Application of grey system-based benchmark models in marine industry forecasting

Purpose This study aims to systematically evaluate the forecasting performance of different grey system models in the marine economy. Its goal is to identify models that can accurately predict trends in key marine industries. The findings are expected to provide a valuable reference for assessing the applicability and reliability of grey models in supporting decision-making, policy formulation and sustainable development planning. Design/methodology/approach This study examines three key sectors of the marine economy, including marine fisheries, seawater utilization and marine transportation. Annual output data from 2006 to the most recent year are utilized. Several grey forecasting models with different structural formulations are applied, and their forecasting accuracy is evaluated using multiple performance metrics. Findings The fractional non-homogeneous discrete grey (FNDGM) model demonstrates the best overall forecasting performance across all three marine economy sectors. Overall, the results confirm that grey prediction models are particularly well suited to various sectors of the marine economy. Originality/value This study presents one of the few systematic comparisons of grey forecasting models in the marine economy. It extends the application of the FNDGM model to multiple marine sectors, demonstrating its robustness across marine economy datasets. The study highlights the strength of grey methods in handling small-sample data and provides practical insights for policymakers and industry planners to support sustainable marine development.