City Science academic and industry presentations (7 minutes each) on emerging tools, techniques, data and models for collaborative work. At workshop on “Finding Connections Between City Systems and Subsystems”, University of Toronto Cities Centre, October 10-11, 2012.
This digest was created in real-time during the meeting, based on the speaker’s presentation(s) and comments from the audience. The content should not be viewed as an official transcript of the meeting, but only as an interpretation by a single individual. Lapses, grammatical errors, and typing mistakes may not have been corrected. Questions about content should be directed to the originator. The digest has been made available for purposes of scholarship, posted by David Ing.
Introduction by Steve Easterbrook
[Mark Ho] [Rhys Goldstein] [Ramtin Attar] [Jose Lobo] [Slobodan Simonovic] [Chris Kennedy] [David Ing (systems)] [Andre Sorensen and Paul Hess] [Seanna Davidson] [Marc-Paul Gauthier] [David Ing (MRM)] [Michael van der Laan]
Mark Ho, Esri, “Esri Geodesign”
- See profile at https://sites.google.com/site/markhogis/cv
Demo of Esri CityEngine
- New product, acquired from Procederal, in Switzerland
- World of geodesign
Main role to create high quality 3D
Can genearate content
- Cars 2, Total Recall backgrounds were built with this
- Done with rules
3D design tool: geodesign
- Interactively change parameters of a complex
Example: 2D building footprint map
- Apply rule to show in 3D
- Zoom into a block
- Rules prewritten or can be written, e.g. footprintRooftop
Standalone product from other ESRI products
Software levels, product is 64 bit
After rule has been applied, can change parameters of a building, e.g. level of detail (low to high), building height, windows types
Can add in street data, change data
Hard surfaces, e.g. empty lot, would like to add development: add some partial data, and then apply a rule for some buildings, the implement scenarios, could change height from 30m to 50m, spacing between buildings, generate a facade, generate some greenspace, generate interiors as well as exteriors
Can also generate reports: floor area ratio, gross floor space
Can generate from scratch
2012 release, can work with the web: City Engine Web Scenes, so that can publish on the web
- Example for the City of Philadelphia
- Bookmarks for navigation
- Can see highways generated with cars, streets with trees
- Building proposed, can be shown to users in different ways, with what’s currently available: two panes with wipe between
- Can change sunlight
[Questions]
Level of materials, e.g. amount of concrete?
- Yes, can design into a rule, then can report
Rhys Goldstein, Autodesk “Design DEVS”
- See profile at http://www.autodeskresearch.com/people/rhys
Would like to present Rob Ford’s transit model in a simulation
Collaboration between programmers
Two steps:
- 1. Form teams to model certain systems, e.g. traffic, power, food supply
- Go away for 2 years, and write independently
- This is easy to do
- 2. Hard part would be to put these together, as a system of systems
DEVS was developed in 1976, the description from Wikipedia is difficult, so the first job will be to make it more accessible
Formula presented as maps
- Flowchart as simulation progresses
Could try simulation with input messages, e.g. someone getting on a subway
- e.g. output, coming out of a subway
Then statistics, e.g. wait time
Collaboration of two DEVS models, e.g. city traffic with ambulance service
Software called DesignDEVS, hope to circulate for free this spring
Ramtin Attar, Autodesk, “Project Dasher”
- See profile at http://www.autodeskresearch.com/people/ramtin
“We shape our buildings; thereafter they shape us” — Winston Churchill
How do we measure sustainability at a building level?
New York Times: “Some Buildings Not Living Up to Green Label”
- Can we measure sustainability, in complex systems?
Gap between design and operation of the built environment
- Have a lot of tools in the design space, not so many in operation
- Design process is only 3% of cost in the entire life cycle of building
Where is most energy being used in a building?
Started Project Dasher
- Combination of a building information model (an intelligent 3D model, manufacturing, assembly, construction), data collection from building control systems, submeters, cubicle sensors
- Dasher is the context within which data is brought it, otherwise it doesn’t make sense
Can break the building into a hierarchy
- e.g. energy usage for floor, cubicle
Goal: visualize a massive amount of data over time
Not just looking at high visualization, combined with HVAC, etc.
Prototype: load onto iPad, then can walk around a building
- Currently some use Google Docs to transfer
Pilot projects include Government of Canada; Netzero building in Cradle to Cradle; NASA
Can we think of a building that gets better of time, as it ages?
