Environment-Aware Adaptive Zero-Trust Networking Framework for Reducing Network Complexity and Latency in Enterprise Hybrid Cloud Environments

Abstract

Zero-trust architecture strengthens resource-centric security but can increase access-decision latency, policy volume, and operational coupling in hybrid cloud, edge, Internet of Things, and remote-access environments. This study presents the Adaptive Environment-Aware Zero Trust Orchestration Framework (AEZTOF), a closed-loop architecture that coordinates continuous trust estimation, risk-adaptive authentication, software-defined routing, traffic classification, and policy optimization. The enterprise-scale experiment covered 12 sites, four hybrid-cloud regions, 85 edge nodes, 10,500 IoT devices, 28,000 user endpoints, 720 microservices, and 31,800 security policies over a 30-day logical workload window. Eight approaches were evaluated through 300 runs per method, giving 2,400 method-level runs across 14 attack scenarios. AEZTOF achieved 99.4% authentication success, 17.9 ms authentication latency, 99.5% decision accuracy, 22.6 ms end-to-end latency, 10.1 Gbps throughput, 0.61% packet loss, 98.9% threat detection, and 99.3% policy compliance. Relative to conventional ZTNA, authentication latency fell by 31.9% and end-to-end latency by 28.3%; relative to a traditional enterprise network, policy conflicts fell by 77.2% and the complexity index by 51.0%. Omnibus and paired tests were significant at p < 0.001, with a large standardized effect (Cohen's d = 0.82). Ablation, sensitivity, calibration, unseen-attack, controller-failure, concept-drift, encrypted-traffic, and scalability analyses confirmed that the gains remained robust across operational conditions. The study contributes an explicit multi-objective orchestration model and a reproducible enterprise evaluation protocol for adaptive zero-trust deployments.