Despite the use of different indicators by existing methods for black-start path optimization to represent the importance of transmission lines and nodes in a power grid， they are not complete. A new method for black-start path optimization based on path and node weight factors is put forward， which are calculated to quantitatively judge the importance of path and node. This method considers each of the factors that influence the importance of path and node to increase the completeness of the indicators. Based on the calculation and comparison of relational weight factors， the identification of the restoration path can be made to realize the optimization of the black-start path. The restoration strategy is divided into four parts: system start-up， pre-formation of backbone network， backbone network restoration and all-direction restoration. Corresponding path optimization models are built to meet the demands of different phases. Finally， the new method combined with the restoration strategy is used to build black-start paths for an actual power grid.