THE ROLE OF PREDICTIVE ANALYTICS IN ENHANCING AGRIBUSINESS SUPPLY CHAINS

Abstract

This study investigated the role of predictive analytics in enhancing agribusiness supply chains using a quantitative, explanatory design that tested direct, mediated, and moderated relationships among predictive analytics capability, supply chain integration, and supply chain performance. This study empirically tests supply chain integration as a mediating mechanism linking predictive analytics capability to agribusiness supply chain performance. Integration—operationalized as multi-tier visibility and planning synchronization—is hypothesized to transmit the benefits of predictive insights into coordinated action. Using structural equation modeling on data from 240 firms, results confirm strong mediation: predictive analytics significantly increases integration (β = 0.64), and integration subsequently enhances performance (β = 0.45). The indirect effect accounts for a major share of the total relationship. Findings indicate that predictive analytics improves performance not solely through technical accuracy gains but through organizational processes that align expectations, reduce information asymmetry, and stabilize decision cycles across farms, aggregators, processors, logistics providers, and retailers. This study develops and tests a comprehensive framework linking predictive analytics capability, supply chain integration, contextual moderators, and agribusiness performance. It conceptualizes predictive analytics as a quantifiable capability embedded across demand forecasting, yield estimation, quality prediction, logistics risk assessment, and cold-chain monitoring. Using cross-sectional data and structural modeling, the study demonstrates that predictive analytics improves performance directly and indirectly by enhancing integration. Moderators—including perishability, digital maturity, market volatility, and chain length—shape the magnitude of these relationships. The resulting model offers a validated structure for explaining how prediction-based capabilities generate measurable efficiency, reliability, and waste-reduction gains across complex agribusiness networks.