The rapid densification of 4G and 5G radio access networks (RAN) across the Kingdom of Saudi Arabia is driving a steep rise in energy consumption and operating expenditure, placing telecom operations in direct tension with the decarbonisation commitments of the Saudi Green Initiative and the Kingdom's 2060 net-zero target. The RAN is the dominant energy consumer in a mobile network, and conventional static energy-saving features are poorly matched to the extreme diurnal traffic and temperature swings of the Arabian Peninsula, where ambient heat both raises baseline power draw and inflates cooling load. This paper proposes the Hierarchical RAN Energy Optimization (HREO) framework, a three-tier, AI-driven, multi-vendor architecture that converts energy efficiency from a static configuration into a predictive, closed-loop control problem. A reactive cell tier exposes the advanced sleep modes (SM1–SM4) of heterogeneous Tier-1 vendor equipment; a predictive AI tier, hosted as a Near-RT RIC xApp, pairs a per-cell LSTM traffic forecaster with a reinforcement-learning sleep-policy agent whose reward explicitly trades energy saving against a hard quality-of-service penalty and a switching-cost term; and a network orchestration tier distributes policy across the multi-vendor estate through the O-RAN O1 interface and prioritises sites powered by solar-battery micro-grids. The framework is localised to Saudi conditions through a temperature-dependent power model and a high-irradiance solar profile, and all targets are expressed against published vendor power ratings and 3GPP baselines rather than proprietary operator data. The result is a deployable, vendor-neutral pathway that aligns national RAN operations with the Saudi Green Initiative while protecting user experience.
Siddiq Bin Salam, Sajjad Waqar Ahmad· Technium Sustainability· 0 citations
— The Kingdom of Saudi Arabia's ambition under Vision 2030 to rank among the world's top-ten logistics economies places its maritime gateways — principally Jeddah Islamic Port on the Red Sea and the adjacent King Abdullah Port — at the centre of a transformation toward autonomous, data-driven cargo handling. Legacy connectivity built on industrial Wi-Fi and best-effort 4G cannot guarantee the deterministic latency, mobility robustness, and device density that automated guided vehicles (AGVs), remotely operated ship-to-shore cranes, and real-time container tracking demand. This paper proposes the Hybrid Multi-Vendor Private 5G Blueprint (HMP-5G), an engineering and orchestration framework purpose-built for the topology, electromagnetic conditions, and procurement realities of Saudi industrial ports. The framework is organised across three coupled layers — a Radio Access layer mapped to port micro-zones, a Slicing layer that isolates safety-critical, automation, and enterprise traffic, and an Interoperability layer that allows heterogeneous Tier-1 vendor equipment (Huawei, Ericsson, Nokia) to coexist through standardised O-RAN O1/E2 and 3GPP service-based interfaces. The framework commits explicitly to a Standalone Non-Public Network (SNPN) model with on-site User Plane Function for data sovereignty, specifies hard slice isolation through dedicated physical resource block (PRB) allocation enforced by a Near-RT RIC xApp, and supports the isolation claim with a reserved-resource queueing model demonstrating that the URLLC control latency target is preserved under a one-million-device-per-square-kilometre telemetry surge. Performance targets are expressed against 3GPP TS 22.104 baselines rather than proprietary data, and every design choice is mapped to the National Transport and Logistics Strategy, the giga-project logistics layer, and the Saudi Green Initiative.
Siddiq Bin Salam, Sajjad Waqar Ahmad· Technium· 0 citations