Optimizing customer service for the number of drivers in an area through real-time transportation service industry online to scale up. In this paper, the dataset used is traffic management accompanied by attributes such as level 6 geohash, day, timestamp, and demand. The dataset used is a sample from geohash6 coded qp0991, containing online transportation demands from 01/04/2018 until 31/05/2018 (61 days). The training datasets are from the qp0991 code sample, starting from 01/04/2018 until 10/05/2018 and the remaining datasets are used as the testing datasets. The percentages for training and testing are respectively 67% and 33%. The methods applied to construct the objective function are three different forecasting methods, namely the Naïve approach, auto-regressive integrated moving average (ARIMA), and simple exponential smoothing. The results of this study indicate that the simple exponential smoothing method is better than the naïve approach and auto-regressive integrated moving average based on the root mean square error (RMSE), mean absolute error (MAE), and mean absolute percentage error (MAPE). The simple exponential smoothing has an accuracy rate of 98.7% for the RMSE value, 98.9% for the MAE value, and 88.81% for the MAPE value. Keywords: time series, traffic management