This paper presents the development of an automated Python-based simulator for MPLS networks, enabling flexible evaluation of network performance under varying bandwidth, latency, and link cost configurations. Unlike traditional tools such as GNS3, which are limited in scalability and automation, the Python implementation supports dynamic traffic modeling, prioritization strategies, and detailed quality of service (QoS) analysis. The study focuses on assessing critical traffic types (VoIP, VoD, and IoT) along with high-demand experimental flows, analyzing how route optimization and bandwidth adjustments impact latency, throughput, and packet loss. Results confirm that prioritizing critical routes and increasing bandwidth on congested links significantly enhance service quality, particularly in latency-sensitive and high-throughput applications. Additionally, the inclusion of synthetic traffic flows reveals the network’s robustness under extreme load conditions and helps identify residual bottlenecks. Overall, the findings validate the applicability of adaptive MPLS optimization strategies in dynamic environments supporting modern convergent services.