The resistance seam welding is a welding technique that is often used for the creation of the leak-tight welds. It is a technique where the weld is created as a set of overlapping weld nuggets between the cylindrical welding electrodes. The creation of the weld nuggets is depending on many external and internal parameters. Some of the external parameters are pressing force, welding electrodes, electrical current and time, while the internal parameters are electrical resistance, resistivity, and temperature. The main issue with the resistance seam welding is to properly set up internal and external parameters, create a weld nugget of a certain quality, reconfigure the parameters, relocate, and create another weld nugget of the same quality. The modern welding machines must have monitoring systems capable to make decisions and recalibration of the parameters. These systems are very complex and expensive, and as such, non-affordable for many small enterprises. This paper is presenting an effective, temperature-based algorithm for selecting the optimal welding parameters before the welding begins. The algorithm bases on the data from the experimental welding and numeric simulation of the welding process. The verification of the algorithm is done after testing the quality of the welds.