A stochastic computer simulation model was developed to evaluate the effects of selection strategies based on heifer pelvic area and sire birth weight EPD on dystocia in first-calf heifers. Parameters and relationships used were taken from the literature. Unrelated sets of yearling heifers were simulated. Replacement heifer selection was based on breeding the top 40, 60, 80 or 100% (i.e., no selection) of available heifers as determined by yearling pelvic area (YRLGPA). Selected heifers were bred to bulls that were selected based on EPD for birth weight (SEPD). Levels of SEPD were -2.7, -1.8, -.9, 0, .9, and 1.8 kg. Simulated responses included precalving pelvic area, calf birth weight, calving difficulty score (1, 2, 3, 4), and incidence of dystocia (0, 1). Both measures of dystocia were predicted from the ratio of precalving pelvic area to calf birth weight. Simulated data were analyzed using least squares procedures. The effects of SEPD were significant (P < .05) for birth weight, calving difficulty score, and the frequency of dystocia. Differences among levels of heifer selection for YRLGPA were significant for YRLGPA, precalving pelvic area, birth weight, and calving difficulty score but not for the frequency of dystocia. Based on partial regressions calculated from weighted replication means, a 1-kg decrease in SEPD will decrease average calving difficulty score by .12 units and decrease the frequency of dystocia by 4%, whereas each 10% increase in selection pressure on YRLGPA will reduce the average calving difficulty score by only .02 units and have no detectable effect on the frequency of dystocia.(ABSTRACT TRUNCATED AT 250 WORDS)