Modeling and optimization of service center inventory management: case study

Abstract

The paper addresses inventory management in a small electronics service center that combines repairs (B2C) with selling spare parts to other workshops and end customers (B2B/B2C). In this hybrid model, inventory is both an input to service delivery and a product for sale, so a priority conflict arises: sales may reduce availability of parts for urgent repairs. The purpose is to propose an integrated inventory management system that links SKU segmentation (ABC/XYZ), replenishment parameter settings (EOQ/ROP/safety stock), and data requirements for channel‑separated demand accounting. The methodology combines inventory theory and operations research: an “as‑is” review of warehouse processes and replenishment routines, KPI calculation (repair time, stockout frequency, turnover, dead‑stock share), combined ABC/XYZ analysis to select differentiated control policies, and illustrative calculations of EOQ, reorder point (ROP) and safety stock under lead‑time uncertainty. The empirical illustration is based on inventory information from the «ProxyS» service center. The diagnostics shows that current replenishment is reactive and relies on intuitive minimum‑stock thresholds without systematic consideration of demand, SKU value and lead time. A key information issue is identified: inventory withdrawals are not tagged by consumption channel, so repair demand and sales demand are aggregated, which triggers cross‑channel cannibalization and repair delays due to waiting for parts. ABC/XYZ segmentation supports a differentiated policy set: continuous review and ROP/safety stock for high‑value A‑class items, simpler periodic review for B‑class items, and minimalistic rules and limits for C‑class items. Integrating ABC/XYZ with EOQ/ROP/safety stock and data‑governance rules provides a practical pathway from ad‑hoc to controlled inventory management. A critical prerequisite is channel‑separated consumption accounting (repairs vs. sales) and recording a minimal set of attributes at issue time (reason, customer type, priority, and job/order linkage). The approach is feasible for small service businesses and can be implemented in stages: first data discipline and KPI tracking, then segmentation and a policy map, then replenishment parameter tuning for the most valuable and critical items.