The methods of management of metrological support of the measuring instrument fleet are considered. The modern situation is described, when the fleet of measuring instruments is heterogeneous and includes both obsolete samples of measuring instruments with a long service life and significant operating time, and the newest, high-tech samples of measuring instruments. At the same time the proportions between the mentioned groups of measuring instruments change over time, as the ageing of measuring instruments and their transition from one group to another takes place. Also the measurement instrument park is renewed as a result of purchases of new and modernization of existing samples. The heterogeneity of the measuring instrument park leads to the necessity of development and application of new methods of management of metrological support of the measuring instrument park, including the use of mathematical modeling. One of the promising methods of metrological management of the measuring instrument fleet based on the risk-oriented approach is proposed. The probability of finding a randomly selected sample of measuring instruments from the measurement instrument park at an arbitrary moment of time in the state of unreadiness for intended use is used as a risk indicator for the measuring instrument park. In accordance with the risk-oriented approach the measuring instrument park is divided into risk classes. The algorithm of assigning measuring instruments to different risk classes is developed, based on solving a series of optimization problems of the fleet operation taking into account the processes of aging and renewal of the fleet of measuring instruments. The results of application of the risk-oriented approach in the problem of modeling of metrological support of heterogeneous fleet of measuring instruments including both modern and obsolete samples with different metrological characteristics, MTBF and lifetime are presented. It is shown that by dividing all samples of measuring instruments into risk classes and carrying out verifications with their close to optimal periodicity and with their tolerance for controlled parameters in each risk class it is possible to minimize the total average risk for the fleet of measuring instruments and at the same time to achieve resource saving.
