Freeze-dried mango powder is prone to quality degradation during storage. Variations in raw material quality and production conditions across batches further complicate long-term quality control. Repeated and lengthy shelf life testing is both time-consuming and costly. Therefore, mathematical models play a crucial role in estimating shelf life, reducing costs, and tailoring storage conditions for each production batch. The objective of this study was to develop and propose mathematical models to quantitatively predict changes in moisture content, bioactive compounds, and antioxidant activity in mango powder over time. Mango powder was stored at four different temperatures (20°C-55°C) corresponding to RH = 33%-55%. Mathematical models were developed based on statistical parameters ( , RMSE, ). The results showed that moisture content was dependent on storage temperature, with the most suitable model being Vu (  > 0.97) at all four temperatures. Zero-order, first-order, and Weibull models were considered suitable to describe the mechanisms of bioactive components and antioxidant activity with  > 0.99. The models predicted the shelf life of the product from 85 to 551 days depending on the storage conditions. Additionally, the equilibrium moisture content of mango powder under the current storage conditions was found to be less than 5%. The study's results provide experimental models and form the basis for improving product quality, predicting shelf life, and establishing automated sample control in production.