Humanitarian Relief Logistics Network Design Using Distributional Robust Optimization for Disaster Management

Le résumé fourni par la source

World’s growing population and the frequency of natural disasters as well as managing disasters and continuous improvements in methods and strategies adopted has become an essential global concern. The current paper introduces a two-stage mathematical model designed to minimize operational costs while improving service delivery through incorporating smart city infrastructure. In pre-disaster phase, multiple suppliers, warehouses, and regions are considered, along with such key objectives as allocating suppliers, locating warehouses, managing inventory in addition to identifying regions for smart city development. Post-disaster, the model focuses on routing drone flights for collecting data, distributing relief items, and establishing make-shift relief centers. The second stage comprises both ground and aerial vehicles for logistics and data collection. To handle uncertainty and the model's dual-level nature, a distributionally robust optimization (DRO) approach is exploited. Sensitivity analysis of a numerical example points to the fact that storage costs, demand correlation, and average demand significantly affect total costs. So much so that an increase in these parameters brings about an increase in total operation costs. First-stage decisions yielded 12 efficient solutions for the second-stage model. The obtained results indicate that for reducing shortages in the humanitarian logistics network, there should happen an increase in total costs. The increase should be directed towards using more drones for distributing aid items and gathering information. Given the NP-hard nature of the model, hybrid algorithms were employed, which outperformed exact methods in terms of efficiency. In a real-world case study in Isfahan Province, seven cities (Fereydunshahr, Kashan, Najafabad, Ardestan, Varzaneh, Isfahan, and Shahreza) were identified as smart city infrastructure sites. Five warehouses in Naeen, Kashan, Fereydan, Isfahan, and Shahreza were selected for disaster relief logistics operations. Findings highlight the trade-off between cost and service level, emphasizing the importance of drones in reducing shortages and enhancing disaster response efficiency. Managerial insights highlight the cost-effectiveness of IoT deployment in reducing demand uncertainty and enhancing response efficiency. The proposed framework offers a practical and scalable solution for disaster preparedness and post-disaster management in diverse urban contexts.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.