Implementation Of Building Seismic Loss Prediction In The Business Interruption Loss Framework

An earthquake is a natural disaster that can have a huge impact on human society.The energy and seismic waves released during an earthquake will move in the ground and have the potential to spread to the communities.Then,many commercial and civil buildings will be severely affected as a result,and the urban economy will also be hit unexpectedly.Commercial buildings and urban infrastructure are the foundation of social and economic development.Therefore,the prediction of building losses,recovery time,and economic shock in the event of an earthquake is very important for economic reconstruction after a disaster.Physical asset loss estimation can also help the government and businesses predict economic losses,calculate the time for building reconstruction,and even formulate more effective risk aversion plans.A business interruption loss（BIL）framework has been introduced to tackle the above problem.However,this BIL framework requires prediction methods for its inputs that have not yet been identified.The goal of this thesis is to implement a prediction method for the building losses and recovery time variables of the BIL framework to estimate BILs.Sensitivity analyses are also performed on the results to test the significance of different inputs in the prediction of building losses and recovery time and business interruption losses.Regional Workflow for Hazard and Loss Estimation（rWHALE）is the main experimental framework to specifically predict and analyze building loss and recovery time.The paper further performs a sensitivity analysis to study the correlation between the accuracy of predicting building loss results and the accuracy of the predicted business interruption loss results.This paper uses the 2008 Wenchuan earthquake as an actual case analysis to verify the credibility of the experimental data.That is,by inputting the 2008 Wenchuan local building data to rWHALE framework to predict the buildings losses and recovery time,including manufacturing,retailing,and residential buildings,and then feeding the results to the BIL framework.The research goal of this paper is achieved through the implementation and analysis of the BIL framework.Although the accuracy of the prediction results is not very high,the prediction results are understandable based on the existing limitations in this thesis.The major limitation is that the data collected for prediction is not accurate in the first place.This thesis also explains in detail the other reasons for the low prediction accuracy.The main source of data for this study is questionnaire surveys,and most of the data collected are based on the memory of the respondents and limited public resources.Even though the predictions of building loss and recovery time were not accurate,the final forecast output of the BIL framework showed higher accuracy.This thesis suggests that building losses are not a dominant impact for the manufacturing industry.Large influencing factors,other data analysis,such as labor or equipment,included in the framework of BIL and reduction,may have a greater impact on the manufacturing industry than building losses.Further analysis of the BIL model will require more accurate data and a better understanding of the influence of the other input variables of the BIL model.