Agent-Directed Resource Conversion Process Model Design Principles

摘要

As comes from the name, the paper discusses design principles for the multi-agent resource conversion process models. To study various kinds of processes, including technological, logistical and business, we use the multi-agent resource conversion process model. To run simulations on such processes more and more computational resources are required. Due to this, an important problem is search and use of new principles for building and analyzing multi-agent models of resource conversion processes. In order to analyze and simulate these processes we suggest a method for analysis and debottlenecking the multi-agent resource conversion process, which is based on integration of the named process model with operational analysis of probabilistic networks, multi-agent approach and expert systems. The paper presents the results of experimental comparison of the new method, suggested by the authors, with the critical path method. The developed principles of building the multi-agent process models and the analysis and debottlenecking method for the named process have been implemented in the automated system for metallurgical production.