Firm dynamics in the adoption of artificial intelligence in plant maintenance practices within the manufacturing sector in Gauteng

Abstract

This study investigated the impact of firm dynamics on the adoption of artificial intelligence (AI) and its use in maintenance and aftermarket support in the manufacturing industry in Gauteng, South Africa. Based on conceptualizations of technology adoption and the Diffusion of Innovations (DOI) theory and dynamic capabilities theory (DCT), this study aimed to understand how both technological perceptions and organizational capabilities influence AI adoption. A quantitative, cross-sectional research design was utilized using structured questionnaires distributed to maintenance and engineering professionals across manufacturing firms. Quantitative data from 113 valid responses was gathered and analysed using SPSS 30, and structural equation modelling using partially least squares (PLS-SEM) was used to produce the statistical evaluation of the relationships among the constructs. The results of the study demonstrated that while technological perceptions such as relative advantage and complexity were influential to awareness of and attitudes towards AI, they were not significant predictors of actual adoption behaviour. Instead, organizational capabilities and environmental capabilities were identified as the strongest determinants of adoption. In particular, environmental capabilities and absorptive capacity had the most significant positive effect, followed by organizational agility and sustainable competitive advantage. These findings suggest that the driving factors of AI adoption in manufacturing contexts of emerging markets depend less on perceived usefulness and more on the firm's ability to learn how to readapt and reconfigure resources in dynamic environments. From a theoretical contribution, this study makes an integration of the DOI and DCT with a unified framework describing AI adoption as a capability rather than a purely perception/attributes process. Practical contributions show the potential in using PLS-SEM for statistical evaluations of complex organizational constructs in resource scarce contexts, and provides practical insights for managers and policymakers, that can be acted on to improve industrial competitiveness in maintaining organizations through the adoption of AI-enabled maintenance. The study suggests that proactive actions such as investment in digital skills development, knowledge-sharing partnerships and organizational agility, leveraging on absorptive capacity in ways that foster sustainable technology-adoption.