In view of the problems like fewer samples and various influencing factors in the prediction of complex equipment costs, the defects in the previous grey relational degree are analyzed and a new grey correlation model of influence factors and cost is established by using improved convex grey correlation. The size of the convex grey correlation degree and the proximity are used to screen the cost driving factor, and a multiple-regression model is established based on the chosen driving factors. Compared with the existing methods, the proposed method has higher prediction accuracy, which means the improved convex grey correlation can diagnose key factors of complex equipment cost.