Abstract. The article introduces the knowledge-based agricultural decision support system. The system includes database and knowledge base as well as modules used for formation of models, optimization, simulation, decision analysis and inference. The linear programming model contains several hundreds of variables and restrictions. The farm model, using entered data, is formed according to "if-then" type rules that are stored in the knowledge base. Having solved the task of optimization, the particular values of variables indicate what kind of crops should be grown and in what area, as well as what animals and how many of them should be kept and what resources and how much of them have to be used for achieving the biggest benefit under the environmental and other conditions. The simulation was employed to test the sensitivity of the plan to weather and market variations. Having applied a set of production rules to the given facts and modelling results within the module of decision analysis and inference, conclusions and suggestions are made. Decision support system performs the analysis of production efficiency, resource reserves and shortage, and with the help of the Internet in real time provides a farmer with conclusions and suggestions necessary to increase the efficiency of production conforming to environmental constraints. The integration of optimization calculations and knowledge management into the agricultural decisions-support system expands its possibilities and improves the quality of solutions.