TY  - JOUR
AU  - Stojadinović, Zoran I.
AU  - Kovačević, Miloš
AU  - Marinković, Dejan
AU  - Stojadinović, Božidar
PY  - 2021
UR  - https://grafar.grf.bg.ac.rs/handle/123456789/2396
AB  - This paper proposes a new framework for rapid earthquake loss assessment based on a machine learning damage classification model and a representative sampling algorithm. A Random Forest classification model predicts a damage probability distribution that, combined with an expert-defined repair cost matrix, enables the calculation of the expected repair costs for each building and, in aggregate, of direct losses in the earthquake-affected area. The proposed building representation does not include explicit information about the earthquake and the soil type. Instead, such information is implicitly contained in the spatial distribution of damage. To capture this distribution, a sampling algorithm, based on K-means clustering, is used to select a minimal number of buildings that represent the area of interest in terms of its seismic risk, independently of future earthquakes. To observe damage states in the representative set after an earthquake, the proposed framework utilizes a local network of trained damage assessors. The model is updated after each damage observation cycle, thus increasing the accuracy of the current loss assessment. The proposed framework is exemplified using the 2010 Kraljevo, Serbia earthquake dataset.
T2  - Earthquake Spectra
T1  - Rapid earthquake loss assessment based on machine learning and representative sampling
EP  - 26
SP  - 1
DO  - 10.1177/87552930211042393
ER  - 

@article{
author = "Stojadinović, Zoran I. and Kovačević, Miloš and Marinković, Dejan and Stojadinović, Božidar",
year = "2021",
abstract = "This paper proposes a new framework for rapid earthquake loss assessment based on a machine learning damage classification model and a representative sampling algorithm. A Random Forest classification model predicts a damage probability distribution that, combined with an expert-defined repair cost matrix, enables the calculation of the expected repair costs for each building and, in aggregate, of direct losses in the earthquake-affected area. The proposed building representation does not include explicit information about the earthquake and the soil type. Instead, such information is implicitly contained in the spatial distribution of damage. To capture this distribution, a sampling algorithm, based on K-means clustering, is used to select a minimal number of buildings that represent the area of interest in terms of its seismic risk, independently of future earthquakes. To observe damage states in the representative set after an earthquake, the proposed framework utilizes a local network of trained damage assessors. The model is updated after each damage observation cycle, thus increasing the accuracy of the current loss assessment. The proposed framework is exemplified using the 2010 Kraljevo, Serbia earthquake dataset.",
journal = "Earthquake Spectra",
title = "Rapid earthquake loss assessment based on machine learning and representative sampling",
pages = "26-1",
doi = "10.1177/87552930211042393"
}

Stojadinović, Z. I., Kovačević, M., Marinković, D.,& Stojadinović, B.. (2021). Rapid earthquake loss assessment based on machine learning and representative sampling. in Earthquake Spectra, 1-26.
https://doi.org/10.1177/87552930211042393

Stojadinović ZI, Kovačević M, Marinković D, Stojadinović B. Rapid earthquake loss assessment based on machine learning and representative sampling. in Earthquake Spectra. 2021;:1-26.
doi:10.1177/87552930211042393 .

Stojadinović, Zoran I., Kovačević, Miloš, Marinković, Dejan, Stojadinović, Božidar, "Rapid earthquake loss assessment based on machine learning and representative sampling" in Earthquake Spectra (2021):1-26,
https://doi.org/10.1177/87552930211042393 . .

TY  - CONF
AU  - Kovačević, Miloš
AU  - Stojadinović, Zoran I.
AU  - Marinković, Dejan
AU  - Stojadinović, Božidar
PY  - 2018
UR  - https://grafar.grf.bg.ac.rs/handle/123456789/2176
AB  - Rapid earthquake damage and loss assessment is crucial both for insuring the safety of inhabitants in the immediate aftermath of an earthquake and for the recovery of the stricken communities in the long run. This paper investigates the potential of different machine learning methods for building a rapid earthquake loss assessment system intended for residential houses in a municipal area. The system is trained on a pre-earthquake selected representative set of residential houses, after observing their damage and loss states. Two representative sampling strategies and three machine learning algorithms are described and evaluated on the 2010 Kraljevo M5.4 earthquake data set. The proposed models showed satisfactory accuracy in predicting the total expected repair cost (less than 20% error with the representative sample size of 10% of the inventory). The approach is independent of geological and earthquake data and does not require local peak ground acceleration values.
C3  - 11th National Conference on Earthquake Engineering
T1  - Sampling and Machine Learning Methods for a Rapid Earthquake Loss Assessment System
UR  - https://hdl.handle.net/21.15107/rcub_grafar_2176
ER  - 

