Fatigue is a complex concept that is simultaneously a physiological, psychological, and social phenomenon which pathophysiology is still not fully understood. Solutions offered by modern technologies are also still unable to prevent economic losses, but even more importantly not to lose human resources in high-risk sectors such as transport and operations. If the fatigue topological model (TM) is created in the form of a directed graph, the analysis provided by such a model in the main problem solution on the level of expression of human fatigue remains to be performed. First, the model makes it clear that in practice the often-combined terms fatigue (F) and drowsiness (D) are neither the same nor the opposite in nature. The graph configuration will clearly indicate the routes belonging solely to F and belonging solely to D, as well as the routes common to F and D. Secondly, based on the graph theory, a formal algorithm enables a selection task to be performed to select the most efficient set of parameters to measure fatigue. The method of selection is derived from the so-called table overlay method. In practice, this method is justified because there are no parameters that can be measured only under laboratory conditions, or the sensor is inconvenient to perform the work. The resulting parameters selected from TM can be obtained with one touch sensor placed on the head. The rest of the information comes from a remote camera, test results and self-survey data.