Shipboard spare-parts stores must accommodate heterogeneous item dimensions within limited space while remaining reliable under vessel-induced vibration, intermittent network connectivity, and dependence on manual inventory records. This paper presents the conceptual design of an intelligent spare-parts shelving system for shipboard storage environments. The proposed architecture combines a four-tier modular frame with adjustable dividers and motion-restraint components, a fixed sensing-unit matrix with dynamic virtual-compartment mapping, and an edge-based inventory-management layer. Weight filtering, stable-state detection, and unit-mass-based quantity estimation are formulated to identify removal, placement, replenishment, and anomaly events. Authenticated user sessions, local event-log caching, and recovery after network or power interruptions are incorporated to support inventory-ledger updates and personnel traceability. Engineering feasibility is assessed from the structural, sensing, communication, power-supply, and management-loop perspectives. The analysis indicates that the concept provides a coherent technical route for improving the accuracy and traceability of shipboard spare-parts management within the defined scope of small- and medium-sized, low-hazard items. Prototype development and shipboard testing are required to quantify measurement accuracy, event-recognition performance, structural safety, and operational reliability.