The source-based routing using quality of service (QoS) metrics results in alternate path discovery considering link-availability time, link costs, and path delays to overcome the barrier of information blocking on a selected path by choosing alternative routes. However, multipath selection, as well as the cost of selecting a path, make the route selection a challenging task. This paper proposes a vehicle orientation based QoS routing in vehicular ad-hoc networks (VANETs), called OBQR that exploits vehicle orientational information instead of magnitude information. The cosine similarity concept with a preliminary scalarization model converts the multi-constraint objectives into a single constraint objective to find a set of possible paths to the destination. We evaluate the performance of our approach and compare it with existing state-of-the-art schemes based on QoS routing and clustering. Experimental results demonstrate that the proposed scheme significantly improves path selection and load balancing with better QoS routing performance.