Leveraging Statistical Thinking for Digital Innovation: Reframing Uncertainty in Engineering Decision-Making

INTRODUCTION: Contemporary engineering operates in a data-rich yet uncertainty-laden landscape, particularly under the technological shifts introduced by Industry 4.0. While foundational, deterministic models frequently fail to address the ambiguity, variability, and incompleteness inherent in real-world data, this paper examines the growing need to embed statistical reasoning within digital engineering decision-making processes to ensure robustness and interpretability. OBJECTIVES: The study aims to investigate how statistical thinking contributes to innovation, transparency, and adaptive decision-making in digitalized engineering systems. It identifies conceptual gaps and underexplored themes in current literature and emphasizes the strategic relevance of probabilistic reasoning in addressing uncertainty across complex industrial settings. METHODS: A hybrid scoping review methodology was applied, combining a semantic AI-driven search via Elicit with a structured bibliographic query in Scopus. The resulting corpus of 928 curated publications was analysed through bibliometric techniques and social network analysis using VOSviewer. This comprehensive process enabled the identification of co-occurrence patterns, thematic clusters, and evolving disciplinary linkages, ensuring the credibility and reliability of the findings. RESULTS: Five primary research clusters emerged: decision optimization, risk management and human factors, machine learning integration, digital information systems, and sustainability. These clusters represent key areas where probabilistic modelling and uncertainty quantification can significantly enhance engineering practices. Although AI and big data analytics are increasingly prevalent, the underrepresentation of probabilistic modelling and uncertainty quantification in these clusters reveals a disconnect between data-centric innovation and risk-aware engineering practice. CONCLUSION: A conceptual shift toward probabilistic reasoning is advocated as a necessary response to the complexity of modern digital engineering environments. Repositioning statistical thinking as a central enabler of digital transformation supports the development of resilient, interpretable, and future-ready engineering systems. Integrating these methodologies into engineering curricula, AI pipelines, and industrial decision-support infrastructures is essential for advancing strategic, uncertainty-aware innovation.