Good morning. My talk is about smart cities – what are they, how can we create them. And what, if anything, is the role of technology?
Many people think we already have our answer…technology is the answer. But what was the question? Cedric Price asked this in 1966. Has anything changed? I’d like each of you to consider this as we go through today’s programme.
So let’s start to define a Smart City as this will help us define our questions. In my experience there are three fundamental requirements: first, a clear vision of what is needed – what the future looks like second, ongoing financial support, over many decades, to make sure that the strategy is supported in the long term third, capable leadership to make the decisions that matter. Once we’ve established these fundamental requirements we can think about how technology can help with each.
And a Smart City requires the balancing of multiple objectives: social, economic, environmental, technological and political. These objectives do not always pull in the same direction. For example, economic progress might risk environmental damage – unless the right strategy creates an effective balance. So how can technology help us achieve that balance? If you are being pitched a technology, which of these fundamental objectives is it addressing? One? Some? Or none?
But a clear vision, financial backing and strong leadership aren’t guarantees of success. Good intentions haven’t always created places that people want to move to or invest in. Like Skelmersdale. It was designed by the best minds at the time. But they didn’t have the right approach: too car-centric. Too fragmented. Too monofunctional. Skelmersdale is not a smart city and the most powerful computer in the world will not create a smart city unless it is programmed with the principles of good placemaking. The urban planning and design principles that make the difference. So this is what I want to turn to next.
For a city to be truly smart, we need to avoid the fragmented city and, instead, to create the integrated city. To reduce car dependency. To promote active travel and public transport. To overcome loneliness and to create the urban buzz that you feel when you’re in a great place. …to make cities places for people and, in doing so, to make them resilient.
The key to this is spatial connectivity. My organisation, Space Syntax, has developed spatial modelling techniques that measure strong connections in red and orange and weaker connections in blue, based on the efficiency of the street grid. We’ve written algorithms that analyse which streets are more likely to be travelled along as people navigate from A to B - algorithms that more closely follow the workings of the brain than conventional traffic engineering. Whereas traffic models rely on calculations of distance, the spatial layout models that we’ve created also factor in the geometrical complexity of the route. This makes them work much better. [The science of cities that we’ve helped create has produced new techniques to measure the effectiveness of planned connections and avoid the mistakes of the past. Much has been learned from historic cities built before the invention of cars, for example how their streets have been laid out in highly connected grids that support walking, cycling and public transport as well as driving.]
The more we study connectivity the more we learn – for example the Net Present Value differences between a disconnected layout on the left and a more connected version on the right. Connections such as bridges over canals and railway tracks might cost a little more to construct but can deliver a lot more in terms of footfall and trade in the long term.
And before we go too much further it’s worth emphasising how important streets are to how places work. Great streets may seem obviously important – so why aren’t we still building them? Because streets are what make smart villages, towns and cities smart. The Ramblas make Barcelona smart. The Champs Elysees make Paris smart. We need to give the design of a main street a greater importance than the design of a major building. The street itself should be at the heart of a Smart City solution.
But it isn’t just about the local scale of the street. It’s also about how places connect at the larger scale. Fortunately spatial layout modelling allows us to measure local and large-scale connectivity at the same time.
In summary, we can say that spatial layouts have five key influences…
At Space Syntax we combine datasets into Integrated Urban Models, which explain the ways that existing cities work then predict the way they will work if we re-plan them. These models use established GIS platforms but, crucially, they incorporate a Spatial Layout layer that makes sense of so much data.
Armed with this knowledge we can help to address the UK’s housebuilding challenge, designing urban extensions that integrate with existing places. Spatial accessibility modelling provides an evidence platform to test proposals on. As in this Spatial Layout Model of Didcot. The existing movement hierarchy is picked out by the model.
Working with Savills Urban Design, Space Syntax helped to design a street layout that follows the historical urban tradition: a few main streets for the shops and a hierarchy of medium to less well connected other streets for non-retail and residential uses. High movement on the main streets where we want it and sufficient movement on the side streets to provide “natural surveillance” that facilitates safety. A new centre to the west of Didcot at the intersection of key movement routes.
And being careful not to create unwanted movement through places where it isn’t wanted such as this quiet village.
By toggling between before and after analyses we can see that the spatial hierarchy is barely altered – the only real difference being the creation – deliberately – of a walking and cycling route northwards out of the village.