For Dasher, can see a trend towards data growing in the Internet of Things, will have trillions of sensors in cities
- CitySense: An Open, urban-Scale Sendor Network Testbed, Harvard U. and BBN Technologies
- Portugal: City as a Living Lab
- At Carleton University, Batawa research, in conjunction with Lady Bata, data from 1940 – 1960 – 1980, can see road formation and building envelopes at different stages, combined building information models and infrastructure models, enables public imagination
- First Nations Land Planning and Infrastructure Management: have challenges combining with subsystems and systems, little intelligent data from city buildings
Conference: SimAUD, http://simaud.org/2013
Jose Lobo, School of Sustainability, Arizona State University, “Metabolism and Economic Activity for Global Urban Areas: Empirical Challenges”
- See profile at http://sustainability.asu.edu/people/persbio.php?pid=7908 and publications at http://www.experts.scival.com/asu/expertPubs.asp?n=Jose+A+Lobo&u_id=1143
Background as physicist
- Building theories as a goal
- Can only build good theories, if have good data
Urban economics: have good data
- e.g. characteristics been urban environment, and innovation
- Mostly rely on data in North America and Europe, because that’s where we have data
How do we build models for other places in the world?
We know how many people live in Mumbai
- But how much activity occurs?
- From census bureau in the U.S., have the value of economic activity coming out of cities
Metaphor: urban metabolism
- Taking inputs, generating outputs
- To turn metaphor into model, how much energy does a city use?
- Difficult to answer in the U.S., even more difficult for global cities
Defense department has Night Time Light (NTL) data, as a byproduct
- Defense Metereological …
Has led to research, e.g. “The Rise of the Mega-region” by Richard Florida; “Shedding Light on the Global Distribution of Economic Activity”
Also axiomatic will be some relationship between economics and light
Challenges:
- What is relationship between NTL (night time lighting), and economic activity?
- Can easily download data, be used to publish
Problem: saturation
- Light emissions are recorded on a scale of 0 to 63; past a certain level of intensity, the sensors are not sensitive enough
- Data is collected, passed on to NOAA, then available
- Unsaturated data is available only for one year, 2006
As organisms get larger, they get more efficient of use with energy
- Are cities like or unlike organisms?
Some research at U. of Toronto
- Research proposal to NASA rejected
[Questions]
Any outliers on relationship? Russia?
- Haven’t examined at global scale
- For North America, the differences between richest and poorest not so large
Slobodan Simonovic, University of Western Ontario, “City Resilience Simulator”
- See profile at http://www.eng.uwo.ca/research/compendium/faculty/ssimonovic.htm and web site at http://www.slobodansimonovic.com/
Building resilience as part of a new paradigm
Resilient city is seen as sustainable goal, for social and institutional gain
Trying to develop measures of resilience, covariant with large urban environments
Quantifying the concept of resilience
- Single number
- Performance of a system over time, through a multidimensional space of performance measures
- Could be a generic way of looking at different ideas
Definition of resilience (as per engineering)
- Ability to reduced the chance of shock, to absorb shock if it occurs and to recover quickly after a shock
- May or may not recover to prior state
Graph: vertical axis as quality of infrastructure, or quality of living
- Horizontal axis: time
- Resilience is area under curve, the time to recover to initial stable state
Trying to map time and space
Impacts: non-climatic or climatic
- Vulnerability and exposure
Adaptive capacity
Now working on research with coastal cities: Vancouver, Lagos
- Have developed generic simulator
[questions]
New centre for Resilient Infrastructure at U. of Toronto
Chris Kennedy, University of Toronto “Mathematical Model of Urban Metabolism”
- See profile at http://www.civil.engineering.utoronto.ca/staff/professors/kennedy.htm and 2011 book on “The Evolution of Great World Cities: Urban Wealth and Economic Growth at http://www.utppublishing.com/The-Evolution-of-Great-World-Cities-Urban-Wealth-and-Economic-Growth.html
- This talk is based on a chapter, “A Mathematical Model of Urban Metabolism”, Sustainability Science, http://dx.doi.org/10.1007/978-1-4614-3188-6_13
One slide, may be through in 5 minutes
Bottom up development
Metabolism at water flows, material flows, energy
A lot of studies published
- Any city that has done a greenhouse gas study has the data
Moving to new level of research
- An accounting framework of flows inside and outside of city
Published as book, not subject to peer review
Breaking down components:
- Heating fuels relate to breathing
- In U.S., some relationship with July temperatures
Research learning about metabolism and greenhouse gases
Have Canadian municipalities with data about how they’ll be dealing with greenhouse gases
Published “The Evolution of Great World Cities: Urban Wealth and Economic Growth” published in 2011
Mathematical model is driven by urban infrastructure, e.g. square footage of buildings, roads
David Ing – Service Systems, Natural Systems
See profile at http://en.wikipedia.org/wiki/David_Ing , blog at http://coevolving.com/blogs/, microblog at http://twitter.com/daviding
See slides at http://coevolving.com/commons/20121011-citysciences-ing-service-systems-natural-systems .
