Dynamic Stock Buffer Management Method Based on Linguistic Constructions

Abstract

The paper proposes a dynamic stock buffer management method based on linguistic constructions. The novelty of the research lies in the fact that a control system based on fuzzy logic and linguistic constructions was created for the dynamic stock buffer management and two artificial neuro-fuzzy network models were created. Three criteria for evaluating the effectiveness were selected and the parameters of the proposed models were identified based on the backpropagation algorithm in batch mode and the genetic algorithm, which are oriented on the parallel information processing technology. The proposed models and procedures for their parametric identification make it possible to increase the speed, accuracy and reliability of decision making. The proposed dynamic stock buffer management method based on linguistic constructions can be used in various intelligent systems that exercise control in natural language.