Moving to a different project, the recently opened headquarters of Bloomberg in the City of London incorporates a street on the ancient alignment of Watling Street.
The idea of the street has been around twenty years in gestation. It emerged from analysis of the surrounding street network. Like the staircase at Trafalgar Square the idea is obvious once it’s drawn. But no one had drawn it until we did. And then spatial modelling helped convince decision-takers that it is an idea worth having. As well as working with individual developers on individual sites, Space Syntax has also constructed a Pedestrian Movement Model for the whole of the City of London. In a few years I would expect any serious economic centre to have a Pedestrian Movement Model.
Closer to home, Space Syntax has been using an Integrated Urban Model in Milton Keynes to measure car dependency. If you live in a convoluted spatial layout that distances you from public transport then you are reliant on a car. If you haven’t got access to a car you are physically and socially isolated as well as disadvantaged in terms of accessing employment.
By running Age UK’s social isolation model through the Integrated Urban Model it’s possible to see the correlations between spatial isolation and social isolation and – in so doing – to provide the hard evidence on the social effects of poor planning that have such profound impacts on mental health.
We’ve then tested the proposed future plans for Milton Keynes…
… and been able to highlight where these exacerbate car dependency.
In Greenwich we’ve used an Integrated Urban Model to analyse the quality of the care provided by GP surgeries – producing a more nuanced and sophisticated analysis than simple catchment analysis. It’s all about connecting digital datasets then exploring the associations and correlations in a highly visual way.
And here in Oxford we have created a Spatial Layout Model that can be used in many different ways. First, as a regional level traffic model, identifying the hierarchy in the main road network based on national and regional movement patterns…
…then at a finer scale, analysing regional and city-scale movement…
…and at the fine-scale of local pedestrian and cycling movement patterns.
…the more we work with this model the more we discover about how Oxford works. First in terms of transport – here we can see the model on the left and, on the right, the travel to work patterns as reported in the last Census. There is a clear link between where you live and whether you travel to work in a vehicle…
…and also how healthy you are likely to be – the more central the more your health is likely to be very good… I must say these results don’t surprise us but they do help confirm the importance of spatial planning and the risks of building isolated, fragmented housing developments. I believe that, once we understand the health risks – mental and physical health – we’ll stop doing it.
As well as Smart planning tools you need Smart working methods. If you pair Smart City planning tools with Smart Working processes then you have a rounded approach. What I mean by Smart Working processes are the tried and tested methods of Creative Workshops and Design Review processes - bringing people together to find solutions as a multi-disciplinary team, then scrutinising those proposals through expert review. Employing the best of computational analysis in combination with the best of interpersonal working methods.
We present analysis to inform workshops and inspire discussions…
And from these discussions the first design ideas often emerge…
a discussion prompts an idea…we can then test the idea in the model…
I want to show one last case study from Darwin, Australia, which brings together some of the themes I’ve introduced. Our brief was to create a masterplan to expand the existing city centre. To almost double it.
Smart Working, through many design meetings as well as consultation meetings with residents and businesses.
Smart City analysis of the way the existing city worked. Its patterns of movement…
…its property values…
…its land use pattern…
...its three-dimensional form…
…its transport patterns…
…its parking areas and payment structures…
…the quality of its streets…
…and its spatial connectivity.
These datasets were combined into an Integrated Urban Model that explained the way the existing city was working…
…and then allowed us to test design ideas…
…in a rigorous way that measured the likely impact of the proposals on movement patterns…
…and, importantly, on property values so that we could identify the best layout that created the most valuable future land in the new city, as well as its impacts on the existing city.
This work led to the creation of the masterplan, which is designed around great streets, parks and public spaces for people.
This is the process we followed. It is rigorous but highly creative.
So what is a Smart City? Well, I want to leave you with a mental map. Imagine a box, with width, depth and height: three axes.
The first axis is spatial. Smart Cities should connect physical and digital flows across different spatial scales of activity: a. the macro scale of Urban Planning b. the meso scale of Urban Design c. the micro scale of Building Design We see technology developing rapidly at each of these scales.
Indeed, Integrated Urban Modelling will, I believe, be a key tool in the creation of future places. By using GIS systems to geo-locate data and then by employing “predictive analytic algorithms” to link between different kinds of datasets, Integrated Urban Modelling is already transforming urban planning.
