A hybrid chimp optimization algorithm(HChOA) with artificial preference weight is proposed to improve the convergence speed, solution accuracy and local extreme escape ability of the chimp optimization algorithm. Firstly, combined with the actual situation of the ChOA, a new nonlinear convergence factor is designed for balancing global and local search capability. Secondly, the concept of “dissimilarity” and an artificial preference weight which can be described as “tendency difference and repulsion similarity” are introduced into the chimp population to optimize position update rule, enhance the population diversity at the end of iteration and accelerate the convergence speed of the algorithm. Finally, an improved arithmetic optimization algorithm(IAOA) and mixing into ChOA are proposed, we extract some chimp individuals to perform the IAOA optimization strategy to avoid group search stagnation and premature convergence caused by leader falling into local optimum. Through the numerical comparison experiments of 8 benchmark functions in various dimensions and the solution of one engineering design problem, the comprehensive analysis verifies that the HChOA has significant superiority, robusticity and the value of engineering application.