Maintaining high operation performance and at the same time having high transporting capacity is a challenge for urban rail lines during unexpected disruption. In this paper,we focus on adjusting the train timetable in case of a complete blockage with fixed disruption length. A mixed integrated linear train rescheduling model, where short-turning, cancelling and adjustable dwell time and running time are included, is proposed to minimize total train delays and services cancellation under different dispatchers reaction time towards disruption. To evaluate the rescheduling impacts on passengers, we later establish an accurate passenger flow model considering time-dependent passenger ODs. The main objective is to minimizing passenger travel time. A case study is carried out for the disruption scenarios on Beijing Metro Line 7. It is illustrated that optimal rescheduling solution is sensitive to reaction time. Shorter reaction time mitigates train delays as well as passengers'travel time. Longer reaction time increases the numbers of affected train services during the disruption and make it more difficult recover to original timetable.

The increasing use of the private vehicles makes it increasingly fundamental to understand how to induce users to prefer public transport. In this regard, studying the quality of public transport and the methods to increase them is central. There are several factors involved in the definition of the quality level of a public transport system: reliability, regularity of the service, crowding level, on-board comfort, etc. In this paper, the research is focused on the analysis of on-board comfort, in particular, on the comparison between the level of comfort of two different public transport systems: metro and bus. The investigation of on-board comfort has been carried out based on the collection of kinematic data (accelerations along three axes), through a sensor installed inside public vehicles. The acceleration values were analysed according to the ISO 2631-1. The aim is to identify which of the two transport systems is the most comfortable and to identify strategies for improving the level of comfort.

Traffic performance has many positive and negative consequences to the environment and society. These external effects are ever more often considered in the traffic system planning and administration. Desired effects of traffic can be thought as traffic performance policies. It is also possible to support these policies through traffic management and traffic signal controllers. In this study we introduce a general framework for a process flow which allows signalised junction controllers to adapt into desired policy. Also, we present an example implementation of the processes of the framework, and experiment with it by optimising a signal controller in a microscopic traffic simulation environment.