An improved cuckoo search algorithm for constrained optimization problem is proposed to improve the convergence accuracy and convergence speed in solving constrained optimization problem. Firstly, the shortcomings of global search and local search in the basic cuckoo search algorithm are analyzed, the global search and local search are redefined, and then the search is carried out in the vicinity of the optimal solution with a certain probability. The 12 standard constraint optimization problems and 4 engineering constraint optimization problems are tested and compared with a variety of algorithms, Experimental results and statistical analysis show that the proposed algorithm is superior in solving constrained optimization problems.