Today the basic criteria of estimation quality of functioning of computerized air traffic control systems don’t exist. Dissolving of the problem of multi-objective optimisation is achieved due to scalar approach to a vector optimisation through Saaty analytic hierarchies process. The analysis of features of a system of key performance indicators for the estimation the state and control quality of service of air cluster composite network are executed. The methodology of current estimation of network parameters and state, and their impact on final efficiency of air traffic service is developed with application step-by-step regression. The results of calculations of correlation matrix of key performance indicators are resulted and statistical dependence between basic parameters are analysed, which a network performance and quality of service depend on and provide optimal air traffic management. An ad hoc assumption about errors of relative priorities of partial criteria allows excluding risk of singularity of priority matrix in applied method. Moreover, the presence of mentioned errors decreases risk of optimisation algorithm looping on local extremes.
Nowadays, application of modern technical means ensures implementation of flexible air space management structure, which provides immediate changes of air traffic control sector edges depending on current air environment. Thus, air traffic optimization occurs ensuring the required air traffic safety rate, "apron-to-apron" air traffic service method and application of area navigation technology, which is known as "free-flight" method.
Application of the computerized control network key parameters as a complex system with signaling information delay can afford its state forecast and real-time service quality control problem solving.
Key performance indicators are selected in a way ensuring comparative performance evaluation of different hub networks, and most significantly, providing guaranteed service quality, which is the highest priority task for the air traffic management systems as critical application systems. This article stipulates separated network performance indicators and scale types for their evaluation.