Optimization of building thermal systems is treated in the paper in the framework of fuzzy mathematical programming. This new approach allows to formulate more precisely the problem which compromises energy saving and thermal comfort satisfaction under given constraints. Fuzzy optimization problem is solved analytically under some assumptions. An example illustrates the viability of the approach proposed. A solution which significantly (with 38%) improves comfort is found which is more energetically expensive with only 0.6%. (c) IFSA