Multi-agent systems of semi-automatic design based on object-functions model in engineering

A methodology for creating semi-automatic design systems for engineering products using artificial intelligence technologies is proposed. The methodology is based on multi-agent methods for creating knowledge bases and is suitable for the development of design and control systems for digital intelligent industries. As a unified agent model, an object-function mechanism is proposed with a partition of attributes into a subset of the input and output attributes of the agent method. Object-function is considered as a unified module for creating multi-agent systems. Graphic and textual models of agent representation are proposed. It is shown that with this representation, the agent is equivalent to the production rule, which is a knowledge base module. The names of the variables of the knowledge module should be selected from a dictionary that can be compiled using various natural languages. The mechanisms of knowledge modules should ensure the implementation of all the functions that may be required in the formation of knowledge bases. It is shown that the mechanism of an object-function mechanism can be a system of agents. When using expert programming technology, the process of generating knowledge modules, translating them into object or executable modules in one of the traditional languages, and testing is performed as a single operation. After obtaining the necessary set of modules, a method is generated that uses a subset of the generated knowledge modules. The generated method can be used as a mechanism of the knowledge module, which allows the use of a hierarchy of rules. It is shown that a cooperative solution to complex problems is provided by a system of agents.