[en] To analyse air crew exposure to cosmic radiation, both experimentally and by means of the calculation with the goal to assure the individual assessment of this type of exposure. The experimental measurements were realised by means of several different equipments, both passive and active, to characterise all components of radiation fields on board aircraft. The calibration experiments in the on-Earth high energy particle reference fields were also regularly repeated since 1993. The most of measurements was performed in the periods between 1991 and 1993, and, subsequently, during 1999. The calculation of air crew exposure level on board was performed by means of the transport code CARI, in the versions 3N (H*(10)-ambient dose equivalent is calculated), and/or 5E (E-effective dose is calculated). The experimental data directly measured in the period 1991 to 1993 were interpreted on the base of a detector energy response, the results of calibration experiments in high energy reference fields and, also, taking into account independent data on particle's energy spectra and their energy deposition in the human body. Complete set of data obtained in the period mentioned was treated using a regression analysis to get the generalised dependence of exposure level on the flight altitude in the period of rather low solar activity. The experimental data directly measured during the series realised during 1999 were interpreted in the same way and they were compared with the data calculated for the same flight profile by means of the code CARI in its version 5E. It was found that the experimental data are in both cases about 20-25% higher than the calculated ones. The experimentally measured ambient dose equivalent rate vary between 4 and 8 μSv per hour for flight altitudes between 10 and 13 km. The code CARI 5E was also used to estimate the annual exposure of the air crew of a small company for the year 1998. Some model assumptions concerning the flight altitudes distributions have been adopted, their influence would compensate the underestimation observed when calculated values are compared with the experimental ones. It was found that, for the average number of flight hours about 450, three maxima can be distinguished in the effective dose distribution. The highest one (about 2.5 mSv) corresponds to the personnel flying frequently on northatlantic flights. The average annual effective dose was 1.54 mSv, when exposures below 1 mSv are omitted, it decreases to 1.85 mSv. Both figures are higher than it is usual for the most of other categories of occupationally exposed persons. We believe that the estimated values of effective doses are correct within ±25%, which fulfill the requirements on a personal dosimetry system. Both calculation and experimental approaches can be improved, we continue our effort in both directions. (author)