Strategic inventory management decision framework for CTO manufacturing

Purpose<br/>Configure-to-Order (CTO) manufacturing combines product customization with partial standardization, creating substantial heterogeneity in SKU demand patterns. Conventional inventory policies based on uniform service levels and ABC classification often fail to reflect this heterogeneity, leading to excessive stock levels and persistent service risk. This study develops a demand-segmented inventory management decision framework that aligns inventory control logic with observable SKU-level demand characteristics in CTO environments.<br/>Design/methodology/approach<br/>The study employs an embedded single-case research design combined with quantitative analysis of two years of ERP transaction data covering several thousand stock-keeping units (SKUs). Demand volume, variability (coefficient of variation), and replenishment lead time are analyzed to profile demand heterogeneity. Inventory behavior under existing reorder-point policies is evaluated and compared with differentiated inventory control algorithms, including Kanban replenishment, forecast-based reorder points, selective buffering, and buy-to-order strategies.<br/>Findings<br/>The analysis demonstrates substantial heterogeneity in demand variability and lead times, with weak correlation between demand volume and variability. Uniform inventory policies systematically inflate stock levels for high-variability SKUs while failing to mitigate stockout risk. Assigning differentiated control algorithms based on demand regimes significantly improves alignment between inventory levels and service risk.<br/>Research limitations/implications<br/>The findings derive from a single CTO manufacturing case and support analytical rather than statistical generalization. Future research should validate the framework across multiple industries and explore automated algorithm-selection models.<br/>Practical implications<br/>The framework provides managers with a structured method for linking SKU-level demand characteristics to appropriate inventory strategies, enabling reduced working capital and improved service performance.<br/>Societal implications<br/>How will it influence (corporate) social responsibility or environmental issues? How could it inform public or industry policy? How might it affect quality of life? Not all papers will have social implications.<br/>Original/value<br/>The study reframes inventory management in CTO systems as an algorithm-assignment problem and integrates demand profiling with differentiated inventory control logic in an empirically grounded decision framework.