Rectangular tunnel boring machine (TBM) is applied for the tunnels’ excavation including a cross section of circular and
rectangular shape within the various rocks and soil strata. Excessive structural forces, which is produced within tunnel linings,
might affect the serviceability and safety of tunnels whose forces acting on tunnels linings during the initial design period
have to be accurately calculated. Few numerical studies have been conducted on different soil-rectangular tunnel systems to
clarify the critical response characteristics of cut-and-cover (rectangular) tunnels adjusted to transversal ground shaking. In
this case, predicting the soil dynamic shear stresses developed around the tunnel is an elaborate task due to the interaction of
TBM in the rectangular form and the rock. Despite doing the empirical studies in analyzing the rectangular tunnels systems,
using artificial intelligence (AI) methods could significantly develop the optimization of TBM tunneling and decreasing the
cost, error percentages, disturbance, and time-consuming complications related to tunneling. In this study, two algorithms,
namely, support vector machine (SVM) and gene expression programming (GEP), were used to accurately predict the soil
dynamic shear stresses developed around the tunnel. The models were developed and measured resulting that SVM can
indicate a high-performance capacity in predicting the soil dynamic shear stresses developed around the tunnel through the
rectangular TBM excavation machine