Measuring Transport Connectivity — a useful tool or a new burden?

IT WAS REFERRED TO as having been significant in the preparation of the National Infrastructure Commission’s autumn publication of its proposed transport priority projects, an award was presented for the Department for Transport’s work on developing a connectivity measurement tool just before Christmas, and now it has been highlighted by the DfT as an important new requirement in the next round of local authority bus service improvement planning.

The last few weeks have seen the emergence of renewed interest and activity in the field of transport connectivity analysis.

Measuring transport connectivity - and the accessibility it provides - is not an entirely new concept, but it took strides forward as part of the digital revolution about 30 years ago. What appears to have been kindling a renewed interest in the subject is the ever more sophisticated digital data analysis that can be used to produce displays of actual and relative connectivity for different places, and identify gaps in it which inhibit economic development, social cohesion and the opportunities available to individuals for personal access to facilities, services and employment. And to judge which interventions are cost-effective.

Isochrone mapping has been used since analogue days to show catchments and travel journey times for both service planning and facility location decision making by identifying catchments and potential users.

But accessibility, as distinct from mobility, emerged as a significant part of transport thinking in the final couple of decades of the last century. One of the key contributory developments was the emergence of geographical information systems that made it possible to track transport and travel movements and plot them to show patterns and problems.

Accessibility analysis thus became a powerful tool for planners to assess how mobility changes could enable people to travel more easily to various destinations and what are the barriers to better connectivity. One early application of related thinking was in the TRICS system of trip generation analysis. First launched in 1989, the Trip Rate Information Computer System offered a new and important part of the Transport Assessment process, and has expanded into a comprehensive database of traffic and multi-modal transport surveys, covering a wide range of development types.

The system allows its users to establish potential levels of trip generation for their development scenarios using a series of database filtering processes, and it is widely used by both transport planning consultants and local authorities to audit Transport Assessments. TRICS was valuable for providing quick ‘what if’ projections of the impacts of particular new types of developments on the transport system, so appropriate measures could be taken to cater for, or mitigate the consequences.

Other applications of greater understanding and computer firepower on how the transport system and elements of it perform, and the access they do and do not provide, were also being developed around the same time, including the innovative PTAL approach that was developed within Transport for London.

The Public Transport Accessibility Level (PTAL) calculation was originally developed by the Borough of Hammersmith and Fulham in 1992, and was later adopted by TfL in 2004 as the standard method for calculation of public transport access in London. It is not yet commonly used outside Greater London or the south east of England, however, though Transport for Greater Manchester meanwhile developed the Greater Manchester Accessibility Level (GMAL) method. This is similar to the PTAL calculation but uses “crow flies” distances to Service Access Points. It also includes an additional score for “Local Link” demand responsive transport services.

PTAL is a simple, easily calculated approach that hinges on the distance from any point to the nearest public transport node, and service frequency at those points. This accessibility index (AI) can then be converted to a PTAL grade from 1–6 (including sub-divisions 1a, 1b, 6a and 6b), where a PTAL of 1a indicates extremely poor access to the location by public transport, and a PTAL of 6b indicates excellent access by public transport. TfL also have software to calculate PTALs across wide areas using GIS and timetable data, the typical result being a map with coloured bands relating to PTAL grades. TfL introduced the WebCAT automatic calculator in 2015.

The PTAL is used as a development planning tool in London, to determine both permitted parking standards and development densities. Large site developments (those the London boroughs refer to the Greater London Authority) must follow planning guidelines that allow more parking in areas with low PTALs (i.e., poor public transport) and vice versa – and that also relate the allowed density of development to PTAL (i.e., areas with better public transport may have higher density housing or offices).

Other geographical computer-generated display developments have similarly made the opportunity for ‘heat mapping’ of transport data both a possibility and an exciting visual presentation of what previously was usually held as tabulated material of little interest beyond those schooled in numerical data use.

Systems like PTAL offer an obvious indication of the density of public transport provision in an area, but have some shortcomings in not taking into account where services actually go to – for example, a bus that runs every ten minutes to the bottom of the road is considered better than a bus that runs every twelve minutes to the city centre. The use of cut-offs to exclude more distant service access points similarly underestimates the ability to access locations just outside those cut-off distances. For example, a point 960m from a rail station could have a PTAL of 6, whilst a point 961m from the same station could have a PTAL of 1 or 2.

