In the paper  we compare the results achieved with off-line and on-line tuning approaches. In particular, we report the results of an experimental study based onMAX–MIN Ant System (MMAS) for the traveling salesman problem (TSP). In this framework, we analyze the impact on the performance of online tuning of the number of parameters adapted. Moreover, thanks to a social experiment carried out, we observe the effect of the expertise of the implementer on the behavior of the algorithm with on-line tuned parameters. Here, the results of the whole experimental analysis are reported.