The purpose of this study was to investigate the effect of the number and the geometry of resorbable screws (RSs; Inion CPS System; Inion Ltd, Tampere, Finland) on the biomechanical stability of the in vitro model with sagittal split ramus osteotomy. The sagittal split ramus osteotomy polyurethane hemimandible (Synbone, Malans, Switzerland) was fixed by 7 osteosynthesis methods after 5 mm advancement of the distal segment (n = 5 for each method): 1TP (1 titanium miniplate and 4 screws), 3RL (3 RSs with linear configuration at the retromolar area [RMA]), 2R1B (2 RSs at RMA and 1 RS at the mandibular body [MB]), 2R1A (2 RSs at the RMA and 1 RS at the mandibular angle [MA]), 3R1B (3 RSs at RMA and 1 RS at the MB), 3R1A (3 RSs at RMA and 1 RS at the MA), and 3R1A1B (3 RSs at the RMA, 1 RS at the MA, and 1 RS at the MB). Values of linear compressive load were measured at 1- to 5-mm displacement of the lower first molar with a 1-mm interval and were statistically analyzed. From 1- to 5-mm displacement, there were significant differences in load values among groups (P < 0.05, P < 0.01, P < 0.01, P < 0.001, and P < 0.001, respectively). When the amount of displacement was increased, the difference in load values between 1TP, 3RL, and 2R1B became significantly prominent. There was a significant difference in total load values according to number and geometry of RSs (P < 0.001). All kinds of geometry with more than 3 RSs were more rigid than 1TP. The 3R1A1B method showed better biomechanical stability than 1TP, 3RL, and 2R1B. In 3 RS and 4 RS groups, fixation in MA (2R1A, 3R1A) exhibited a tendency of better stability than fixation in MB (2R1B, 3R1B). Fixation with 2R1A could provide better biomechanical stability than 1TP and similar rigidity with 3R1A1B.