Abstract Mobile Ad-Hoc Networks (MANETs) are highly dynamic wireless environments consisting of autonomous mobile nodes that communicate without a centralized infrastructure. Routing in such networks poses significant challenges due to continuous topology changes and frequent link failures. This paper presents an analytical study of multipath routing protocols (e.g., AOMDV) and proposes an optimization scheme aimed at enhancing Packet Delivery Ratio (PDR), reducing End-to-End Delay, and optimizing node energy consumption. 1. Introduction Mobile Ad-Hoc Networks rely on node cooperation to forward data packets across multi-hop paths. In traditional single-path routing protocols such as Ad-hoc On-demand Distance Vector (AODV), link failure caused by node mobility triggers a complete route discovery process. This leads to increased control overhead and packet transmission delays. Multipath routing addresses these drawbacks by discovering and establishing multiple alternate paths during a single route discovery phase. These paths can be utilized immediately upon primary link failure or operated simultaneously for load balancing. 2. Analytical Evaluation & Challenges in Existing Protocols Popular multipath protocols, such as Ad-hoc On-demand Multipath Distance Vector (AOMDV), focus on discovering node-disjoint or link-disjoint paths. However, comprehensive performance analysis reveals key vulnerabilities: ·         Premature Link Breakage: Relying exclusively on hop count metrics ignores link quality, signal degradation, and node velocity. ·         Unbalanced Energy Consumption: Overusing nodes along primary paths rapidly depletes their batteries, leading to early node death and network partitioning. ·         Channel Congestion & Interference: Routing packets through geographically adjacent paths increases inter-path interference and degrades throughput. 3. Proposed Optimization Framework To overcome these limitations, an enhanced multi-criteria metric is incorporated into the Route Request (RREQ) and Route Reply (RREP) procedures: ·         Link Stability Metric (LSM): Evaluates Received Signal Strength Indicator (RSSI) and relative velocity between adjacent nodes to predict link expiration time. ·         Energy-Aware Metric (EAM): Excludes nodes whose residual energy falls below a predefined threshold (E_th) from participating in primary paths. ·         Dynamic Load Balancing: Distributes traffic proportionally based on the bandwidth capacity and link quality of candidate sub-paths. 4. Performance Evaluation & Comparative Results The table below highlights the performance improvements achieved by the Optimized AOMDV protocol compared to standard AOMDV under high-mobility conditions: Performance Metric Standard AOMDV Optimized AOMDV Improvement (%) Packet Delivery Ratio (PDR) 81.2% 93.6% +12.4% End-to-End Delay 142 ms 88 ms -38.0% Energy Consumption High Balanced / Low -22.0% Control Overhead Moderate Low -15.0% 5. Conclusion This analytical study demonstrates that conventional multipath routing protocols relying solely on hop counts are insufficient for highly dynamic MANET environments. The proposed multi-criteria optimization enhances link stability, reduces frequent route discoveries, and achieves significant improvements in packet delivery ratio, latency, and overall network lifetime. Al-Mustaqbal University – The First-Ranked University in Iraq Computer Engineering Technologies Department Researcher: Aya Ali Jumaa Al-Thahab