@conference{
author = "Kovačević, Miloš and Stojadinović, Zoran I. and Marinković, Dejan and Stojadinović, Božidar",
year = "2018",
abstract = "Rapid earthquake damage and loss assessment is crucial both for insuring the safety of inhabitants in the immediate aftermath of an earthquake and for the recovery of the stricken communities in the long run. This paper investigates the potential of different machine learning methods for building a rapid earthquake loss assessment system intended for residential houses in a municipal area. The system is trained on a pre-earthquake selected representative set of residential houses, after observing their damage and loss states. Two representative sampling strategies and three machine learning algorithms are described and evaluated on the 2010 Kraljevo M5.4 earthquake data set. The proposed models showed satisfactory accuracy in predicting the total expected repair cost (less than 20% error with the representative sample size of 10% of the inventory). The approach is independent of geological and earthquake data and does not require local peak ground acceleration values.",
journal = "11th National Conference on Earthquake Engineering",
title = "Sampling and Machine Learning Methods for a Rapid Earthquake Loss Assessment System",
url = "https://hdl.handle.net/21.15107/rcub_grafar_2176"
}

Kovačević, M., Stojadinović, Z. I., Marinković, D.,& Stojadinović, B.. (2018). Sampling and Machine Learning Methods for a Rapid Earthquake Loss Assessment System. in 11th National Conference on Earthquake Engineering.
https://hdl.handle.net/21.15107/rcub_grafar_2176

Kovačević M, Stojadinović ZI, Marinković D, Stojadinović B. Sampling and Machine Learning Methods for a Rapid Earthquake Loss Assessment System. in 11th National Conference on Earthquake Engineering. 2018;.
https://hdl.handle.net/21.15107/rcub_grafar_2176 .

Kovačević, Miloš, Stojadinović, Zoran I., Marinković, Dejan, Stojadinović, Božidar, "Sampling and Machine Learning Methods for a Rapid Earthquake Loss Assessment System" in 11th National Conference on Earthquake Engineering (2018),
https://hdl.handle.net/21.15107/rcub_grafar_2176 .

TY  - CONF
AU  - Marinković, Dejan
AU  - Stojadinović, Zoran I.
AU  - Kovačević, Miloš
AU  - Stojadinović, Božidar
PY  - 2018
UR  - https://grafar.grf.bg.ac.rs/handle/123456789/2175
AB  - Earthquake resilience starts with a sudden drop of performance when an earthquake strikes followed by a relatively long recovery phase. The dynamics and volume of investment directly affect the rate and level of recovery. In this paper, the recovery process of the city of Kraljevo, Serbia following the M5.4 November 3, 2010 Kraljevo earthquake is analyzed. The base for this analysis is recorded reconstruction data that includes building types, damage states, damage survey dates, repair methods, and repair design, permit, and completion dates. Tracking the rate of building re-occupation and the rate of investment in repairs during the recovery process made it possible to construct empirical housing resilience curves for a community of about 70,000 people. The recorded data also provides insights into the reasons for different recovery rates for different building types and damages states, as well as a basis to evaluate the housing recovery management process and identify the significant influence of the reconstruction funding volume and rate on the sequencing, design and construction of the post-earthquake repairs. A comparison between the housing recovery after the 2010 Kraljevo and the 2009 Yunnan earthquakes shows significant simulates, pointing to the need to further improve, optimize and standardize post-earthquake housing recovery strategies worldwide.
C3  - 16th European Conference on Earthquake Engineering
T1  - 2010 Kraljevo Earthquake Recovery Process Metrics Derived from Recorded Reconstruction Data
UR  - https://hdl.handle.net/21.15107/rcub_grafar_2175
ER  - 