If the first axis of a Smart City is therefore Space, the second axis is Time. Smart Cities should connect through time, across different phases of activity: a. Design (before construction) b. Construction (during construction) c. Operations (after construction)
And it is a SMART process in which we first sense data then map it to make it come alive, then analyse using the Integrated Urban Model. We react to it by creating design ideas, which we then test using predictive modelling. And the first letter of each word in this process - Sense, Map, Analyse, React and Test - spells SMART. This SMART process is, I believe, key to the delivery of successful results as it has been at the Queen Elizabeth Olympic Park in London. The same process of spatial analysis led to the creation of urban planning proposals that connect that new park to its historic surroundings.
And this then gives Smart City practice a clear space/time organisational framework, allowing us to now consider the third and final dimension.
Which is data. And it’s a problem. Even knowledgeable, experienced professionals can be daunted by the abundance of Big Data: the information generated from traffic cameras, pollution detectors, temperature sensors, population censuses, never mind social media feeds, emails and web traffic. How can this great volume of information be handled? How can city officials and private organisations possibly consider all of it? And, if not, which data streams are more important than others?
Well it helps to have a data structure. We find it helps to organise data into four categories: a. People – connecting people with each other; supporting social, economic, personal and professional relationships. b. Urban form and infrastructure – connecting buildings, streets, spaces, bridges, highways, rail lines, air and sea corridors. c. Resources – natural, “extracted” resources such as water, gas and minerals as well as artificial, “created” resources such as data, finance, power and machines including cars and computers. e. Environments – connecting human interventions to the natural environment; incorporating the effects of climate, topography and geology. Each of these datasets is generated in and channeled along physical and digital connections in which interactions occur. Cities are “smart” when flows are smooth and when interactions are effective. And Smart City policy should anticipate the crossovers that occur between different kinds of datasets. For example, digital connections like Skype, WeChat and email don’t only enable communications across large distances; they also help people meet later in physical space. And, while people are the focus of any city, a truly Smart City will encourage effective interaction from machine to machine.
So we need to be smart in every dimension. This means that, in considering a Smart City strategy, we need solutions that fit across the entire span of each of the three axes: of space, time and data. There is no single Smart City solution currently available that can do this and nor, I believe, will there ever be. The implication for those people responsible for procuring Smart City strategies is that they need to work with multiple providers of systems and solutions. This means they need to establish clear objectives and clear standards. In my experience, the most important element of a Smart City approach is the establishment of an overall organisational framework. At first this may appear complex since that framework needs to include large-scale planning as well as detailed engineering; construction as well as operations; bricks and steel as well as energy and data flows. And this is where the simple concept of a three-dimensional approach is helpful. It means that Smart City clients can ask Smart City service providers, “Where does your solution fit in my Smart City framework?”. It means that Smart City clients can consider all the solutions they have been offered within the framework and identify overlaps that need managing and gaps that need filling.
In summary, I believe that, in developing technology, we should pursue the three drivers of great property and place: that it should look good, last long and work well. But none of this is new. We may think that digital technology offers new opportunities. And perhaps it does. But it also exists in a context that stretches back into history. At least as far as Vituvius, who anticipated the needs of the Smart City over two thousand years ago, giving us a set of technology drivers that are as fresh today as they must have been at the time. So what is genuinely new? This is the question I'd like to pose and encourage each of you to answer through the rest of this conference. Thank you.
Smart cities Technologies for people places BOB-MK Oxford, 6th March 2018 @Tim_Stonor
"But what was the question?” Cedric Price "Technology is the answer." 1966
The Smart City 3 fundamental requirements Clear vision. Ongoing financial support. Capable leadership. @Tim_Stonor
The balancing of multiple objectives Social Economic Environmental @Tim_Stonor
Skelmersdale is not a Smart City @Tim_Stonor
The fragmented "city" of disconnected developments is not smart. The integrated city of continuously connected neighbourhoods is smart. @Tim_Stonor
Spatial connectivity is the key to a Smart City @Tim_Stonor
Connected layout Disconnected layout High accessibility score Low accessibility score Key discovery Spatial layout influences land value
Key “Smart City design element” The Street
Five key influences of spatial networks Movement Land Use Value Crime Carbon
0. Spatial Layout 1. Movement 2. Land Use 3. Crime & Safety 4. Land Value 5. Carbon Emissions Integrated Urban Modelling
Bloomberg Headquarters, City of London Ⓒ Foster + Partners
Bloomberg Headquarters, City of London Before After
Car Dependence Car is equal to Public Transport Car is 1.0 x better Car is 1.0 - 2.5 x better Car is 2.5 - 5.0 x better Car is 5.0+ x better Car dependence & social isolation
58,000 Dwellings 139,200 @ 2.4 people per household Blanket growth per LSOA based on average predicted growth rate between 2014 and 2039. Infrastructure impact assessment 2050 Growth scenario 35,400 jobs Blanket growth per WPZ based on average predicted growth rate of population between 2014 and 2039.
