In Vehicular Ad Hoc Network (VANET), vehicles on the road provide mainly two types of services to drivers or passengers inside vehicles, namely, safety and comfort services. Safety messages are time-critical, therefore in this paper we have proposed a novel Fuzzy Logic based Greedy Routing (FLGR) protocol which assists in delivering safety messages to the destination vehicle with minimum delay. The proposed FLGR is a multi-hop routing protocol that uses multiple metrics of neighbor vehicles to decide which neighboring vehicle will be the approximate best next-hop node for further forwarding the packets. We have used the concept of fuzzy logic that helps in decision making in VANETs. Fuzzy logic deals with reasoning that is approximate rather than fixed and exact. Fuzzy logic variables may have truth values that range in degrees between 0 and 1. The fuzzy logic inference system became important and useful when the values of the decision criteria are not only vague but uncertain in nature. FLGR is used within the communication range of the current forwarding node that employs fuzzy logic for selecting the best next-hop node in multi hop VANETs. We have used the fuzzy logic tool box of MATLAB to evaluate our protocol. Metrics such as distance, speed, direction, and position of next-hop forwarding node are put into fuzzy logic system. Thus, our protocol helps in finding the approximate best next-hop node for routing the safety messages by employing fuzzy logic. It selects the next-hop node having maximum distance, speed, and progress and minimum angular deviation of nodes towards destination from current forwarding node. Further, FLGR is more suitable for safety applications in vehicular networks.