RELATIONSHIPS AMONG GENETIC GOALS AND HERDING BREEDING EXPENSES

Techniques of linear programming were applied to 400 genetic breeding goals to select semen from 331 sires to achieve or exceed an average of each goal for minimum semen cost. Goals were combinations of average repeatability 30 to 90%, predicted differences milk 0 to 675 kg, fat percentage −.20 to +.20%, and type −.5 to +1.5 points. Sires included all Holstein bulls available July, 1977, from nine major artificial insemination organizations. Eighty of the most restrictive goals could not be satisfied with any combination of sires. Minimum costs per unit of semen ranged from $2 to $53, averaging $5. Number of bulls required to meet each goal ranged from one to five. A typical goal of 90% repeatability, predicted differences of +450 kg milk, 0% fat percent, and 0 type selected semen from five sires at a cost of $3.93 per unit. Only five goals exceeded $15 per unit in cost. Minimum costs to attain goals were affected by linear and quadratic repeatability and milk, quadratic type, and all linear products involving milk and each other measure. With judicious selection of bulls, dairymen need not pay excessively to achieve respectable genetic goals. Semen cost can be minimized further by increasing fat percent and decreasing milk, holding constant predicted difference dollars. © 1979, American Dairy Science Association. All rights reserved.