The article considers the road monitoring weather-stations which collects raw observations that are processed to be able to make the necessary forecasting for future decisions. For the road maintainers those predictions are crucial to make decisions daily. When it comes to the winter season when road safety is very important; however, the road condition is also affected by the snow and icing. In order to improve safety on the road network the road maintainers are trying to use every possible way to be able to provide it. A number of methods have been studied and compared to clarify the parameter required by Kalman filter, which can be improved by making forecasting more accurate. Several road monitoring weather-stations are merged into one region because they are relatively close to each other and it is assumed that there are common conditions in one region that may indicate changes in road conditions. The corresponding algorithms are applied for each region and then compared to each other. Adaptive Kalman filter is generalized in the relevant article in order to have a general understanding of how to correctly apply the approach. The main result of this article is a comparison with the different methods, which are finally compiled in a single table.