Track: Decision Sciences
Abstract
The biggest challenge in earthquake emergency logistics lies in determining the demand for emergency logistics support. To forecast the need for emergency logistics support plays a vital role in optimal disaster logistics management. An accurate demand forecasting can prevent an out-of-stock, can save time, and ensure a proper allocation of emergency logistical relief to overcome the long-suffering of victims. This paper aims to design a model for estimating emergency logistical assistance requests after an earthquake. The methodology of Case-based Reasoning (CBR) is applied to build the model. At the same time, the implementation of the Internet of Things (IoT) able to supports retrieving data to the model to produce the forecasting results quickly. The research results show that the error forecast for relief logistics includes blankets, tents, food are respectively 16.78%, 15.99%, and 10.48%. All errors forecast in the range of 10% -20%; thus, the results indicate that the forecast output model is valid to use for predicting emergency logistical assistance requests immediately after an earthquake occurs.