Friction on road and runway surface plays a very important role in the safety of road traffic and aircraft taking-off and landing. Loss of friction or skid resistance is closely related to incidents of accidents on the road and at the airport, especially in winter conditions when the surfaces are covered by ice, frost, snow or water. In theory, the friction coefficient (µ) is defined as the ratio of the resistive forces (Fr), a result of the movement between two opposing object surfaces, to the normal force (N) exerted on the surface. In reality, however, the coefficient is complicated and is affected by many factors, such as: tyre pressure, rubber composition, tyre tread configuration, pavement surface material composition, pavement micro- and macro-texture, pavement surface temperature and surface state etc. Such complexity was dealt with in this paper by introducing the concepts of surface texture and friction index (0.0-1.0).

In this paper, we present a numerical road weather and friction index forecast model and results of verification of that model. The data for verification was obtained from Helsinki International Airport. The model was run up to 3 hours ahead, in hourly automatic modes with forecasting output at 10-min intervals. Verification of the model considered both the meteorological variables (pavement surface temperature, water depth on the surface, air temperature, dew point etc) and the friction index. The comparison between forecasts and measurements, with thousands of samples, showed that both bias and RMS errors of the meteorological variables and friction index forecasts were suitably small; with the overall friction forecasting bias being slightly negative and the RMS error being 0.32. In the case when the friction index was <0.3 and driving, landing and taking off conditions became risky, the friction forecast error fell to less than 17% of the overall forecast error.