Finite-element analysis of recycled-fiber reinforced concrete segmental linings for shield tunnels across construction and service stages

Precast segmental linings for shield tunnels experience substantially different loading conditions during storage, tunnel boring machine (TBM) advancement, synchronized grouting, and long-term service. This study presents a comparative finite-element assessment of conventional reinforced-concrete segments (CRCS) and recycled-fiber reinforced-concrete segments (RFRCS) incorporating 0.3% by volume of recycled tire-derived steel fibers together with conventional reinforcing bars. A shield-tunnel lining with an internal diameter of 5.5 m, external diameter of 6.2 m, thickness of 0.35 m, and ring width of 1.2 m was analyzed using Abaqus. Three construction-stage conditions storage loading, TBM jack thrust, and synchronized grout pressure were investigated, while service-stage behavior was evaluated using homogeneous-ring and beam-spring representations. Displacement magnitude and maximum principal strain were used as the primary comparative response indicators, while von Mises stress and reaction force were reported as supplementary numerical outputs. Across all five numerical analyses, RFRCS consistently exhibited lower displacement and maximum principal strain than CRCS. The calculated relative displacement reductions ranged from 60.0% to 70.1%, while the corresponding reductions in maximum principal strain ranged from 55.6% to 63.2%. The reported peak von Mises stress and reaction-force responses remained approximately unchanged between the two systems. The finite-element procedure was benchmarked against full-scale tunnel-segment experiments, with displacement differences of approximately 4.5%, 11.0%, and 6.8%. These comparisons support the adopted procedure for global deformation and elastic response. The consistently lower displacement and maximum principal strain predicted for RFRCS across all five analyses indicate a numerical trend toward reduced deformation under both construction and service-stage loading within the adopted modelling assumptions. However, the validation does not extend to crack propagation, residual tensile behavior, fiber pull-out, or post-cracking response, which were not explicitly simulated.