Artificial insemination constitutes the main method of pig breeding in Europe. According to literature, statistical models which allow prediction of productive capacity can be used as additional tools which may be helpful in making selection decisions concerning breeding boars. The aim of the research was to evaluate the influence of boar's age, standardized daily growth rate, standardized meat content in the carcass, the number of nipples, the age when the boar starts to be used for breeding, the number of semen collections per month, and the interval between semen collections on selected qualitative and quantitative parameters of the semen and to calculate multiple regression equations, which would allow prediction of productive capacity of AI landrace boars. The material for the analysis comprised 47596 ejaculates obtained over the period 1995-2014 from 306 Polish Landrace boars. The obtained results show that semen parameters of the examined boars depend on physiological and organizational factors included in the analysis. It was observed that the following factors, measured with the number of produced semen doses, were of particular significance for the efficiency of use of the boars: boar's age, the daily growth rate, meat content in the carcass, as well as the number of collections and the interval between collections. The conducted statistical analysis points to the possibility of predicting semen parameters accurately using regression analysis. The models are characterized by very high accuracy, which points to their potential for application. The calculated determination coefficients for the multiple regression equations assumed the R 2 value of above 0.92. The highest value of the determination coefficient (R 2-0.99) was obtained for the percentage of sperm cells showing progressive motility. The calculated regression equations may be used in the management of the stock of breeding boars and when making selection decisions.