Uncertainty is inherent in engineering decision-making, still statistical thinking remains underutilized across many domains. Traditional deterministic models and heuristics often fail to capture variability, limiting effective risk management. This study examines the role of statistical thinking in managing engineering uncertainty, identifying barriers and opportunities for its broader application. A scoping review using Elicit and Scopus retrieved 928 documents from the past 15 years. Bibliometric analysis with VOSviewer identified five key research clusters: decision-making and optimization, risk assessment, machine learning, information management, and sustainability modeling. Findings reveal that machine learning and AI are increasingly integrated into decision-making, yet gaps remain in their adoption for uncertainty quantification. Furthermore, Bayesian inference, probabilistic modeling, and experimental design remain underutilized in critical engineering applications. This study emphasizes the growing intersection between statistical thinking, AI, and sustainability, highlighting the need for greater statistical literacy and methodological integration. Future research should focus on bridging deterministic and probabilistic approaches to enhance decision accuracy and risk mitigation.