
Safety stock is a buffer quantity kept on hand to absorb unexpected demand spikes or supplier delays. It is typically calculated using the desired service level, demand variability, lead time variability, and average consumption. The goal is to maintain product availability without holding excessive inventory, striking a balance between service and cost.
Classic formulas assume demand follows a normal distribution and lead times stay constant — assumptions that rarely hold in volatile markets. Dynamic safety stock models use probabilistic forecasting to recalculate buffers item-by-item as demand patterns and supplier performance shift, preventing stockouts during peaks while avoiding excess stock during quieter periods.