Post-stack seismic inversion tightly integrates different datasets and provides an accurate and high-resolution image of the subsurface. Selecting a suitable inversion algorithm for reservoir characterization using seismic data is very important, especially in geologically complex areas. In this study, five post-stack inversion algorithms were applied to select the most optimum algorithm required for the delineation of thin-bedded reservoir sand of Lower Goru Formation, Kadanwari gas field, Pakistan. Inversion results demonstrate that linear programming sparse spike inversion (LPSSI) provides better results than band-limited inversion (BLI) and coloured inversion (CI), respectively. The other two algorithms, maximum likelihood sparse spike inversion (MLSSI) and model-based inversion (MBI), are not able to clearly resolve the thin-bedded E-sand reservoir. Probabilistic neural network (PNN) in combination with LPSSI was applied to predict the spatial distribution of porosity, which showed 98% correlation with log porosities. The combination of LPSSI and PNN can be used to better characterize the thin-bedded Cretaceous sands having similar depositional environments around the globe. [ABSTRACT FROM AUTHOR]