The ant colony algorithm is a bionic evolutionary algorithm derived from the natural biological world. It has the characteristics of selforganization, positive feedback, strong robustness, and distributed computing, and it is easy to combine with other algorithms. it is therefore of great applied value in the complex combinatorial optimization field. We firstly introduce the theory of the ant colony algorithm and its important parameters. Then we analyze the improvement and applications in parameter optimization and intelligent fusion. Thirdly, we summarize the progress of applications in jobshop scheduling problem, vehicle routing problem, image processing and electric power system optimization. Finally, we discuss the potential problems in research work in theory and application domain, as well as some possible countermeasures.