Use of SEM-PLS analysis to predict sports injuries in professional football players through warehouse technology data.

Sports injuries (SP-INJ) substantially affect athlete health and team performance, underscoring the need for precise risk assessment in professional football. This study develops and tests an interpretable partial least squares structural equation modelling (PLS-SEM) injury-risk model that integrates the following constructs: body mass index (BMI), mechanical load (ML), training-program structure (TP), context-specific risk factors (RF), warehouse-technology effectiveness (WT), which is supported by Electronic Athlete Information System (EAIS), musculoskeletal development (MD), performance, and SP-INJ. The model integrates continuous WT data with self-reported/contextual measures in a unified framework, demonstrating a scalable, warehouse-backed pipeline for team-level risk stratification. Data were collected from 153 professional football players across multiple leagues worldwide. WT captured workload and physiological signals to derive ML, while questionnaires obtained BMI, TP, RF, MD, Performance, and SP-INJ outcomes. PLS-SEM (two-stage: measurement → structural) was used to test the hypothesised pathways among these constructs. Higher ML and elevated BMI were positively associated with SP-INJ incidence. Well-structured TP and effective WT were negatively associated with injuries. MD and performance mediated the impact of ML and RF. The final model explained 61% of the variance in SP-INJ. Integrating ML metrics with EAIS-supported WT and modelling them within a PLS-SEM framework enables interpretable team-level risk stratification and can inform periodised, targeted prevention strategies in professional football. The cross-sectional design, reliance on BMI as a body-composition proxy, modest sample size, and football-specific data may limit generalisability; future longitudinal, multi-sport validation is warranted.