With the increasing complexity of construction projects, enhancing decision-making and managing operational risks have become significant challenges for enterprises. This paper designs and implements an engineering decision support system based on data analysis and machine learning, aimed at providing scientific decision analysis support to enterprises. The system integrates knowledge from engineering management with advanced data mining techniques to construct a series of decision models for cost prediction, schedule control, and quality analysis. These models are efficiently deployed using microservices architecture and offer interactive visual decision interfaces. Comprehensive testing demonstrates the system’s excellent functionality, high concurrency processing capabilities, and scalability. In practical applications, the system significantly improves decision quality, reduces operational costs, promotes the transformation of decision-making from empirical to data-driven and model-driven approaches, thereby contributing robust technical support to the high-quality development of the construction industry.