Abstract: The establishment of Quality of Service (QoS) guarantees as required by applications through conventional routing protocols is a challenging endeavor in MANET due to its dynamic topology, lack of load balancing capabilities, limited power supply and a deficiency of unified authority. Hence, it is essential to have an efficient routing algorithm to satisfy a set of QoS constraints in spite of network dynamics. In this research, we afford a new improvement in AODV routing protocol to deliver QoS guarantees for hard real time applications in MANET. Conventional AODV routing protocol uses priority-only-based packet scheduling to arrange the packets in its queue. Existing scheduling algorithms in AODV are not efficient enough to provide QoS for hard real time packets in uncertain and dynamic environments because these approaches assume that the network is deterministic and predefined decisions will be statically implemented during scheduling. To tackle this problem, we present an interval number theory to explore the network dynamics and an argency aware architecture for packet scheduling to enhance the service demands. According to our architecture, we develop a novel Urgency aware Packet Scheduling (UaPS) algorithm and priority based buffer management policy for real time packets. We perform extensive experiments in order to examine the efficiency of the UaPS-AODV in the NS-2.34 (Network Simulator) and compare with basic AODV and Delay aware AODV (AODV-D) protocols. Our results reveal that the UaPS-AODV outperforms those existing algorithms in terms of performance metrics.