Globalization has improved the efficiency of supply chains but has also introduced vulnerabilities to unforeseen disruptions such as natural disasters or pandemics. While prior studies have investigated supply chain resilience through digital transformation and Industry 4.0 technologies, challenges persist—particularly when key suppliers take long time to recover or reluctant to share data due to privacy concerns. Cloud manufacturing, with its enhanced visibility and resource sharing capabilities, presents a promising new avenue for addressing disruptions through the dynamic coordination of the available manufacturing resources within the production network. Building on this potential, this study proposes a bi-level supply chain model that integrates cloud manufacturing to enhance supply chain resilience. To ensure effective coordination while safeguarding sensitive information, the model is divided into network-level and node-level models, supported by a novel three-level data privacy classification that guides secure data sharing. At the network level, we utilize Non-dominated Sorting Genetic Algorithm II (NSGA-II) and Mixed-Integer Linear Programming (MILP) to balance the objectives of lead time, transportation cost, and capacity redundancy, thereby ensuring timely order delivery. At the node level, mixed-integer programming is employed to while ensuring suppliers can meet order requirements via dynamic resource allocations. Additionally, we build mathematical models for evaluating and optimizing resilience across both the network and node levels. Through case studies examining both base (15 orders) and expanded (30 orders) scenarios across 6 manufacturing service providers, we demonstrate how the model effectively balances system resilience with operational efficiency. The results reveal trade-offs between system resilience and performance metrics such as total lead time and transportation cost. The bi-level coordination approach maintains operational efficiency while protecting supplier privacy, with only safe-to-share and aggregated data utilized at the network level. These findings demonstrate that cloud manufacturing offers a promising approach for enhancing supply chain resilience, though careful balance between redundancy and operational efficiency is crucial.