世界银行-全球灾后快速损失估算评估综述_2015-2024年发展中国家灾后快速损害估算的前沿（英）-2025_90页_8mb

Summary of GRADE Assessments (2015-2024)

The Global Rapid Post-Disaster Damage Estimation (GRADE) is a remote, desk-based methodology developed by the World Bank to estimate direct economic damages to housing, non-residential buildings, infrastructure, and agriculture rapidly after disasters. Launched in 2015, GRADE has been deployed 66 times by November 2024, covering 62 events in 54 countries, including earthquakes, floods, tropical cyclones, volcanic eruptions, and human-induced disasters.

From 2015 to 2024, GRADE assessments were completed in an average of 2.6 weeks, making them significantly faster than traditional post-disaster needs assessments (PDNAs) or damage and loss assessments (DaLAs), which took an average of 14.7 weeks. GRADE results are typically within 88–90% accuracy of these assessments when compared sectorally. GRADE estimates direct economic damages, while PDNAs and DaLAs often include losses and broader needs assessments.

GRADE uses innovative techniques like remote sensing, drone footage, social media feeds, and civil engineering analyses. Case studies highlight its effectiveness in events like the 2023 Türkiye earthquakes, where it informed USD 1.74 billion in rapid recovery funding, and the 2019 Cyclone Idai in Mozambique, where damages were estimated at USD 715 million.

Several challenges persist, including data limitations, methodological differences, and uncertainties due to evolving disasters. GRADE’s future focuses on expanding capacity, leveraging technologies like machine learning, and enhancing partnerships to meet growing demand. Opportunities include integrating artificial intelligence, improving datasets, and addressing vulnerabilities like fragility, conflict, and gender impacts.

Overall, GRADE provides timely and reliable damage estimates, supporting faster and more informed post-disaster recovery efforts, though continuous refinement is needed to maintain accuracy and adapt to new disaster scenarios.