Key findings Modal split Comparison 150 300 450 600 750 Households Current modal split based on trips to employment
Access to Services GP Surgeries Capacity and Quality Number of GP practices within 15 minutes of walking Mean for Greenwich: 67% < 75% ≥ 75% Overall satisfaction percentage missing data 0.63 – 5 5 – 10 10 – 15 Number of GPs per 1,000 registered patients
Regional scale accessibility National/Regional traffic Oxford Space Syntax Limited © 2018
City scale accessibility Regional/City traffic Oxford Space Syntax Limited © 2018
Local scale accessibility Pedestrian movement Local scale integration Active centres Oxford Space Syntax Limited © 2018
Oxford Space Syntax Limited © 2018 Travel to work Drive car or van Local scale integration Active centres
Health Very good health Local scale integration Active centres Oxford Space Syntax Limited © 2018
A “smart” approach 1. Smart City planning tools Big Data analysis Integrated Urban Modelling Transport, property value, health, safety & social wellbeing 2. Smart Working processes Creative Workshops Design Review The best of computational analysis in combination with the best of interpersonal working methods.
Pedestrian movement Hourly average Weekday Pedestrian movement people per hour 800 - above 500 - 800 200 - 500 100 - 200 50 - 100 0 - 50
Darwin City: $1,525/m2 Larrakeyah: $1,230/m2 Stuart Park: $759/m2 The Gardens: $687/m2 (UCV taken from NT government data in 2011) Building attraction Land values 5,000 to 200,000 2,500 to 5,000 2,000 to 2,500 1,750 to 2,000 1,500 to 1,750 1,250 to 1,500 1,000 to 1,250 750 to 1,000 500 to 750 250 to 500 0 to 250 Unimproved Capital Value (UCV) in $ per m2
Building attraction Land use Community Retail Residential Medical Storage Catering Offices Education Leisure Under construction Industry Parking Ground floor land use Services Hotels Vacant Civic Agriculture Transport Empty or abandoned site Utilities Green open space
Building attraction Building height 20 to above 6 to 9 3 to 5 2 1 10 to 15 Building storeys
Transport attraction Bus routes & bus stops Bus routes Bus stops
Transport attraction Parking zones & spaces Parking space Zone A ($2.20/hr for up to 2 hours) Zone B ($1.60/hr for up to 3 hours) Zone C ($1.10/hr or $6.50 full day) 15-minute zone (Free of charge)
Building attraction Active frontages
Spatial Accessibility High Spatial layout attraction Low
0. Spatial Layout 1. Movement 2. Land Use 3. Crime & Safety 4. Land Value 5. Carbon Emissions Integrated Urban Modelling
Existing land value Land Value Model predicts > $AUD 3.7 billion total land value
Urban Strategy Production Part 1 Urban Data Collective Part 2 Urban Performance Model Part 3 Urban Strategy Part 4 Urban Impact Review #ULITech17 @Tim_Stonor
So what is a Smart City? #ULITech17 @Tim_Stonor
Dimension 1 Space Macro Urban Planning Meso Urban Design Micro Building Design #ULITech17 @Tim_Stonor
Masdar, Abu Dhabi Integrated Urban Modelling Land use Density Layout Space Syntax © 2017
Dimension 1 Space Dimension 2 Time Policy Before construction Planning Design Construction During construction Operations After construction
Sense Map Analyse React Test
Time Sense | Map | Analyse | React | Test Policy | Planning | Design | Construction | Operations
Dimension 3 Data Data structure 1. People 2. Urban form and infrastructure 3. Resources 4. Environments
Time Sense | Map | Analyse | React | Test Policy | Planning | Design | Construction | Operations Data Resources | Environment People | Urban Form The Smart City
Looks good Lasts long Works well Venustas Firmitas Utilitas Vitruvius, 30-15 BC A definition of a Smart City @Tim_Stonor