Directional overcurrent relays (DOCRs) are the essential protective devices in distribution networks which are usually set without considering any contingencies. However, the current challenge in power systems is the existence of uncertainty and its unfavorable consequences. It sometimes appears that some elements simultaneously fail which makes other parts to be overloaded to the extent that it leads to cascading outages. Therefore, DOCRs may have mal-operation which ends in unwanted trips when there is no fault, or they may not operate in the case the fault is located within their reach point. In such cases, the coordination setting will need complex programming with many related non-linear inequality constraints. In this paper, a novel hybrid method is proposed based on multi-objective optimization including new objective functions by using the genetic algorithm (GA). Also, the cascading outages are considered in the presented method based on network data analysis. This approach is performed on distribution part of the IEEE 14-bus meshed system, and a real industrial radial feeder named TOSEE, located in Iran. The simulations have been implemented in MATLAB and Power Factory-DIgSILENT software packages in different models, and the results are evaluated.