Demands of motor traffic have often overridden or generally compromised the quality of cycling infrastructure provision.
With a focus on securing the expeditious movement of people and goods, for decades Councils have concentrated on motor vehicle traffic. But as congestion is increasing, obesity is growing and air quality is being compromised in many of our cities, it is time to start thinking alternatives.
Introducing transport modeling
Transport modelling is a mathematical representation of a transport system used to estimate usage patterns of the road network under certain assumptions formulated as a simulation scenario.
The modelling happens when data is collected and organised as the foundation for later creation of scenarios, i.e. making assumptions about the rationale of people's that condition how simulation evolves.
Classical transport model is built by following consecutive stages:
- Trip generation - how many travellers go to and from a given location.
- Trip distribution - how many people travel between each pair of locations.
- Mode choice – what mode of travel people choose.
- Route assignment - what route do people take when going from one location to another.
In some cases there are datasets covering some of these stages so there's no need to estimate them, but in many cases numbers have to be derived using general formulas that require specific factors to be assumed.
Values for these important factors, reflecting our assumptions, are arrived at either by interpolating global observations onto individuals’ choices or extrapolating data from small samples onto larger population. This is why it is important to work with multiple scenarios using different assumptions to see which combination of factors gives outputs that are closest to real-world observations.
On the other hand when working with speculative scenarios e.g. modelling growth sites or considering new or improved infrastructure, the assumptions used in scenarios giving most promising outputs are the ones to apply for wider policies.
An example could be seeing best results ('best' could mean most concentrated flows or of highest volume) when simulating short distance journeys from growth sites being translated into not only building infrastructure to make these journeys convenient, but also targeting behaviour change projects at populations doing short distance journeys to work.
What are the benefits of computer modelling?
Computer modelling is a collection of services that bring following benefits:
- Developing a structured vision for a network that separately highlights long/medium/short distance commuter routes.
- Understanding of gradients and how to shape a convenient network that bypasses steep hills.
- Highlighting routes that serve accessing rail stations and local commercial centres.
- Deciding where to target behaviour change projects aimed at commuters living in areas that are most likely to benefit from new/improved cycle routes.
- Exposing routes that carry and areas that generate a lot of short distance travel to work that can be easily shifted to cycling.
- Supplementing economic development plans (new housing, retail, enterprise zones, etc.) with plans for active travel infrastructure when knowing what are likely travel patterns from/to these development areas.
Computer modelling can be an amazing tool to help councils design cycle and walking routes in areas where people want to travel. I work with programmes such as MapInfo or Manifold, which are standard for analysis and visualisation of spatial data.
But there are caveats in making predictions whether someone will choose to go by foot or get on a bike.
Propensity to cycle
Transport surveys show popular transport habits between residential and business centres, but what makes someone more likely to get active on their commute?
Distance is an obvious factor.
We know that in some areas most people are unlikely to walk for more than 2 km (1.2 miles), but are most likely to cycle between 2 (1.2 miles) and 5 km (3.1 miles) for their daily commute, so we can create maps based on both short and longer distance commutes between popular ‘journey pairs’ (start and end points). But in some places the picture is more complicated and we need to input other factors into the model.
One council in the hilly North of England asked us to design maps identifying routes with low gradients which they could use to plan cycle routes. To create a visual representation of this through modelling, I mapped what speeds cyclists can reach when travelling on different streets in that area, and in another map I proposed sub-networks made of roads that do not exceed certain gradients and allow for maintaining cruising speed throughout the journey.
We can apply any assumptions we like for building scenarios, for example putting more weight on trips generated from student populations or employees of a predominantly IT-orientated business park, where people sit at their computers all day or residents of more affluent housing estates with many households having access to 2 or more cars.
The possibilities are practically limitless and the actual choice most of the time depends on how the client Council perceives cycling (e.g. is it more of an option for young people or do they want to target aging populations that don't have many opportunities for physical activity), if they want to see active travel opportunities for some specific planned developments or is there any particular theme in their strategic documents that they want to be also represented in the transport model (e.g. supporting multi-modal journeys with people in suburban areas cycling to train stations to get to the metropolitan centre).
We have worked with Newcastle City Council, Northumberland County Council, Sheffield City Council, South Yorkshire Combined Authority, Tees Valley Combined Authority, Transport for Greater Manchester and Liverpool City Region providing these services.
Computer modelling helps us set future cycling expectations and inform development
Computer models are fantastic tools to help us set expectations for the future. Recently a client asked me to find out how future economic and residential development would affect peoples’ likelihood of cycling or walking.
Using client supplied data about size and location of the proposed developments, I created maps of predicated transport looking at different scenarios including intensification of retail land use, housing growth, peripheral growth and realising current land use planning applications.
The results show transport planners how the new developments will affect demand on the road network, how it will grow over time with developments being built in phases and where to put measures offsetting these pressures.
We can use models to assess a route’s feasibility, through making sure there is enough space for all road users and where extra care will be needed to design safe and convenient transitions. Using OS MasterMap or more detailed survey maps we run a process which sweeps through the whole route and checks building to building and kerb to kerb widths.
There’s always an element of surprise with computer modelling.
You never know quite what the mapped result will be when you put the data in and let the simulation run.
One of the main challenges for convincing results is getting other offices within councils where we work to share their information, particularly records of developments that happened since the last census in 2011 and, oftentimes confidential, plans for future developments.
Good computer models need reliable data to bring useful information about the future.
A bright future
In an ideal world, cycling and walking infrastructure precedes the demand and where demand for active travel is not matched with safe and convenient infrastructure - it's quickly fixed with immediate interventions.
Also, inconclusive meetings with stakeholders and steering groups and their conflicting ideas are solved by testing various solutions with flexible methods on the ground.
If tomorrow was the first day of such perfect reality then it should start with curbing the demand for short car journeys and driving children to school, by de-incentivising such behaviours and providing attractive active travel alternatives.
Change to our streets will not come from analysing their issues over and over again but by taking decisive actions that firmly puts us in the saddle of transition into sustainable travel for people of all ages and abilities.