Abstract
The Author develops die life analyzer working on the internet with using the cooperation with independent databases of cold forging die-life instances. The instance databases are established with HTML and XML on the internet, that the forging engineers access them with www browser usually as references for their forging process design. Furthermore, the analyzer deals these databases as a knowledge base to predict and analyze forging die-life and the cause of short life. The author considers die-life analysis as the risk analysis that the life of target tool is to be worse than a life of instances. The author has developed die-life analyzer using risk tree networks and fuzzy linguistic risk calculation. The author investigates the causes of short life to perform risk analysis tree network. A cause, we call a parent node, is divided to more detail causes, we call daughter nodes. Daughter nodes are connected to its parent nodes with weights, and risk value of parent node is calculated as average of daughter nodes' value with the extension principle. The risk values of causing shorter life than instances, that is to be the risk value of the top of all nodes, are calculated through the above process finally. After then, the system predicts die life with interpolation using die life instances and the corresponding risk values making die life shorter.
