In this paper a best fitted Auto Regressive Integrated Moving Average model commonly known as ARIMA model has been developed for future projection of tea production in the state Assam by using various statistical tools and techniques viz., SPSS, R-software. We have also used various statistical test to study the nature of data and other time series related test. Data are collected from Tocklai Tea Research Institute, which is a pioneer institute of tea research in Assam. The type of data used in this study are secondary in nature. From the study we found that ARIMA (1,1,1) is suitable for the data set and we forecasted upcoming Fifteen years.