Globally, exponential data growth is observed with mobile traffic generated from devices likerntablets, smartphones and other devices. Likewise, Addis Ababa cityâ€™s cellular network data trafficrnis increasing exponentially. To absorb this high traffic demand ethio telecom, the telecom servicernprovider in the city continuously expands and optimizes the cellular network. Having knowledgernof the growing data traffic demand in advance at a given time and space will assist ethio telecomâ€™srnplanning strategy and optimization. rnRecently, few studies are conducted to forecast Addis Ababa cityâ€™s Universal MobilernTelecommunication System (UMTS) network traffic using statistical time series models and neuralrnnetwork models. However, the studies deal with only time-domain forecasting and recommendrnto do from a spatial point of view. Moreover, another study modeled the spatiotemporal mobilerndata traffic, which can capture the space and time variation of UMTS data traffic in the city; thernstudy recommends the need of spatiotemporal data traffic prediction. rnIn this thesis, a deep neural network model, specifically Convolutional Long Short-Term Memoryrn(ConvLSTM), is used for spatiotemporal data traffic demand prediction of Addis Ababa city. Threernmonths' real dataset from 739 base stations is collected and preprocessed from ethio telecomâ€™srnUMTS network. After defining geographical grids, the ConvLSTM model is applied, which canrncapture spatial correlations through convolution operators and temporal dynamics through thernLSTM network for prediction. rnThe proposed model can predict up to six hours of future data traffic with a root mean squarernerror (RMSE) of 1.37. Additionally, the predicted data traffic demand is analyzed with respect tornblocked data traffic at a given space and time which gives significant insight to the optimizationrnprocesses like load balancing.