Abstract: On the basis of establishing the multi-objective decision-making model for partner selection, a self-adaptive genetic algorithmwas proposed to optimize partner selection in a supply chain alliance, and an example was given. In this genetic algorithm, each member in a supply chain alliance has a self-adaptive ability to changes in environment in the genetic process through designing crossover probability and mutation probability. The example result shows that in the genetic algorithm the optimum global solution can be gained and the convergence speed is fast.