More detailed accessibility modelling has been proposed as a solution to these problems. It uses GIS to calculate door-to-door travel times by public transport to a grid of points around the point of interest, resulting in a set of isochrone journey time contours within which the number of workplaces, households or residents can be calculated using census data. This method takes into account many more factors than PTAL, but is much more time-consuming and requires a level of expertise with GIS software and methodologies.

Not surprisingly, potential improvements to, and wider applications of PTAL-like thinking were occurring to those working in other parts of the transport system, especially those with access to new, much more rich and fine grained sources of data, and the ways of computing and presenting them.

Innovative work was also being done on data mapping and presentation in a number of UK and international consultancies and research bodies like the Centre for Advanced Spatial Analysis (CASA), an interdisciplinary research institute focusing on the science of cities within The Bartlett Faculty of the Built Environment at University College London (UCL). CASA was established in 1995 to lead the development of a science of cities drawing upon methods and ideas in modelling and data science, sensing the urban environment, visualisation and computation.

The Department for Transport’s own connectivity tool’s development was first mentioned by the DfT in 2014 in a bulletin it issued about New ‘Connectivity’ statistics and travel time data. This said “The Department has developed a new statistical dataset and indicators which aim to measure transport ‘connectivity’ in a nationally consistent way. We have published experimental statistics and are seeking feedback from potential users on usefulness, the approach and potential further developments and applications.”

The project would feature new experimental statistics presenting travel time data measuring connectivity to key transport destinations (major airports, stations, road junctions and ports) from each area of England, by road and by public transport.

“The statistics are based on a large dataset of estimated travel times from each neighbourhood in England (origins) to a set of major transport network access points (destinations) – airports, larger stations and major road junctions – separately for travel by car and public transport, and for peak and off-peak periods’, DfT explained. “Conceptually, this data is equivalent to results of millions of journey planner queries. Although these statistics focus on transport-related destinations, the methods could easily be extended to other types of destination (for example major trauma centres or distribution centres), and to time periods and geographies other than those shown.”

The Bulletin said potential applications could include analysing travel times, visualising patterns at national and local level, calculating summary statistics, validating transport (or other) models, comparing destinations, calculating travel time catchments, and measuring local access (ease or difficulty of getting to a particular destination).

It explained that these connectivity statistics are a development of the established accessibility statistics which DfT has published for many years, and which provide a local-level measure of the availability of transport to key local services such as schools, food stores and GPs. The content of these statistics was currently being reviewed.

“Although this data has been commissioned by DfT, there are software packages, used for accessibility planning, which enable similar calculations to be performed and ‘what if’ type analyses to be carried out” the 2014 Bulletin noted. “We are currently exploring the use of these.”

It seems, in the event, the Department pressed on with its own work and revealed in 2021 that it was reviewing the statistics it produced for journey times to key services, covering food stores, education, health care, town centres and employment centres. Its Connectivity model had been designed to be used for both monitoring and appraisal purposes to understand the impact of policy interventions. It aimed to calculate a connectivity score for all of the geographic areas covered by the Journey Time statistical series, based on the purpose of travel, time of day and mode. The connectivity score was expected to range from 0 to 100 where 100 represents the most connected area. Within the model, ‘Connectivity’ was defined as ‘someone’s ability to get to where they want to go’.

The model as it now exists aims to provide a broad assessment of how well people can get to places they wish to access for a range of needs and services covering 180,000 output areas in England and Wales. The tool measures how easy it is for people to get to jobs, services, retail and each other. It combines a newly developed methodology, substantial datasets, and is scaled on the cloud to perform trillions of calculations.

The Department has most recently suggested it will utilise outputs from the Model of Connectivity, to monitor progress against its “Grow and Level Up the Economy” strategic aims, a role that was previously fulfilled by the Journey Time statistics series. It was promised in 2021 that “the new Model of Connectivity and its methodology will be made publicly available as soon as is practicable.”

Though such public availability has yet to be forthcoming, The Department for Transport has entered and won the award in the Evaluation and Analysis category in the 2023 Civil Service Awards for the new Connectivity Model.

