Integrating logistics planning and population sampling decisions for humanitarian needs assessment

Abstract

After a disaster, humanitarian organizations must quickly assess the needs of affected populations spread across a large area to inform effective relief distribution. This process involves site visits and collecting information through interviews, where balancing travel time and assessment accuracy is critical. This paper introduces the Assessment Routing Problem with Coverage and Accuracy, which addresses the logistical planning and sampling decisions of a post-disaster needs assessment team. The problem involves determining which sites to visit, in what sequence, and how much time to spend at each site to collect information within a total assessment time limit. We present a mathematical model that balances site coverage and sampling accuracy, and propose a constructive heuristic and a tabu search algorithm to efficiently solve large-scale instances. Computational experiments demonstrate that the tabu search algorithm can quickly obtain optimal or near-optimal solutions for large instances, while the performance of the heuristic is influenced by network topology. We illustrate the effectiveness of our approach through a case study based on the 2011 Van Earthquake in Turkey. Our findings emphasize the trade-offs involved in prioritizing certain areas and the advantages of leveraging reliable prior information in the assessment process to achieve better site coverage and sampling accuracy.