Abstract
The paper presents the idea of an agent-based information processing system, which can be recognized as an alternative approach to the optimization of industrial processes. The main objective of the proposed approach is to find similarities between the current production and the production in the past. The goal of this search is to choose the period of the past production, which is characterized by the conditions similar to the current ones and simultaneously, which presents desired quality of production. The found period provides useful information on how to control the current production and can be used as the starting point for the additional optimization analysis. In order to verify proposed approach, an agentbased information processing system was developed, and tested on the real industrial process. Preliminary results were shown.