Author(s)

J. Vargas Florez, M. Lauras, L. Dupont & A. Charles

Abstract

Humanitarian supply chains have received a lot of attention over the last fifteen years, and can now be considered a new research area. But a gap exists between the research work proposals and their applications in the field. One of the main issues is that the demand, in the case of disaster, is hard to assess because of the high-level of uncertainty. Gathering knowledge about future demand is of prime importance to be able to propose models, which are relevant to implement for a real problem. This paper tackles this problematic proposing a four-step methodology for forecast disaster impact, and in this way, the future demand, such as cyclones in the Caribbean or earthquakes along the Pacific Ring of Fire. This approach uses data analysis techniques such as Principal Component Analysis and Multivariate Regression Analysis. An application case on Peruvian earthquake demand is proposed to illustrate the benefits of our approach. Keywords: forecast, disaster, demand, principal component analysis, multivariate regression analysis.

Keywords

forecast, disaster, demand, principal component analysis,multivariate regression analysis