Simulation of a beef cattle herd to examine different selection strategies

A stochastic simulation program was developed to quantify differences in economic measures in simulated herds of commercial beef cattle using bulls from sire lines with different genetic trends. Individual animals were processed through six phases of production: breeding, summer grazing, weaning, winter grazing/feeding, calving, and post calving. Breeding values were simulated for 15 traits. Parameter covariance matrices were constructed from literature estimates and adjusted to avoid negative eigenvalues. All genetic effects were normally distributed. Conception and calving difficulty were modeled as threshold traits. Old and open cows were culled in the fall. Cows without a calf after calving were culled. Prices received per kg of output and costs incurred per unit of input were constant over the length of simulation. The model was used to determine the probability of selection for efficiency. Simulated selection was for 40 years. Eighty commercial herds of 350 to 394 breeding cows were simulated. Four groups of twenty commercial herds chose bulls at random from four different sire lines with different genetic trends: (1) increasing yearling weight (GT); (2) increasing index (yearling weight minus 3.2 times birth weight) (ET); (3) no trends (NT) and (4) increasing conception rate liability (FT). Results for the ET line were compared with results for the GT line. Selection from the NT line was compared with that of the FT line. Commercial herds selecting bulls from the ET line made $4,235 more in year 40, on average, than herds selecting from the GT line (P $\u3c$ 0.05). An advantage of the ET herds was that mature cow weight (thus cow costs) did not increase as rapidly as those of the GT herds. Commercial herds selecting bulls from the FT and NT lines lost money, as FT herds lost \$2,856 compared to the NT herds that lost $21,087 on average in year 40 of selection. Differences were caused by increased weaning rates in FT herds