The DoD is interested in delivering systems such as missile defense systems and services to the warfighter at a reasonable cost. To achieve this, it would be helpful to collect sufficient knowledge on supply bases, analyze and predict risks in supply chains, and take risk mitigation approaches to improve supply chains. However, current methods of data collection and analytics do not efficiently and reliably predict risk, or mitigate it effectively to improve supply chains. To address this critical need, IAI and collaborators at the University of Tennessee, Knoxville will design and implement the Deep Machine learning system for risk Analysis and Prediction (D-MAP) in supply chains. This system will include some novel components, including a data management module that automatically collects raw data from multiple data sources, improves data quality, and merges and processes raw data to obtain both features and ground truth to facilitate further analysis. It will have a data analytics module that analyzes the supply chain data via advanced machine learning techniques to predict future risks. Further, a risk identification module will identify risks based on mission requirements and discover new risks from data analysis results. Finally, it will employ a novel risk mitigation module that recommends approaches to mitigate risks to improve the supply chain and enable it to meet its mission requirements. The D-MAP software system can be directly applied to improve reliability and efficiency of MDA and other military applications with complex supply chains. Using big data and advanced data analysis, D-MAP can also help to reduce risk and improve management in commercial supply chains.