In this study, a method is proposed to analyse the capacity improvement ability of using virtual coupling technology to the sharing-corridor metro lines for solving the problem of capacity bottleneck. A mixed integer linear programming model is developed to optimise the train timetable of sharing-corridor lines under the assumption of using the virtual coupling technology, considering the constraints related to train timetabling, virtual coupling technology, and passenger assignment, with the objective to improve the service quality. The model adopts the departure times, arrival times, and whether each train service is planned as virtually coupled, with its adjacent front one or the following one, as decision variables. Furthermore, a method is proposed to evaluate the capacity utilisation referring to UIC 406 and considering the feature of virtual coupling technology. The experiments on Shanghai Metro Line 3 and Line 4 are implemented to verify the effectiveness on transport capacity and passenger service quality of the proposed method. The results show that using virtual coupling technology can launch eight more train services during morning peak hours and at most reduce passengers’ waiting time by 73.8 s. •A passenger-oriented timetable optimisation method is proposed for sharing-corridor lines.•Passengers’ heterogeneous vehicle selection behaviours are considered.•A transport capacity evaluation method considering the features of virtual coupling is proposed.•Numerical experiments based on real-life data of Shanghai Metro Lines is performed.