Background: The objective of the research has been to provide an answer to the question of what the possibilities of determining the direction of approach of the auditory danger signal emitted by an overhead crane appropriately are. Cases of use and no use of earmuffs (in the passive mode and level-dependent ones) were all taken into consideration. Material and Methods: The auditory danger signal and ambient noise were recorded in an industrial hall. Signals were reproduced at an experimental set-up, using a large number of speakers. Eight speakers for reproduction of the auditory danger signal were placed above a subject’s head. The study participants would indicate the direction from which, according to them, the auditory danger signal was being emitted. Results: The average percentage rate of the correct localization amounted to 75.8% when the overhead crane’s signal wasn’t masked. The presence of ambient noise caused a reduction of the number of correctly identified localization to 66.6%. The use of earmuffs in the passive mode resulted in the worst results (44.5%). There is some improvement when level-dependent earmuffs are used (57.3%). Conclusions: In situations where it is important to identify the direction from which the auditory danger signal generated by the crane’s signaling device is approaching, it is beneficial to use level-dependent earmuffs rather than earmuffs in the passive mode. Correct identification of whether the auditory danger signal generated by the crane’s signaling device is approaching from the left or right side is almost perfect, however correct identification of whether the signal is approaching from the front or back of a person is not always possible. Med Pr 2016;67(5):589–597