Opportunistic Mobile Networks (OMNs) are an extension of MANETs where the end-to-end transfer delays in these networks are much larger than conventional networks, such as Internet. Consequently, routing algorithms in OMNs only rely on locally available information when making routing decisions. To improve the delivery performances, most of the OMN routing schemes exploit more than one routing metrics. For instance, SimBet routing considers two social properties of the peer nodes, namely social similarity and betweenness centrality, to select better relay nodes. Typically, the algorithm assigns the same weight to both the metrics when making routing decisions. However, our investigation shows that this assignment may lead to the suboptimal performance of the algorithm in real-life OMNs. Therefore, we propose the entropy weight method (EWM) that dynamically assigns the weights of the SimBet’s routing metrics for a given node mobility scenario. Eventually, simulation results show that the proposed algorithm is able to improve the delivery performances of SimBet routing, in terms of delivery ratio, average delay, and overhead ratio, in several real human movement cases.