Many factors need to be considered when designing the timetable of intercity railway lines, where the passenger demand is variable and unbalanced along time in both directions of the line. This paper introduces a new multi-objective nonlinear integrated optimization model for designing full-day train timetables incorporating coupling-decoupling operations and rolling stock assignment on a bidirectional railway line. The model explicitly considers train capacity constraints and oversaturation while minimizing the total passenger waiting time and operating costs. To solve the problem, a hybrid algorithm governed by a genetic algorithm is proposed. Before evaluating the fitness of each individual, to complete the definition of each chromosome, a truncated branch and cut algorithm is used to solve a capacity and train unit allocation subproblem. A good-quality initial population is partially generated using the spatiotemporal synchronous coupling algorithm. The proposed approach is applied to a real case using the Shanghai-Hangzhou intercity railway line. The results of different experiments show the effectiveness of the model and algorithm, both in terms of passenger convenience and cost savings.