Natural disasters can devastate the critical infrastructure of the affected regions, including the power and energy systems. Power grid resilience can be achieved during operation, by adopting risk-based dispatch strategies, or through proactive grid reinforcement and hardening strategies. A solution is proposed in this paper for optimal line reinforcement in order to enable microgrid formation during the course of a natural disaster event. The problem is formulated as a mixed-integer nonlinear multi-objective optimization model to minimize the reinforcement cost and maximize the amount of load served. The problem is solved subject to power flow and network topology constraints. A case study is presented to illustrate that through a selective and targeted line reinforcement, multiple microgrids can be formed in order to continue the local supply of the loads until damaged components are repaired or replaced, and service is restored to the main grid.