@conference{
author = "Marinković, Dejan and Stojadinović, Zoran I. and Kovačević, Miloš and Stojadinović, Božidar",
year = "2018",
abstract = "Earthquake resilience starts with a sudden drop of performance when an earthquake strikes followed by a relatively long recovery phase. The dynamics and volume of investment directly affect the rate and level of recovery. In this paper, the recovery process of the city of Kraljevo, Serbia following the M5.4 November 3, 2010 Kraljevo earthquake is analyzed. The base for this analysis is recorded reconstruction data that includes building types, damage states, damage survey dates, repair methods, and repair design, permit, and completion dates. Tracking the rate of building re-occupation and the rate of investment in repairs during the recovery process made it possible to construct empirical housing resilience curves for a community of about 70,000 people. The recorded data also provides insights into the reasons for different recovery rates for different building types and damages states, as well as a basis to evaluate the housing recovery management process and identify the significant influence of the reconstruction funding volume and rate on the sequencing, design and construction of the post-earthquake repairs. A comparison between the housing recovery after the 2010 Kraljevo and the 2009 Yunnan earthquakes shows significant simulates, pointing to the need to further improve, optimize and standardize post-earthquake housing recovery strategies worldwide.",
journal = "16th European Conference on Earthquake Engineering",
title = "2010 Kraljevo Earthquake Recovery Process Metrics Derived from Recorded Reconstruction Data",
url = "https://hdl.handle.net/21.15107/rcub_grafar_2175"
}

Marinković, D., Stojadinović, Z. I., Kovačević, M.,& Stojadinović, B.. (2018). 2010 Kraljevo Earthquake Recovery Process Metrics Derived from Recorded Reconstruction Data. in 16th European Conference on Earthquake Engineering.
https://hdl.handle.net/21.15107/rcub_grafar_2175

Marinković D, Stojadinović ZI, Kovačević M, Stojadinović B. 2010 Kraljevo Earthquake Recovery Process Metrics Derived from Recorded Reconstruction Data. in 16th European Conference on Earthquake Engineering. 2018;.
https://hdl.handle.net/21.15107/rcub_grafar_2175 .

Marinković, Dejan, Stojadinović, Zoran I., Kovačević, Miloš, Stojadinović, Božidar, "2010 Kraljevo Earthquake Recovery Process Metrics Derived from Recorded Reconstruction Data" in 16th European Conference on Earthquake Engineering (2018),
https://hdl.handle.net/21.15107/rcub_grafar_2175 .

TY  - CONF
AU  - Stojadinović, Zoran I.
AU  - Kovačević, Miloš
AU  - Marinković, Dejan
AU  - Stojadinović, Božidar
PY  - 2017
UR  - https://grafar.grf.bg.ac.rs/handle/123456789/2174
AB  - This paper presents an earthquake damage and repair cost prediction framework for individual residential buildings and portfolios of residential buildings in a municipal area in a region where the seismological networks are sparse and the structural engineering data on the existing residential building stock is poor. 
The proposed data driven framework is based on the damage and reconstruction data from an actual earthquake, in this case, the M5.4 November 3, 2010 Kraljevo, Serbia, earthquake. It belongs to a more general class of hybrid building portfolio vulnerability models. The earthquake in the model is defined by its magnitude and epicenter location. The geographical distribution of the intensity of the earthquake at the location of the buildings is modeled using the 2013 Akkar-Sandikkaya-Bommer ground motion prediction model suitable for seismically active crustal regions in Europe, with the peak ground acceleration as the intensity measure. The data on the soil type distribution was collected form the municipality building department sources. The residential building stock was classified into six types by identifying typical architecture layouts, structural systems and elements. The residential building damage was surveyed after the 2010 Kraljevo earthquake by local engineers using a locally-developed survey form. The form contained the information about the individual damage, classified into four categories ranging from slight damage to collapse, varying amount of building-specific details, and addresses from which geographic locations of the buildings were derived. 
A random forest machine-learning algorithm was used to derive a predictive model for residential building portfolio seismic damage and repair cost using a portion of the 2010 Kraljevo data as the learning dataset. The model outputs both the individual building fragility and the aggregate portfolio-level vulnerability data. The calculation of the expected repair cost for each building type was done using an expert-defined matrix that specifies average repair costs for each building type and damage category. The model is verified on a separate test portion of the 2010 Kraljevo dataset, yielding a satisfactory relative error when comparing total predicted to total actual repair costs. 
The model is limited to regions with similar seismicity and similar building stock. However, there many regions in the Balkans that fit this constraint. The proposed framework is, however, more general. It can be applied to other regions with different seismicity and building stock using the data from a recent earthquake as its learning input dataset and an expert-defined repair cost matrix for analyzing the repair cost scenarios.
C3  - 16th World Conference on Earthquake Engineering (16WCEE)
T1  - Data-Driven Housing Damage and Repair Cost Prediction Framework Based on The 2010 Kraljevo Earthquake Data
UR  - https://hdl.handle.net/21.15107/rcub_grafar_2174
ER  - 

