In this study, the weighted model averaging (WMA) was applied to calibrate ensemble forecasts generated from Local ENsemble prediction System (LENS). The WMA technique, which is easy to implement post-processing technique and lays greater weight on ensemble member forecast that exhibits the best performance, provides probabilistic visibility forecasting that takes the form of predictive probability density function for ensembles. The predictive probability density function used is a mixture of discrete point mass and two-sided truncated normal distribution components. Observations were obtained from the 3 International Airports (Gimpo, Incheon and Jeju) and 13 ensemble member forecasts derived from the LENS that were obtained between December 2018 and June 2019. Prior to applying to WMA, reliability analysis was conducted by using rank histogram and reliability diagram to identify the statistical consistency of ensembles and corresponding observations. Then, WMA method was applied to each raw ensemble model and proposed a weighted predictive probability density function. The performances were evaluated by mean absolute error, continuous ranked probability score, Brier score and probability integral transform. The results showed that the proposed method provide improved performances than the raw ensemble, indicating the predictive probability density function is well calibrated with respected to raw ensemble.