The network reconfiguration procedure after a complete blackout or a local outage usually lasts a relatively long period of time and involves a lot of factors. It can be formulated as a combinatorial optimization problem with multiple variables and objectives. A chance constrained programming（CCP） model for network reconfiguration scheme optimization is developed with the operating time uncertainty taken into account. The node number of the restarted generators and the benefit from the restored loads are taken as the returns from the restoration scheme and the network reconfiguration scheme of maximum returns is searched in the process of network reconfiguration. Then a modified genetic algorithm considering the characteristics of the network reconfiguration scheme optimization problem is proposed. Finally， a sample power system is adopted to demonstrate the essential features of the model and method developed.