I have long thought the only way we can create cities that are economically successful, great places to live, sustainable and resilient, is to treat them as an integrated system of systems.
The problem is that cities are so complicated we must break the operations down into sub-systems to get our minds around the problems. We set up different departments to handle transport, waste, planning, social services etc., and we give each one a budget, targets and a team. This is very sensible and practical but creates silos that make it difficult to manage across the different city systems.
Models need to talk to each other
One response has been to build single integrated models of cities with all of the key systems built into them. It is an understandable approach, but it tends to fail. Each team collects its own data and runs its own models. Because they need the flexibility to meet their own objectives, centralised models will not work for them. Any master model will quickly become out of date, failing to include the latest thinking about the different city systems.
What is needed instead is a way of allowing different independent models to interact with each other.
A few weeks ago I visited Space Syntax to learn about their Tombolo project. A collaboration with the Future Cities Catapult funded through the Innovate UK programme – An Integrated Future For Cities. The purpose of the programme was to develop ways of linking different city models and to help users to explore the complex interactions between different systems.
To find out where Tombolo had got to I spoke to Space Syntax directors Ed Parham and Tim Stonor
Integration of data from different sources
Tombolo started with an understanding that it was critical to make data transferable and interoperable between different systems. Get that bit right, and you have the foundation of a truly flexible system. The Digital Connector tool allows data from multiple different datasets in different forms to be pulled into a local repository and converted to a standard structure. It allows you to create a private tailored dataset that is linked to the original sources and kept up to date.
It doesn’t sound very difficult or exciting, but integrating data from different sources has been the bane of city management. The Digital Connector allows you to add any new data sources to your study with a simple ‘recipe’ that describes how you want to handle the new information.
Visualisation of complex and multi-layered data
Helping decision makers understand the complex data involved in solving city problems is a major challenge. Spreadsheets, tables and graphs do not effectively tell the story when layout and geography are so important to how cities function. The Urban Data Explorer, created by Emu Analytics, is a tool that links key information that you are picking up with the Data Connector to geospatial information about the city.
The Urban Data Explorer allows users to play with the layers of data and the styling to create maps that highlight specific themes. A chance to ‘see’ how the city functions.
Modelling the interaction of different systems
The final capability is to generate new models of the way city systems interact with each other. Combining, physical, demographic, economic and health data to explore alternative options. A way of exploring how decisions made in the past affect the practical choices you can make for the future.
But what sort of thing can I model?
The most exciting part of the discussion for me was the practical ways in which the Tombolo tools are already helping cities to think through issues of planning and management.
Ed and Tim talked about how planning decisions to build houses in small clusters around courts and cul-de-sacs were taken to create communities and safe spaces, but increased car dependence and isolation. We are still making that mistake in new developments today, and once the infrastructure is in place, it is hard to change. What small-scale interventions, such as new pedestrian pathways to increase the permeability of the neighbourhood, could improve things?
In another study, they have been looking at the relationship between where older people live, the availability of public transport and the location of GP surgeries, on their health and social welfare. Where are the risks and what could be done to reduce them?
Early studies that are yielding intriguing results.
Are we there yet?
So, have we solved the problem of integrated modelling of cities? Not quite yet. We can integrate, manipulate, visualise and explore data from many different sources. It is a step forward in giving city managers and planners new ways of exploring the interactions between different city systems, new perspectives and challenging some of the assumed rules about how city infrastructure works to deliver for citizens.
But I don’t think we are yet able to link multiple simulation models together so that their outputs are each other’s inputs, allowing their interactions to be directly modelled. I think that is the next step.
Tombolo is an open set of tools for crafting models attuned to the specific challenges of any city, and Space Syntax and the Future Cities Catapult are experts in using those tools to help cities explore alternative futures. Tombolo should be of interest to anyone looking for a way of thinking about complex city problems.