Datasets and Scripts for Reproducing Results in“UAV-Guided Bi-level Robust Optimization for Multi-hazard Emergency Response: An Exact-Matheuristic Collaborative Approach in Contin-uous-Discrete Nested Spaces”

Description

The dataset combines real-world topological network data with simulated mul-ti-hazard emergency scenarios (earthquakes, forest fires, and urban waterlogging). It provides the foundational spatial and computational data required for modeling du-al-layer emergency response systems, which encompass ground rescue routing and unmanned aerial vehicle (UAV) auxiliary reconnaissance path planning. The geographic environments and disaster simulations are designed to evaluate facility site-selection, resource allocation, and routing efficiency under varying levels of hazard intensity. The foundational geospatial and road network data are extracted via OpenStreetMap (OSM) to construct the abstract and concrete sub-systems for the routing models. The dataset incorporates unified structural formulas with disaster-specific coefficients to represent economic costs across different hazard scenarios without requiring disparate cost equations. The dataset includes: • Geospatial and Network Data (city.oms,forest.oms,N30E104.hgt): Geographic coordinates of demand nodes, candidate facility sites, and edge topologies extracted from OSM. These files define the physical constraints and connectivity for both ground vehicles and UAV reconnaissance paths. • Multi-Hazard Scenario Parameters (Disaster_Environment): Simulation data arrays representing earthquake, forest fire, and urban wa-terlogging events. This includes demand distribution, varying intensity levels, and the unified structural cost coefficients specific to each disaster type. • Map and Topology Visualization (Net-work_Abstract_Concrete.png): Visual representations of the abstract and concrete sub-systems serving as spatial references for the experiments. • Implementation Scripts (main_optimization.m): Core MATLAB scripts and Gurobi interface configurations for reproducing the algorithmic testing, dual-layer routing, and cost evaluation procedures. The raw geographic data is sourced from OpenStreetMap and is subject to the Open Database License (ODbL). All generated simulation parameters, matrix configurations, and algorithmic codes are provided to ensure full computational reproducibility. This dataset supports research on multi-hazard emergency management, dual-layer routing optimization (UAV and ground), data-driven facility location-allocation, and the evaluation of heuristic optimization algorithms in complex disaster environments.

Institutions

• Nanchang Hangkong University

  Jiangxi, Nanchang

Categories

Licence