Abstract. Objectives: To predict short term ionospheric Total Electron Content (TEC) data derived from the International GNSS Service (IGS) station. Methods: A new hybrid technique based on Autoregressive Integrated Moving Average (ARIMA) and Artificial Neural Network (ANN) model for the short-term ionospheric TEC prediction is constructed. The TEC dataset in this research work is utilized from Crustal Dynamics Data Information System (CDDIS), NASA from the year 2014 to 2016 and for each day the 10 minute average TEC data have been utilized for modeling and prediction. The hybrid model than compared with existing developed model such as ANN and ARMA as well as IRI-2016 global model. Findings: In order to estimate the performance of new hybrid model, it is compared with the existing ANN model, ARIMA model and IRI-2016 global model. Based on the comparison results, it is observed that new hybrid model predicts well than other prediction models with RMSE of 10.21 TECU and MAPE 0.034 TECU. Novelty: The proposed model can recognize the usual pattern of TEC in different seasons and it can predict the TEC values with 86{\%} accuracy up to 24 hours.

Full-text: PDF