Vehicular ad hoc network (VANET) is a subset of mobile ad hoc network (MANET) where data communication occurs among vehicles and other road users. Due to the rapid topological changes, designing efficient routing protocols for VANET in terms of reliability becomes a major concern. For dense ad hoc networks, Optimized Link State Routing protocol (OLSR), a link state and a table driven proactive protocol is preferred owing to its multipoint relaying (MPR) feature. The route information of OLSR may include data such as broken link information, which leads to vagueness. Due to this limitation, it is necessary to optimize the OLSR routing for efficient data transmission. The main objective of this paper is to enhance the performance of the proactive OLSR routing protocol by minimizing the uncertainty in the routing information using Fuzzy Rough Set Theory (FR). Our investigation shows that the proposed OLSR named FR-OLSR has shown a significant performance improvement in terms of packet delivery and delay to be apt for data services on roads.