The neural-network based synthesis of developing strategies for regional industrial production

The paper presents a technique to synthesize regional production development strategies under unrepresentative statistics of macroeconomic data. Fuzzy variables are offered for data presentation, while Boolean neural networks, whose structure and parameters are determined from the learning sample, are applied to synthesize control actions under incomplete a priori information. The technique was successfully applied in 1995–2000 to synthesize short-term industrial production development strategies in Mordovia and Penza regions of the Russian Federation.