With regard to a long horizon of the forecasts built, among others, for the needs of foresight deep changes should be expected in the area of the considered phenomenon. In this connection, the variability of the surrounding, i.e., economic, political, legal, social and technological situation as well as the natural environment should be taken into account in the process of creating a long-term forecast. This problem can be partially solved by means of constructing variant forecasts that take into consideration the adopted scenarios of the surrounding development. However, in such a case, we obtain merely point forecasts that, from the point of view of the construction of development scenarios, may be insufficient. The analysis should be enriched by means of forecast intervals, which would take into account the changes in the development of the considered phenomenon with regard to the change in the level of key factors. The purpose of the article is the presentation of the way to build long-term point forecasts that would take into consideration the changes in the surrounding of the analysed phenomenon. The author compares the intervals of forecasts obtained by means of standard forecast uncertainty, key factors aggregation and simulation analysis.