Аннотация
Based on the results of empirical research and a natural psycholinguistic experiment, the study aims to justify the feasibility of an innovative prognostic tool for improving flight safety in low-altitude civil aviation in the Republic of Kazakhstan. This tool is based on a Markov-entropic model for assessing the risk of information-related errors of psycholinguistic nature arising in multilingual communication. Methods. This research uses a triangulation approach, which allows several research methods to be used simultaneously. These methods include: the causality method; multifactor analysis; Markov-entropic mathematical modelling; the entropy method; a natural psycholinguistic experiment; experimental scenario prediction; the graphic method; the method of logical sequence; and the method of generalisation. Results. Based on the results of the natural psycholinguistic experiment, a Markov-entropy model for assessing the risk of communicative errors of psycholinguistic nature arising in multilingual aviation communication was developed and verified. The application of this model allowed not only to quantify the probability of such errors, but also to identify the key factors influencing their occurrence in real operational scenarios. The psycholinguistic approach to risk modelling used has demonstrated its effectiveness in identifying hidden dynamic patterns inaccessible to traditional methods of analysis. It was found that entropic characteristics of the communication flow serve as a reliable indicator of its complexity and potential vulnerability, while Markov chains effectively describe the sequential states of the communicative act and the probabilities of transitions between them, which is critical for the predictive detection of potential failures. Conclusions. The results obtained confirm that the integration of psycholinguistic parameters into the risk assessment model significantly improves the accuracy of communicative error prediction and forms the basis for the development of effective preventive measures aimed at minimising the human factor and improving overall flight safety. The developed model is an innovative predictive tool with significant potential for practical application in the aviation industry.