Pakistan is the largest consumer of tea, but not the biggest producer of it. To fulfill its needs Pakistan expends a large amount of foreign reserve to import tea. To save foreign reserves and fulfill tea needs, we have to make precise decisions at the right time to increase the tea yield to maximize the utilization of land suitable for it. The factors on the basis of which farmers, researchers and the government make decisions to increase tea crop yield are environmental and soil conditions such as soil PH level, humidity level and rainfall level as well as Plucking rounds. We can use these factors to make a right and timely decision by using the decision support system with data mining because it encompasses both classical statistics and modern machine learning techniques. To develop the Decision Support System with Data mining we use different methods for decision making, such as decision trees, naïve Bayes, neural network and linear regression. These techniques utilize the dataset for factors affecting the tea yield, which is collected from black tea farms located in Shinkyari, Mansehra to build and train a data mining model to discover hidden patterns and relationships. Once we get a stable data mining model for decision support system using different techniques, we can have accuracy through cross validation for each. In this regard neural network has higher accuracy followed by decision tree and then other techniques. In this study, decision support system is developed to find the best environmental conditions to maximize the tea plan production.