Darestani YM, et al. Multi-hazard socio-physical resilience assessment of hurricane-induced hazards on coastal communities. Resilient Citi Struct 2023; 2(2): 67-81.

[93] Mazumder RK, Enderami SA, Rosenheim N, Sutley EJ, Stanley M, Meyer M. Estimating long-term k-12 student homelessness after a catastrophic flood disaster. Resilient Citi Struct 2023; 2(2): 82-92.