India’s fast-moving consumer goods (FMCG) sector is rapidly growing but faces challenges in balancing high service levels with efficient inventory turnover. This study examines how predictive demand segmentation – using data-driven grouping of products and customers by demand patterns – can enhance inventory management in Indian FMCG manufacturing. Employing a mixed-methods approach (combining quantitative analysis of sales/forecast data and qualitative insights from industry experts), we analyze segmentation techniques and their effects on key performance indicators. We find that clustering SKUs by demand variability and applying tailored forecasting models significantly reduces stockouts and excess inventory. For example, Gradient Boosting time-series models achieved the lowest forecast error (MAPE ~6.8%), allowing leaner safety stocks and higher turnover (Dhongde & Nanda, 2024). Simulation of demand spikes showed that proactive planning (via segmentation and scenario analysis) helped maintain service levels even with a 20% demand surge (Dhongde & Nanda, 2024). In practice, companies like Hindustan Unilever have applied AI-driven forecasting to optimize supply chains and improve both turnover and service (Unilever Hindustan Unilever Limited, 2025). Our findings suggest that advanced segmentation and predictive analytics together enable FMCG manufacturers in India to tailor inventory policies by segment, boosting inventory turnover and service level performance.