Purpose In the contemporary digital era, innovation represents a fundamental source of competitive advantage and sustainability within the manufacturing industry. Drawing on this premise, this study empirically examines how lean manufacturing (LM), Industry 4.0 Technologies (I4.0T), knowledge management (KM) and digital infrastructure (DI) influence manufacturing industry innovation performance (MIIP). It investigates the mediating roles of I4.0T and KM, and the moderating effect of DI, in elucidating the relationships among these constructs. Design/methodology/approach This study employs a mixed-methods, quantitative approach based on the resource-based view (RBV) and dynamic capabilities theory (DCT). Survey data from 263 respondents in Ethiopia's manufacturing sector were analyzed using partial least squares structural equation modeling (PLS-SEM) to test relationships, while fuzzy-set qualitative comparative analysis (fsQCA) identified configurations leading to high MIIP. Findings PLS-SEM reveals that LM has a significant influence on I4.0T and KM, which in turn enhance MIIP. While LM's direct link to MIIP is positive, it remains statistically insignificant. KM and I4.0T mediate between LM and MIIP. DI strengthens the LM to I4.0T relationship but does not significantly moderate LM to MIIP. FsQCA identifies KM as necessary for high MIIP and reveals several equifinal combinations of LM, I4.0T, KM and DI driving innovation performance. Originality/value This study advances RBV and DC Theory by developing and testing an integrated framework explaining how LM, I4.0T, KM and DI shape MIIP in emerging economy contexts. By combining PLS-SEM and fsQCA, the study captures net effects and configurational pathways through which operational, technological and knowledge-based capabilities drive innovation outcomes. The findings show that DI acts as a boundary condition shaping how lean and technological capabilities translate into innovation performance, revealing multiple equifinal pathways for achieving innovation success.
Zhang et al. (Tue,) studied this question.