Post-mortem studies carried out after recent power system blackouts have shown that time-stamped measurements from Phasor Measurement Units (PMUs) would have helped to avert these blackouts if they were available. Consequently, there is the need to optimally locate these PMUs to provide information about the system’s state. This paper proposes a novel method incorporating modal participation factors computed from the eigenvalues and eigenvectors of the reduced Jacobian matrix obtained from the system’s power flow with Binary Integer Programming (BIP), in solving the Optimal PMU Placement (OPP) problem while considering the system’s voltage stability. The proposed method identifies the critical buses prone to voltage instability, determines the minimum number of PMUs required for complete topological observability, and the best locations to site the PMUs. The proposed method was tested on benchmark IEEE 14-bus, 30-bus, and 57-bus test networks. Analyses of the results show that the proposed method provides a better procedure for the strategic placement of PMUs in practical networks for voltage stability monitoring especially for the cases of phased (multi-stage) PMU placement limited by the availability of funds and communication channels. Thereby making it necessary to give priority to critical/voltage weak buses during the phased installation process.