This work focuses on comparing the performance of three well known metaheuristics for determining the best berthing sequence in terms of minimization of total penalty cost when allocating a fixed number of vessels in a bulk port, located in Santa Marta (Colombia). The above problem is related to the well-known berth allocation problem (BAP) which is NP-hard in nature. Therefore, it is hard to find an optimal solution within reasonable computational time even for small instances which, in general, does not reflect the industrial reality. For that reason, in this work we studied the use of different metaheuristics as an alternative and suitable way of finding high quality solutions within short computational times. Computational tests are carried out with the aim to analyze and compare the quality of solutions based on the penalty function.