Abstract: In this paper, the research and development of the expert system platform for forecast and prediction of agricultural pests were introduced in detail, including the design of the structure and functions, knowledge base and the model of the knowledge form, modules and characteristics of the platform.The platform consists of knowledge base, inference engine, modules of forecast and prediction, management for knowledge base, case database, and explanation. The knowledge base was loaded by ten tables in database supported by SQL server 2000 in the platform, and the relationships among the tables were defined. So, the expert's knowledge in platform, including the description of the knowledge, the critical values for the description, terms for distinguishment, and the grade of pestsharm, formed a web model. Expert-knowledge-based forward chaining reasoning andCase-Based Reasoning(CBR) were combined in the inference engine. Wizard and other patterns of hint in the interface of modules can guide the users to finish themanipulations of maintaining knowledge base, data input, reasoning process, showing results of pest prediction and forecast, management of case database, and explaining the system actions. The platform was of openness, self-education, and easyoperation. It can be used in prediction and forecast of pest in agriculture andforestry, etc.