The increasing complexity of smart industrial power systems necessitates the use of advanced data analytics to enable proactive and predictive maintenance. This paper presents a comprehensive review of advanced data analytics techniques that can be applied to smart power systems to improve maintenance and operational efficiency. We explore various machine learning techniques, including support vector machine, KNN, gradient boosting, gaussian naive bayes, and discuss their applicability to failure data analysis in power systems. Data-driven solutions offer a promising approach for predictive decision-making by leveraging fault analysis to identify potential issues before they occur. Finally, we discuss challenges associated with implementing advanced data analytics in a case study and suggest potential solutions depending on the failure data. Our findings suggest that the application of advanced data analytics techniques can greatly empower smart electrical systems when proper accuretness is identified depending on the nature of past events, enabling them to operate more efficiently and effectively.