
Food and Energy Security, Год журнала: 2025, Номер 14(3)
Опубликована: Май 1, 2025
ABSTRACT Grain supply and demand affect regional food security; however, the drivers are often unclear, making precise forecasting policymaking challenging. This study used Central Asia as a case to integrate Partial Least Squares Structural Equation Modeling (PLS‐SEM) with particle swarm optimization least squares support vector machine (PSO‐LSSVM) separately identify of grain enhance prediction accuracy. We analyzed interannual variations in production, import/export volumes, consumption, inventory wheat, rice, barley, maize, other grains (1992–2019). then decoupled factors affecting wheat production consumption using PLS‐SEM made predictions by integrating PSO‐LSSVM. The results showed that across Asia, primarily driven declined later recovered, turning point between 1995 1998. Kazakhstan exports 44% its whereas countries heavily depend on imports. In path coefficients ( r ) area yield total were 0.36 0.77, respectively, Kazakhstan, they 0.37 0.81, respectively. Climate cultivation indirectly through yield, influence area. Economic growth increased urban population decreased it. reduced = −0.23) but boosted economy 0.33), pattern was not observed Asia. coupling model PSO‐LSSVM enhanced accuracy reducing error 10.21% 32.8% Kazakhstan. offers novel approach decouple driving predicts crop yields regions limited data availability.
Язык: Английский