CONSTRUCTION OF MULTIVARIATE LINEAR TREND MODELS OF FEEDBACK PROCESSES IN AIR TRANSPORT SYSTEMS

Abstract

The trend model proposed in the article provides an approach for methodical assistance to the work to be done in the direction of improving the effective management system of airlines. Aviation has different characteristics from other service sectors, passenger requirements and satisfaction are regulated by geographic, economic factors, context of use, local and international governance, resource use and standards. Taking into account these characteristics, the problems related to passengers were systematized and summarized as flight delays and cancellations, problems related to baggage, overbooking, and a trend model of dissatisfied passengers depending on these factors was established. The model was built using a set of statistical data related to air transport of the United States of America, and the adequacy of the results was evaluated by analyzing the relative error between the actual data and the values calculated based on the model. The analysis of the results of the calculation based on the years 2019 and 2022-2023 showed that the relative error has decreased in recent years. Decrease during the negative impact of the COVID-19 pandemic on air traffic is finally reversing its impact and airlines are starting to stabilize. The obtained results can be used in determining the forecast indicators of dissatisfied passengers, in the future, based on the proposed model, in the construction of more perfect intellectual models of the service sector with the application of modern IT technologies.