A physiologically-based demographic model (PBDM) was used to predict the population dynamics of Drosophila suzukii immature stages in the Apulia region in the South of Italy. In this work, the model was calibrated by data coming from experimental observations on the development rates of 50 eggs of D.suzukii to adults at different temperatures (5°C, 10°C, 16°C, 28°C, 32°C, 34°C), describing each step of juvenile development (L1, L2, L3, and pupa). Calibration was carried out by using the Marquardt-Levenberg algorithm by which experimental data, obtained in a growth chamber (life table) were sorted out as functions of temperature (range 5.6-32.4°C). Lactin-1 function was selected as appropriate for representing the development rates of egg, larval and pupal stages of D.suzukii. In contrast, Wang-2 function represented at best the mortality rates of D.suzukii immature stages. The calibrated model provided reliable functions of the variation in time of D.suzukii population dynamics within the above-mentioned temperature range. This model is the first PBDM that would predict population dynamics of the pest in Apulia, in which the average range annual temperature falls approximately between 5 and 32° C. Experimental data at 24°C were satisfactorily congruent with the model curve, thus confirming the evaluation of calibration. Thereafter, the validation of the model will be achieved by proving that it matches the open field data.