Estimated probability of the number of buildings damaged by the floods and the cost of repairs

Abstract

Flood disasters often cause buildings damaged, for repairs required considerable cost. This paper analyzes estimates of the probability and the number of buildings damaged by the Citarum River flood in the framework of cost recovery planning. The probability estimation of building damage was carried out using a logistic regression model with a genetic algorithm approach. The number of buildings damaged and risk of losses are estimated using the principle of expected value. The analysis shows that the probability estimator of the building damage significantly follows the logistic regression model. Meanwhile, the estimated the building damage is 349, with a loss of IDR67.80 billion. So the related parties (government, financial institutions and homeowners) should reserve the cost of repair at least worth the risk of loss, to face future flood disasters.

Keywords: Citarum River; logistic regression; genetic algorithm; losses risk; expected value