Planning for Major Rail Closures with the Capacity of London Assessment Model (CLAM)



Planning for Major Rail Closures with the Capacity of London Assessment Model (CLAM)

Authors

Catherine Seaborn, CH2M HILL, Sandra Weddell, Transport for London, Jorge Devesa Gallego, CH2M HILL

Description

The Capacity of London Assessment Model (CLAM) has been developed by TfL to support strategic decision-making, operational planning and customer information and travel advice around major planned rail closures in Greater London.

Abstract

Like many other European cities, the London public transport network is undergoing a period of intense growth and renewal. Two large authorities, Transport for London (TfL) and Network Rail, are jointly responsible for planning and de-conflicting track and station closures related to major engineering works in Greater London. As both passenger numbers and project delivery pressures increase, there is a growing need to assess the combined impacts of major closures on passenger behaviour, and then communicate this back to customers.

The Capacity of London Assessment Model (CLAM) has been developed by TfL to support strategic decision-making, operational planning and customer information and travel advice around major planned track and station closures in Greater London. The aim is to assess closure mitigation scenarios by displaying the combined effects of planned works on network capacity, highlighting areas of over- and under-capacity. This information is then used to inform travel advice and customer communications around the planned closures. This pioneering project has benefited from close collaboration with Network Rail and selected Train Operating Companies (TOCs), and builds on the success of another groundbreaking model developed to support transport planning and operations during the London 2012 Olympic and Paralympic Games. CH2M HILL has provided technical model development support over the past two years.

CLAM is an add-on to TfL’s strategic public transport model (EMME/Railplan). The Railplan model covers all public transport modes in London and represents average conditions for three key time periods (AM, Interpeak, PM) during a typical weekday. The CLAM enhancements add value across four key areas: (1) behavioural responses of passengers to network crowding in both the short- and long-term, (2) granular detail of results down to 15-minutes for all days and seasons, (3) capacity constraints of the network based on real timetables and train station capacities, and (4) direct support to customer advice about the most crowded stations or rail lines and alternative routes.

Recent research around improving strategic models for crowded public transport networks has tended to focus on refining the crowding function. Whilst this is certainly important to get right, the application of the crowding function results in a long-term equilibrium in which passengers have perfect knowledge of crowding conditions. CLAM also reflects short-term behavioural responses of passengers unfamiliar with the network, for example taking the shortest path even though it is overcrowded.

CLAM seeks to meet the differing requirements of multiple stakeholders both within and outside TfL as they aim to improve planning, operations and customer service around major closures and events. As such, the project enables a step change in the application of models for transport planning within TfL and partner organisations. A case study at a major rail and Underground hub (London Bridge) is used to illustrate how the model is providing practical support to strategic and operational planning in London, as well as directly informing travel advice to customers.

Model enhancement and validation are integral to the project. Modelled results are compared to observed data from closures that have already taken place in order to improve the model for future planned closures. A key challenge is to separate changes in passenger travel behaviour resulting from different influences, for example advance customer communications, seasonal demand suppression, and the customer’s own experience.

The paper concludes with a review of the extent to which the model enhancements meet the needs of TfL and its partners. Following on the initial development phase, the future success of CLAM depends on a deep understanding of how CLAM sits within the organisational structure and ”business as usual” operations of the authorities that intend to use it. The paper should be of particular interest to modellers, planners and decision-makers involved in planning for disruptions to urban rail services in crowded metropolitan areas and in providing related travel advice to customers.

Keywords: modelling, public transport, planning, travel behaviour, customer service

Publisher

Association for European Transport