Time-series clustering of global SDG index using DTW for policy insights toward 2030 agenda

The Sustainable Development Goals (SDGs) offer a global roadmap to address interconnected challenges such as poverty, inequality, and climate change by 2030. This study aims to develop a time-series clustering framework based on Dynamic Time Warping (DTW) to uncover global development patterns and group countries according to the similarity of their SDG Index trajectories from 2000 to 2023, covering around 167 countries and 15 groups. Three clustering algorithms, namely K-Means, Agglomerative, and Spectral, are evaluated using Silhouette Score, Calinski-Harabasz Index, and Davies-Bouldin Index to identify the most suitable method. Spectral Clustering outperforms the others, revealing four distinct global development trajectory clusters. These clusters reflect both economic and geographic patterns, which align with existing development theories: wealthier countries tend to achieve higher SDG scores, while countries nearby often share similar trajectories, suggesting regional policy influence and common challenges. The study demonstrates how time-series clustering can uncover actionable insights to support data-driven policy-making, promote regional alignment, and enable cross-country learning. By identifying shared development paths and challenges, the findings contribute to more effective and collaborative strategies for achieving the 2030 Agenda.