Title: MOSES: A Synthetic Spatial Model of UK Cities and Regions
1MOSES A Synthetic Spatial Model of UK Cities and
Regions
School of Geography FACULTY OF EARTH AND
ENVIRONMENT
- Mark BirkinUniversity of Leeds
2- OVERVIEW
- MoSeS
- Modelling and Simulation for e-Social Science
- Project funded under the ESRCs e-Social Science
initiative - One of eight major projects in the National
Centre for e-Social Science (NCeSS) (12 million
programme) - Others include Geographic Visualisation of Urban
Environments (GeoVUE) - And (arguably) a bunch of Computer Science
3- OVERVIEW
- e-Science
- Major research council initiative in the UK over
the last 6/7 years - Matched by the US Cyberinfrastructure programme
- Aims to address the Grand Challenges of
scientific research - Suggestion is that new solutions are brought into
view through a combination of - Data availability
- Simulation and visualisation
- Virtual collaboration
- All supported through a new generation of
computational infrastructure (Grid?)
4Powering the Virtual Universehttp//www.astrog
rid.ac.uk(Edinburgh, Belfast, Cambridge,
Leicester, London, Manchester, RAL)
Multi-wavelength showing the jet in M87 from top
to bottom Chandra X-ray, HST optical, Gemini
mid-IR, VLA radio. AstroGrid will provide
advanced, Grid based, federation and data mining
tools to facilitate better and faster scientific
output.
Picture credits NASA / Chandra X-ray
Observatory / Herman Marshall (MIT),
NASA/HST/Eric Perlman (UMBC), Gemini
Observatory/OSCIR, VLA/NSF/Eric Perlman
(UMBC)/Fang Zhou, Biretta (STScI)/F Owen (NRA)
p4
Printed 09/11/2009
5 myGrid Project
Motivation In silico experiments necessitate the
virtual organization of people, data, tools and
machines. The scientific process also
necessitates an awareness of the experience base,
both of personal data as well as the wider
context of work. The management of all these data
and the co-ordination of resources to manage such
virtual organizations and the data surrounding
them needs significant computational
infra-structure support.
6(No Transcript)
7OVERVIEW
- MoSeS
- The Modelling and Simulation of e-Social Science.
- MoSeS Objectives
- To develop a complete representation of the UK
population at a fine spatial scale - To produce rich, detailed and robust forecasts
of the future population of the UK - To investigate scenarios which relate
demographics to service provision - emphasis on
policy applications within the health and
transport policy sectors
8- MoSeS An Example
- Leeds Social Services
- Requirement to understand the future needs of the
population (morbidity/ mortality) - Allocation of resources
- Service delivery
- Statutory targets e.g. reduction of (spatial)
inequalities in life expectancy - Preparation of strategy demands a relatively long
view 2027?
9Population Projections
Source Office for National Statistics
Source Moses
10Ethnic Projections
Source Moses
11Growth in Elderly Population (85)2006-2031
12Model of disability (1) of Disability (1)
Disabled in Leeds
Disabled in UK
Source BHPS
Source Moses
Estimate of the disabled in Leeds 51,599
13Disabled in Leeds, 2006
Disabled in Leeds, 2031
Source Moses
Source Moses
Estimate of the disabled in Leeds 2031
93,698
Increase of 82!
14(No Transcript)
15Model of Disability (3)Scenario 5Plus1
Assume that a 65 year old in 2031 enjoys the
health of a 60 year old today
Disabled in Leeds, 2031
Disabled in Leeds, 2006
Baseline
Scenario
Source Moses
Source Moses
Revised estimate of the disabled in Leeds 2031
70,359
Increase of only 36!
16Other Estimates of Need
17- Moses Methodology
- What are the functional components of an applied
urban simulation? - Recreation of a baseline population
- A dynamic/ forecasting capability
- A suite of service utilisation and activity
models - A container (spatial decision support system?)
18- Moses Methodology
- We create a synthetic representation of the UK
population - Using data from the 2001 Census Small Area
Statistics and the Sample of Anonymised Records - 24 million households and 60 million residents
are individually represented - The synthetic population looks just like the
actual population but no real citizens are
included - The reconstructed population includes a wide
range of social and demographic attributes age,
ethnicity, housing, economic activity etc
19Moses Population Reconstruction Model
20(No Transcript)
21Health Status (Optimised)
Actual
Model
22Car ownership (Co-varying)
23Moses Activity Model
24Smoking
25Carers
26Diabetes
27 28Moses Dynamic Model
29Migration Model
- We combine two approaches
- A person-specific general model, using
probabilities of migration derived from the BHPS
applied to cloned individuals in households
derived from the 2001 Census SAR - Location specific information about migration
intensities in small areas (2001 Census SMS),
which are used to modify the results of the
person-specific model - The model has a two stage procedure
- Migrant generation protocol
- Migrant distribution protocol
30Migrant generation protocol
- Assess migration probabilities from an analysis
of BHPS data, 2000-2004 for - a) households
- b) groups
- c) individuals
- Major drivers of migration identified using a
stepwise chi-squared estimation procedure - Households age of head, household size, housing
type - Individuals age, household size, marital status
- Groups merged with individuals (small numbers)
- National rates are locally adjusted by age using
the Census Migration Statistics (SMS)
31Migrant generation households
32Migrant generation individuals
33Migrant distribution protocol
- The problem can be described as follows
- Estimate migration rates by location, age,
household size and housing type this process
creates a stock of vacant housing - For each migrant, by location and household type
(age, size) find a destination location by
location and house type - Calibrate this process using data on known moves
(by distance from the census SMS) and known
assignments of household type to house type
(BHPS)
34Simulation Database
Update Location and Dwelling Characteristics
1
5
Migrant generation model
2
2
Aggregate To Migrant Population
Aggregate To Vacant Dwellings
Migration distribution protocol ( See Birkin and
Clarke 1987 Nakaya et al. 2006)
Spatial Interaction Model
3
Compute dwelling preference for each migrant
4
35Migrant distribution model distribution model
Lambda Calibration
36(No Transcript)
37Model Results Aireborough
Observed
Predicted
38Model Results Seacroft
Observed
Predicted
39Model Results Headingley
Observed
Predicted
40Agent-based simulation of student migrants
- Agent-based simulation of student migrants
- We recognise the following groups
- First year undergraduates
- Other undergraduates
- Master students
- Doctoral students
- We apply the following rules
- Each group is allowed set years to stay in an
area - Students stay close to their university of study
- They dont do marriage and fertility
41(No Transcript)
42(No Transcript)
43(No Transcript)
44- Moses Methodology Architecture
45Moses Selection Portlet
46Moses Architecture
47Moses Mapping Portlet 1 Google Maps
48Moses Mapping Portlet 2 SeeGeo
49- Moses Discussion
- 1. Moses is not the only work in this area in
either an academic or a policy environment - But has some interesting and unique features!
50- Moses Discussion
- 2. This work has both an intellectual and a
practical value - Even though it is not critical
- Sometimes it is necessary to be constructive as
well
51- Moses Discussion
- 3. This work is hard
- Maybe too hard?
- Scale back ambition?
- Extend capability/ resourcing?
52- Moses Conclusions and Next Steps
- There is still much work to be done to establish
a convincing set of demonstrator applications for
urban simulation - Enhanced visual representation of simulation
outputs is one key ingredient - Collaboration with GeoVUE has important strategic
value - Embedding this research more clearly within a
paradigm of (generative) social simulation is a
potential means to re-enter the mainstream - Genesis project?