Delay-tolerant network (DTN) is a session of communication networks developed to function in challenging environments characterized by intermittent connectivity and unpredictable delays. These networks have many practical applications such as remote sensor networks, disaster response, etc. However, routing in DTNs is challenging due to the absence of end-to-end paths, making traditional routing protocols ineffective. This paper introduces an optimized routing framework named firefly algorithm-based spray and wait (FAbSW) for DTNs. The proposed FAbSW enables effective message delivery in DTN environment considering its intermittent and unpredictable connectivity by estimating the optimal neighbor node for packet forwarding. The presented framework designed and implemented in the MATLAB environment by creating the DTN with varying number of network nodes. The experimental results demonstrates that the proposed methodology attained greater delivery rate of 98.98%, and minimum delay of 2 s, which is comparatively better than the conventional methods.