Forecasting is a technique to estimate a value in the future by taking into account past data and current data. Forecasting method is divided into two namely qualitative methods and quantitative methods. More qualitative methods are based on an analysis based on intuitive thinking, logical and previous thinking. In contrast to quantitative methods, quantitative methods require the quantitative information of the past to be quantized in the form of numerical data. There are two types of quantitative forecasting models: time series model and regression model. The time series model seeks to predict future conditions by using historical
data and extrapolating them into the future. One of the forecasting methods that can be used in this case is Autoregressive Integrated Moving Average (ARIMA). This method was first developed by George Box and Gwilym Jenkins. ARIMA represents three models: autoregressive, moving average, and autoregressive moving average. The purpose of this research is to know the best model (ARIMA) that can be used to predict the amount of fish production at Pangkalan Pendaratan Ikan (PPI)SeliliSamarinda City. The data used from Januari 2017 to April 2017 to determine the results of forecasting the number of fish production in PPI
SeliliSamarindaCity for the next 15 days from16 April 2017 until 30 April 2017. Based on the analysis obtained that the best model to forecast the amount of production Fish in PPI SeliliSamarinda city is model ARIMA (0,1,1), and obtained result of forecasting amount of fish production in PPI SeliliSamarinda City for 15 days ahead is decreasing.