Objective To explore the effects of functional training based on fragmentation time on spinal joint motion and inflammatory cytokines in patients with ankylosing spondylitis. Methods A randomized controlled trial was conducted on 109 patients with ankylosing spondylitis in the First Affiliated Hospital of Henan University of Science and Technology from April 2021 to September 2022. They were divided into two groups by the random number table method. In the control group, there were 37 males and 17 females, the age was (53.51±6.83) years old, the course of disease was (3.21±1.12) years, and the pain sites were peripheral joints in 18 cases, lumbar spine in 14 cases, and sacroiliac joint in 22 cases. In the observation group, there were 36 males and 19 females, the age was (54.13±6.62) years old, the course of disease was (3.33±1.21) years, and the pain sites were peripheral joint in 19 cases, lumbar spine in 15 cases, and sacroiliac joint in 21 cases. The control group was given routine functional training program, and the observation group was given functional training based on fragmentation time. The improvement of symptoms [Bath Ankylosing Spondylitis Functional index (BASFI), Bath Ankylosing Spondylitis Disease Activity Index (BASDAI), total joint Pain (Pain-VAS), and total joint Stiffness (Stiffness-VAS)], spinal joint mobility [scoliosis, modified Schober index (degree of lumbar flexion dysfunction), chest expansion degree, and cervical spine rotation], and inflammatory factors [C-reactive protein (CRP), tumor necrosis factor-α (TNF-α), and erythrocyte sedimentation rate (ESR)] were compared between the two groups. Independent sample t test, paired sample t test, and χ2 test were used. Results After intervention, the BASFI score, BASDAI score, Pain-VAS score, and Stiffness-VAS score in the observation group were lower than those in the control group [(3.87±0.58) points vs. (5.27±0.62) points, (3.17±0.46) points vs. (4.23±0.63) points, (1.27±0.56) points vs. (2.57±0.46) points, (1.73±0.43) points vs. (3.17±0.51) points], with statistically significant differences (t=12.177, 10.017, 13.254, and 15.948, all P<0.05). The scoliosis, modified Schober index, chest expansion degree, and cervical spine rotation in the observation group were higher than those in the control group [(17.36±0.77) cm vs. (13.28±0.65) cm, (4.85±0.18) cm vs. (3.91±0.17) cm, (4.55±0.17) cm vs. (3.61±0.15) cm, (58.73±1.96)° vs. (53.37±1.84)°], with statistically significant differences (t=29.866, 28.020, 30.590, and 14.714, all P<0.05). The levels of CRP, TNF-α, and ESR in the observation group were lower than those in the control group [(11.74±2.17) mg/L vs. (14.01±2.45) mg/L, (29.72±4.16) ng/L vs. (43.15±5.52) ng/L, (19.85±3.22) mm/h vs. (25.72±3.53) mm/h], with statistically significant differences (t=5.123, 14.325, and 9.073, all P<0.05). Conclusion Functional training based on fragmentation time can effectively improve the clinical symptoms and spinal function in patients with ankylosing spondylitis, and reduce the levels of inflammatory factors.