Incomplete information imputation in limited data environments with application to disaster response

3Citations
Citations of this article
49Readers
Mendeley users who have this article in their library.

Abstract

Following a major disaster, a field operations manager needs to deploy relief activities within the affected region. State-of-the-art humanitarian logistics models have been developed over the past decades to assist relief operations. However, while many models assume availability of information on infrastructure status, this is typically not the case in practice. Often, only partial information about infrastructure status is known. Utilizing the similarities in the known data via attributes, we develop a framework to impute incomplete information in limited data environments. We present an application of this framework to a past disaster, the 2010 Haiti earthquake. We build an ArcGIS model to automate the data collection and processing efforts to the extent possible. The study explores the impact of missing data, dispersion of missing data and imputation techniques used in approximating the incomplete information. Our results suggest that lower granularity yields better estimates of the unknown information above a threshold. We also develop publicly available test cases for the broader community.

Cite

CITATION STYLE

APA

Yagci Sokat, K., Dolinskaya, I. S., Smilowitz, K., & Bank, R. (2018). Incomplete information imputation in limited data environments with application to disaster response. European Journal of Operational Research, 269(2), 466–485. https://doi.org/10.1016/j.ejor.2018.02.016

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free