Andre Sorensen and Paul Hess (University of Toronto), “Urban Form Analysis: Toronto region evolution over 60 years”
- See Andre Sorensen profile at http://www.utsc.utoronto.ca/~socsci/sorensen/ and personal web pages at http://www.utsc.utoronto.ca/~sorensen/Home.html
- See Paul Hess profile at http://www.geog.utoronto.ca/people/faculty/hess and personal web pages at http://faculty.geog.utoronto.ca/Hess/hess_home.html
Work on the Toronto region, based on census data
Census data is bad to looking for urban form
Have been creating a dataset based on parcels, that can be aggregated
- 550 areas, based on pre-settlement farms and concession roads, as basic planning units for suburban development over the past 60 years
Halton, Peel, York, Durham, Toronto — have become a single urban centre
Divided up into study areas, can see by date for housing stock, making a typology of different areas in the region: housing density, plot size, socioeconomic status
Institutions, planning institutions that govern, evolving over 60 year period
- Different rulesets from different approaches over different jurisdictions
- Output of urban form has been consistent
- Robust in creating an urban form, with an inflection in the 1990s smarter growth, denser grids
Over summer, coded all of the housing stock as semi-detached, townhouses, etc.
Have data from 2010, won’t have the same long form data in 2011
[Question]
Condos?
- Don’t have ownership data, so have apartment buildings
Seanna Davidson, University of Waterloo, “Building Social Innovation Tools at the Waterloo Institute for Innovation and Resilience”
See profile at http://sig.uwaterloo.ca/profile/sdavidson and microblog at http://twitter.com/seannalee
Systems Mapping Tool being developed by Dan McCarthy
- How to map systems in a way so that we can have a dialogue
- Questions to work through with a group
- Understanding landscape boundaries
- Context, what’s included and not
- Easy to scale up, not necessarily breaking down into smaller pieces
- What is the current system that we’re working with?
- Feedback with institutions
- Alternatives
- Working towards a transformative pathway
Social Innovation Lab
- Towards the end of process
- Design labs and change labs have been popular, how is a social innovation lab different?
- Design labs: emphasize on quality of ideas, incorporating different perspectives
- Change labs: collaboration, shared understanding between participants, towards shared action
- Social innovation lab: emphasize development of solutions, more attention on integration into systems
- Have been working with other partners, e.g. Helsinki Design Lab, Mindlab, Institute without Boundaries, Stanford Design for Change, to figure out how Social Innovation Lab is different
- Process: design brief, types of research necessary skills, at what stages
- Struuture: where they should happen, how they’re funded, how projects are generated, how they are rolled out
- Design principles: social innovation model of systems with cross-scale and cross-sector perspectives; using ethnographic research methods, to create the design brief; including intensive workshops that bring in emotions; customized model to test sensitivities to ideas
Research to date is on web site
- Social Innovation Generation: a community of practice on how things happen
[Questions]
Helsinki Design Lab different as funded by government, as compared to organization funded by foundations
- Not trying to replicate, want to learn from expertise of colleagues of what has worked and hasn’t worked
Marc-Paul Gauthier, Arup, “IMSEP Integrated Mode Share Estimation Platform for Sites and Station Areas”
- See profile at http://www.linkedin.com/pub/marc-paul-gauthier/4/1aa/54, microblog at http://twitter.com/marcpaulg
Part of some internal research at Arup
Want to understand transit mode share
- Could do case study analyses, areas like what we have, e.g. region with 10%, 30% ridership
- Problem: don’t know what’s interactive
- Other side, could build regional models, based on transportation modes like roads and rails
- Would like to have something between case study analyses and regional models
Look not just at travel time and mode shares, but how the area operates
- 3D: Density, diversity, design
Created an ESRI ArcMap plugin using Python (didn’t want to rebuild everything, relies on GIS)
Research had coefficients, e.g. density of street networks, distance between stops
- In Toronto, generated transit ridership by station
- Block sizes –> station boarding
IMSEP case study, Scarborough Town Centre
- Mapped, and then started making changes, adding residential blocks that weren’t there before
- Impacts density, land use mix, leading to different transit station boarding
- Result would look like City Engine output demoed earlier
Next steps: Would like develop coefficients for Toronto, like those done for Seoul
David Ing (for Roy Wiseman and Jim Amsden) “The Municipal Reference Model: Government by Design”
- See Roy Wiseman profile at http://www.linkedin.com/pub/roy-wiseman/8/387/801
- See Jim Amsden profile at http://www.linkedin.com/pub/jim-amsden/3/90/77a and blog at https://www.ibm.com/developerworks/mydeveloperworks/blogs/home/tags/jim-amsden?lang=en
See slides at http://coevolving.com/commons/20121011-the-municipal-reference-model-government-by-design .
Michael van der Laan, University of British Columbia, “Web Application of Urban Form and Land Use Cases”
In BC, all communities collect energy consumption and emissions data
- Now to run for future development
Develop methods of “measured visualization”: what resonates with citizens?
Venue is engagement with city planners, professionals, stakeholders
Elevate consideration of sustainability
Have tool on web site where people can play with insulation values, seeing different quantitative data in cases
- Cases like a building information model, but maybe not so complicated
- Can then aggregate data into combinations and arrangements to simulate alternatives
Learning how to improve measured visualization tools, test potential of user created content