In this paper, inspired by the biology migration phenomenon, which is ubiquitous in the social evolution process in nature, a new meta-heuristic optimization paradigm called biology migration algorithm (BMA) is proposed. This optimizer consists of two phases, i.e., migration phase and updating phase. The first phase mainly simulates how the species move to new habits. During this phase, each agent should obey two main rules depicted by two random operators. The second phase mimics how some species leave the group and new ones join the group during the migration process. In this phase, a maximum number of iterations will be set to predetermine whether a current individual should leave and be replaced by a new one. Simulation results based on a comprehensive set of benchmark functions and four real engineering problems indicate that BMA is effective in comparison with other existing optimization methods.