Robin Lovelace

‘Truncate, replicate, sample’: A method for creating integer weights for spatial microsimulation

Abstract Iterative proportional fitting (IPF) is a widely used method for spatial microsimulation. The technique results in non-integer weights for individual rows of data. This is problematic for certain applications and has led many researchers to favour combinatorial optimisation approaches such as simulated annealing. An alternative to this is ‘integerisation’ of IPF weights: the translation of the continuous weight variable into a discrete number of unique or ‘cloned’ individuals. We describe four existing methods of integerisation and present a new one. Our method – ‘truncate, replicate, sample’ (TRS) – recognises that IPF weights consist of both ‘replication weights’ and ‘conventional weights’, the effects of which need to be separated. The procedure consists of three steps: (1) separate replication and conventional weights by truncation; (2) replication of individuals with positive integer weights; and (3) probabilistic sampling. The results, which are reproducible using supplementary code and data published alongside this paper, show that TRS is fast, and more accurate than alternative approaches to integerisation.

Development of the Impacts of Cycling Tool (ICT): A modelling study and web tool for evaluating health and environmental impacts of cycling uptake

Background A modal shift to cycling has the potential to reduce greenhouse gas emissions and provide health co-benefits. Methods, models, and tools are needed to estimate the potential for cycling uptake and communicate to policy makers the range of impacts this would have. Methods and findings The Impacts of Cycling Tool (ICT) is an open source model with a web interface for visualising travel patterns and comparing the impacts of different scenarios of cycling uptake. It is currently applied to England. The ICT allows users to visualise individual and trip-level data from the English National Travel Survey (NTS), 2004–2014 sample, 132,000 adults. It models scenarios in which there is an increase in the proportion of the population who cycle regularly, using a distance-based propensity approach to model which trips would be cycled. From this, the model estimates likely impact on travel patterns, health, and greenhouse gas emissions. Estimates of nonoccupational physical activity are generated by fusing the NTS with the English Active People Survey (APS, 2013–2014, 559,515 adults) to create a synthetic population. Under ‘equity’ scenarios, we investigate what would happen if cycling levels increased equally among all age and gender categories, as opposed to in proportion to the profile of current cyclists. Under electric assist bike (pedelecs or ‘e-bike’) scenarios, the probability of cycling longer trips increases, based on the e-bike data from the Netherlands, 2013–2014 Dutch Travel Survey (50,868 adults).Outcomes are presented across domains including transport (trip duration and trips by mode), health (physical activity levels, years of life lost), and car transport–related CO2 emissions. Results can be visualised for the whole population and various subpopulations (region, age, gender, and ethnicity). The tool is available at www.pct.bike/ict. If the proportion of the English population who cycle regularly increased from 4.8% to 25%, then there would be notable reductions in car miles and passenger related CO2 emissions (2.2%) and health benefits (2.1% reduction in years of life lost due to premature mortality). If the new cyclists had access to e-bikes, then mortality reductions would be similar, while the reduction in car miles and CO2 emissions would be larger (2.7%). If take-up of cycling occurred equally by gender and age (under 80 years), then health benefits would be marginally greater (2.2%) but reduction in CO2 slightly smaller (1.8%). The study is limited by the quality and comparability of the input data (including reliance on self-report behaviours). As with all modelling studies, many assumptions are required and potentially important pathways excluded (e.g. injury, air pollution, and noise pollution). Conclusion This study demonstrates a generalisable approach for using travel survey data to model scenarios of cycling uptake that can be applied to a wide range of settings. The use of individual-level data allows investigation of a wide range of outcomes, and variation across subgroups. Future work should investigate the sensitivity of results to assumptions and omissions, and if this varies across setting.

Barriers to investing in cycling: Stakeholder views from England

Background Planners and politicians in many countries seek to increase the proportion of trips made by cycling. However, this is often challenging. In England, a national target to double cycling by 2025 is likely to be missed: between 2001 and 2011 the proportion of commutes made by cycling barely grew. One important contributory factor is continued low investment in cycling infrastructure, by comparison to European leaders. Methods This paper examines barriers to cycling investment, considering that these need to be better understood to understand failures to increase cycling level. It is based on qualitative data from an online survey of over 400 stakeholders, alongside seven in-depth interviews. Results Many respondents reported that change continues to be blocked by chronic barriers including a lack of funding and leadership. Participants provided insights into how challenges develop along the life of a scheme. In authorities with little consideration given to cycling provision, media and public opposition were not reported as a major issue. However, where planning and implementation have begun, this can change quickly; although examples were given of schemes successfully proceeding, despite this. The research points to a growing gap between authorities that have overcome key challenges, and those that have not.

David Boyce and Huw Williams, Forecasting urban travel: Past, present and future

Urban Travel is an imposing book. Its 600+ pages are written in academic prose, interspersed with detailed quotations, high-quality graphics and a smattering of mathematics, deliberately placed in each chapter’s meticulous endnotes to increase readability. The chapters are ordered roughly chronologically, covering the entire gamut of computational transport forecasting models from the early developments in the US and UK (Chapters 2 and 3) through discrete choice modelling approaches (Chapters 4 and 5), activity-based and network equilibrium approaches (Chapters 6 and 7), the practice of travel forecasting (Chapters 8 and 9) to computational aspects of the field (Chapter 10) and prospects for the future (Chapter 11). The introductory and concluding chapters astutely synthesise these substantial strands of thought into a single narrative.