Objective Aiming at the medial prosthetic loosening failure and lateral cartilage degeneration after unicompartmental knee arthroplasty (UKA), the effects of the prosthetic installation errors of joint line in UKA on knee contact mechanics and kinematics during different activities were studied using musculoskeletal multibody dynamic method. Method On the basis of the neutral position of the medial natural joint line, six installation errors of joint line including ±2 mm, ±4 mm and ±6 mm were considered respectively, and seven musculoskeletal multibody dynamic models of medial UKA were established to comparatively study the variations in knee contact mechanics and kinematics during walking and squatting simulations. Results The prosthetic installation errors of joint line in UKA had significantly effects on the knee contact mechanics during squatting and the swing phase of walking. At the 70% of walking gait cycle, compared with the neutral position, the medial prosthetic contact force was increased by 127.3% and the contact force of the lateral cartilage was decreased by 12.0% under a +2mm joint line condition, the medial prosthetic contact force was close to 0 N but the lateral cartilage contact forces were increased by 10.1% under a -4 mm joint line condition. The tibiofemoral total contact forces were increased by 19.7% and decreased by 14.2% under the +2mm and -2mm joint line conditions, respectively. During squatting, the changes in the medial prosthetic contact force and tibiofemoral total contact force were more significant along with increased installation error of joint line, while the change in the lateral cartilage contact force was not marked. At the 100 o of knee flexion during squatting, compared with the neutral position, the medial prosthetic contact force and the tibiofemoral total contact force were increased by 31.6% and 11.1% under a +2mm joint line condition, and decreased were 24.5% and 8.5% under a -2mm joint line condition, respectively. Moreover, during the swing phase of walking, the varus angle was decreased, the internal rotation and the anterior translation were increased along with the elevation of joint line in UKA, while it was just the opposite along with the reduction of joint line in UKA. The trends of the varus-valgus movement and anterior-posterior translation during squatting were consistent with those during the swing phase of walking, but the trend of the internal-external rotation was opposite. Conclusions In order to reduce the risk of medial prosthetic loosening failure and lateral cartilage degeneration, it was recommended that the installation error of joint line in UKA should be controlled in the range of -2 mm to +2 mm. This study provided a theoretical basis for UKA clinical failure caused by the changes in joint line.