A DfT spokesperson described the Connectivity Tool to LTT “as an under-development web-based app that generates a national measure of any place’s connectivity across the country.” They said that, using a combination of land use and transport data, the Connectivity Tool will establish a metric that could be used by policy-makers and decision-takers in local authorities, central government and beyond to better understand what connectivity gaps exist and how new transport schemes improve connectivity. The Connectivity Tool was the result of a collaboration between DfT’s data science and planning policy teams, the spokesperson said. “It is still in alpha testing and not yet available for public use.”

The tool’s metric integrates public transport timetables, National Travel Survey data, and comprehensive data on the location of homes, shops, businesses and services.

Users of the Connectivity Tool will in due course be able to access an overall connectivity score, scores for specific modes and purposes of travel, and impact assessment scores. “It is hoped in the future the tool can inform planning, strategy and investment at both national and local levels,” said the spokesperson.

“This new insight is transformational for our ability to measure levelling-up and sustainability” says the citation for the Civil Service Awards 2023 in the Evaluation and Analysis Award category in which the Connectivity tool was successful. Amongst the applications of the tool are measuring the impact of potential infrastructure investments, or identifying suitable locations for housing developments with sustainable travel options.”

Now, most recently, the DfT has opened up a new avenue for the use of connectivity analysis in the area of bus service analysis and planning. As part of its new guidance about further implementation of its bus policy the Government will this Spring ask all Local Transport Authorities to submit Bus Connectivity Assessments (BCAs).

BCAs are described as a process to identify the best bus network to meet bus connectivity objectives for the local transport authority (LTA) area at varying levels of public funding support. They should be completed annually, says DfT, led by the LTA working closely with bus operators, to feed into the LTA’s budget setting process. As part of this process, operators and LTAs will be expected to report on a range of issues, including but not limited to: connectivity via an assessment of the connections provided by the LTA’s bus network between employment, leisure, services and residential locations; the costs associated with maintaining and improving that connectivity; patronage by identifying actions taken to grow patronage; types of Service by setting out percentage and cost of tendered services, numbers of zero-emission buses, DRT services; and innovation by stating innovative approaches to growing revenue; and funding, by specifying annual transport budget and use of specific grants.

Data from the BCAs is intended to help Government understand the impact of recent funding interventions and to determine what, if any, new policy and/or funding interventions are needed, says the new bus guidance. It seems there may have been Treasury insistence that monitoring of the benefits of the substantial sums now being provided for bus support should be undertaken by the DfT, perhaps explaining the £2.2m contract it has awarded to Ernst & Young for assessing the BCA concept and collecting other bus sector management information.

“BCAs will complement the 2024 BSIP, with evidence and information gathered for the BSIP being relevant for the BCA and vice versa,” the guidance says. “The exercise is designed to support the LTA with identifying the best bus network available at varying levels of funding (including more and less than is currently available), plus informing future Government decisions on bus interventions.” These activities will also provide evidence for and feed into the LTA’s local transport plan, says DfT.

It is not clear whether the new Bus Connectivity Assessments are connected directly – or even indirectly – to DfT’s data science and planning policy teams, who have just won the award for the Connectivity Tool they have been working on as mentioned above, though the Department’s Chief Planner says it is currently looking for local authorities with an interest in piloting the Connectivity Tool over the first half of this year.

It is not really yet clear either what the Department is trying to achieve by this new requirement placed upon local transport authorities relating to their bus networks. This after all, is not an assessment based on actual proposals to see if they are value for money, nor to support the implementation of a masterplan approach underpinned by an appropriate available budget. Indeed, what would seem to be significant related promised guidance on determining how to decide on the provision of Socially and Economically Necessary Services (SENS) has not been published – and there is no clear indication when it will be.

It is to be hoped that there will very soon be a much clearer briefing to local transport authorities on how they are expected to approach their end of this project, and that suitable support to the resourcing of the additional tasks associated with it will entail, rather than eating into the limited money already earmarked for the preservation and enhancement of actual bus services.

It is not to say that Connectivity Assessments are not a potentially useful tool to be deployed in the right circumstances – but that there is a risk of drowning in data and the analysis of it, and the expensive support services to go with that, whilst the actual delivery of services continues to be severely squeezed for resources.

Meanwhile the Transport Planning profession should surely be taking a closer interest in this emerging “connectivity measurement’ territory - not something too much talked about to date. Is it useful, are their core principles and practices to apply, and what useful experience has been obtained about it to date?

Peter Stonham is the Editorial Director of TAPAS Network

This article was first published in LTT magazine, LTT884, 23 January 2024.