This work is aimed at verifying that analytical modelling approaches may provide an estimation of the outdoor performance of TF (thin film) PV (photovoltaic) technologies in inland sites with sunny climates with adequate accuracy for engineering purposes. Osterwald's and constant fill factor methods were tried to model the maximum power delivered and the annual energy produced by PV modules corresponding to four TF PV technologies. Only calibrated electrical parameters at STC (standard test conditions), on-plane global irradiance and module temperature are required as inputs. A 12-month experimental campaign carried out in Madrid and Jaen (Spain) provided the necessary data. Modelled maximum power and annual energy values obtained through both methods were statistically compared to the experimental ones. In power terms, the RMSE (root mean square error) stays below 3.8% and 4.5% for CdTe (cadmium telluride) and CIGS (copper indium gallium selenide sulfide) PV modules, respectively, while RMSE exceeds 5.4% for a-Si (amorphous silicon) or a-Si:H/mu c-Si PV modules. Regarding energy terms, errors lie below 4.0% in all cases. Thus, the methods tried may be used to model the outdoor behaviour of the a-Si, a-Si:H/mu c-Si, CIGS and CdTe PV modules tested ordered from the lowest to the highest accuracy obtained in sites with similar spectral characteristics to those of the two sites considered. (C) 2014 Elsevier Ltd. All rights reserved.