The production of medical cannabis in compliance with the guidelines of Good Manufacturing Practice requires a science-based control strategy for the process. In this work, we hypothesized that industrial tray drying process of cannabis inflorescences can be achieved through an accurate selection and fitting of mechanistic models. In total, three industrial runs were measured and modeled from an initial moisture content of 70–72 wt% to a final content of ≤12 wt%. The temperature, relative humidity, and flow rate of convection air were kept constant at 20 °C, 50 %, and 9800 m3h−1. A two-period model based on external resistance and intraparticle diffusion (FUMM+SPPM) was the best to describe the process, with a fitting error of 16.9 %. A control chart was conceived using FUMM+SPPM and a non-linear relation for the standard deviation along time. The suitability of the FUMM+SPPM was demonstrated with a validation run, resulting in a prediction error of 11.5 %; and where the upper and lower control limits of the control chart were respected. A sensitivity analysis showed that the initial moisture uncertainty can offset by 11.6 % the duration of the drying process. A deeper analysis on the fitted intraparticle diffusion coefficients points the migration of water to resemble osmotic movement. Globally, this study validates the adopted pathway to launch and improve a statistical process control for the tray drying of medical cannabis inflorescences.