Predicting sustainable innovation performance of food MSMEs using SEM-ANN approach

Abstract

Sustainable innovation performance has become increasingly important for enhancing the competitiveness of micro, small, and medium-sized enterprises (MSMEs), particularly in emerging economies. However, evidence on the influence of strategic orientation dimensions remains mixed, and the moderating role of open innovation is still unclear.  In addition, limited research has combined explanatory and predictive approaches to examine both causal relationships and complex patterns. To address these gaps, this study integrates Partial Least Squares Structural Equation Modeling (PLS-SEM) and Artificial Neural Networks (ANN) using data from 183 food-sector MSMEs in Pekanbaru City. PLS-SEM examined causal relationships, whereas ANN captured nonlinear predictive patterns. The findings show that environmental, technological, entrepreneurial, and customer orientations positively influence sustainable innovation performance, with environmental orientation emerging as the strongest predictor, whereas learning orientation exhibits a negative, context-dependent relationship.  Open innovation does not significantly moderate these relationships, suggesting that internally aligned strategic capabilities may be more important than external collaboration in resource-constrained settings. This study clarifies the distinct roles of strategic orientation dimensions and the limited moderating role of open innovation in the MSME context. The findings offer practical insights for MSMEs seeking to strengthen innovation performance through strategically aligned internal capabilities.