@conference{
author = "Stojadinović, Zoran I. and Kovačević, Miloš and Marinković, Dejan and Stojadinović, Božidar",
year = "2017",
abstract = "This paper presents an earthquake damage and repair cost prediction framework for individual residential buildings and portfolios of residential buildings in a municipal area in a region where the seismological networks are sparse and the structural engineering data on the existing residential building stock is poor. 
The proposed data driven framework is based on the damage and reconstruction data from an actual earthquake, in this case, the M5.4 November 3, 2010 Kraljevo, Serbia, earthquake. It belongs to a more general class of hybrid building portfolio vulnerability models. The earthquake in the model is defined by its magnitude and epicenter location. The geographical distribution of the intensity of the earthquake at the location of the buildings is modeled using the 2013 Akkar-Sandikkaya-Bommer ground motion prediction model suitable for seismically active crustal regions in Europe, with the peak ground acceleration as the intensity measure. The data on the soil type distribution was collected form the municipality building department sources. The residential building stock was classified into six types by identifying typical architecture layouts, structural systems and elements. The residential building damage was surveyed after the 2010 Kraljevo earthquake by local engineers using a locally-developed survey form. The form contained the information about the individual damage, classified into four categories ranging from slight damage to collapse, varying amount of building-specific details, and addresses from which geographic locations of the buildings were derived. 
A random forest machine-learning algorithm was used to derive a predictive model for residential building portfolio seismic damage and repair cost using a portion of the 2010 Kraljevo data as the learning dataset. The model outputs both the individual building fragility and the aggregate portfolio-level vulnerability data. The calculation of the expected repair cost for each building type was done using an expert-defined matrix that specifies average repair costs for each building type and damage category. The model is verified on a separate test portion of the 2010 Kraljevo dataset, yielding a satisfactory relative error when comparing total predicted to total actual repair costs. 
The model is limited to regions with similar seismicity and similar building stock. However, there many regions in the Balkans that fit this constraint. The proposed framework is, however, more general. It can be applied to other regions with different seismicity and building stock using the data from a recent earthquake as its learning input dataset and an expert-defined repair cost matrix for analyzing the repair cost scenarios.",
journal = "16th World Conference on Earthquake Engineering (16WCEE)",
title = "Data-Driven Housing Damage and Repair Cost Prediction Framework Based on The 2010 Kraljevo Earthquake Data",
url = "https://hdl.handle.net/21.15107/rcub_grafar_2174"
}

Stojadinović, Z. I., Kovačević, M., Marinković, D.,& Stojadinović, B.. (2017). Data-Driven Housing Damage and Repair Cost Prediction Framework Based on The 2010 Kraljevo Earthquake Data. in 16th World Conference on Earthquake Engineering (16WCEE).
https://hdl.handle.net/21.15107/rcub_grafar_2174

Stojadinović ZI, Kovačević M, Marinković D, Stojadinović B. Data-Driven Housing Damage and Repair Cost Prediction Framework Based on The 2010 Kraljevo Earthquake Data. in 16th World Conference on Earthquake Engineering (16WCEE). 2017;.
https://hdl.handle.net/21.15107/rcub_grafar_2174 .

Stojadinović, Zoran I., Kovačević, Miloš, Marinković, Dejan, Stojadinović, Božidar, "Data-Driven Housing Damage and Repair Cost Prediction Framework Based on The 2010 Kraljevo Earthquake Data" in 16th World Conference on Earthquake Engineering (16WCEE) (2017),
https://hdl.handle.net/21.15107/rcub_grafar_2174 .