Reliable and safe critical infrastructures are crucial for the sustainability of modern societies. To cope with increasing disruptive events such as man-made and natural disasters attacking infrastructures, resilience should be considered as an integrated perspective into the system planning process. This paper presents a p-robust optimization model for infrastructure network planning against spatially localized disruptions. The optimization aims at minimizing the investment costs for system hardening and expansion and the total system costs under nominal operating conditions, while incorporating resilience requirements by the p-robustness constraints. Importantly, instead of only mitigating system vulnerability , the proposed model integrates the arranging of the repair sequence of damaged components under limited repair resources into the preevent system planning. The complexity of the proposed model is analyzed, and a hybrid algorithm that combines scenario-based decomposition and variable neighborhood search is developed for its efficient solution. The effectiveness of the approach is illustrated through an application to a real power transmission system. Quantitative analysis can assist managers in decision making regarding investing in different system protection actions and making a tradeoff between desired resilience and budget constraints. Index Terms-Critical infrastructures (CIs), disruption risk, optimization , system planning, system resilience. NOMENCLATURE Indexes, parameters, and sets G(V, L 0) An undirected graph {G} comprising a set of nodes V connected by a set of links L 0 representing the original network system before the disruption.