Ce diaporama a bien été signalé.
Nous utilisons votre profil LinkedIn et vos données d’activité pour vous proposer des publicités personnalisées et pertinentes. Vous pouvez changer vos préférences de publicités à tout moment.
The European Commission’s science
and knowledge service
Joint Research Centre
From place of residence to place of activity
Emerging data and tools for territorial analyses
Carlo LAVALLE, Filipe BATIST...
Outline
 Emerging geospatial data
 Applications
 Localisation of productive activities
 Cities’ attractiveness
 Day/N...
Emerging geospatial data sources
Data generated as a by-product
Unintentional crowd-sourced data.
 Mobile network operato...
Points of Interest (POI)
 Physical structures on the Earth’s surface with a functionality relevant to human
or societal a...
Density of Points of Interest in Paris per 500 m cells
Commerce
Food
Commerce
Other
Education
School
Education
University
...
POI data – Application: Land Use
Land use characterization
using POI data
Part of a wider project to refine
the thematic a...
CORINE
Land Cover
LUISA Base
Map
Localisation of industry-commerce-service clusters using POI data
CORINE Land Cover LUISA...
1211 Production facilities (ABCDE)
1212 Commercial/service facilities (GHIJKLMN)
1213 Public facilities (OPQ)
121 Industri...
Original Land
Use Map (CLC
2012)
Bologna
Parma
LUISA
base map
Bologna
Parma
Torino
LUISA base map -
Population
Bologna
Parma
Identification of clusters of mixed services as element for walkable neighbourhoods
 Input data:
 Open Street Map points...
High
Low
Agglomeration
index
Service centres - Milan
High
Low
Agglomeration
index
Service centres - Milan
Brussels Wien
Vilnius
Athens
Copenhagen Madrid
Milan
Warsaw
Service centres – various capitals
Service centres – various capitals
New set of POI data from Google Maps
enriched with Popular Hours data.
 Fine spatial resolution and 24/7
temporal detail....
Web mining
Applied to extract useful information from websites.
Many applications:
Non directly geospatial-oriented
 Medi...
FDI data from www.fdimarkets.com (FT)
EU28 investors: most common value (frequency)
Period
Source
Country
Sub Sector Marke...
Mapping hot spots of FDI origin and destination
Source (2003-2018) Destinations
African FDI flows (2003-2018)Egypt investitors analysis: most common value per period (frequency)
Period Source Cntr Sub S...
Malta: 1230 houses
Density: 3.89 point/km2
Housing (ask) prices from Remax (Italy) Features (Total 34):
lat - Latitude
lon...
Convergenc
e
Housing Cost: Milan time-series (Sep 2016 – Aug 2018)
Conventional + Big Data: Mapping tourism
Goal
 Tourism density maps at the highest possible temporal and spatial resoluti...
Mapping tourism
Summer
Summer
Easter
Skiing
Summer
Summer
Santa
Claus
Aurora borealis
Seasonality: Very low
Seasonality: Very high
Seasona...
Application: Spatiotemporal population
Day & night-time 24 grids
Integrating conventional and non-conventional data in
a c...
Difference in population between day-
and night-time
+ in night-time
+ in daytime
Spatiotemporal population
Ljubljana
Night-time
population
Day-time
population
Functional Border Areas
Modelling travel time to land border
crossings
‘Origin’ points
 Regularly spaced points at 10 Km ...
Functional Border Areas
Dashed polygon = Polish NUTS3 of
Szczecinecko-pyrzycki (PL427).
Officially considered a cross-bord...
Functional Border Areas
Functional Border Areas
Functional Border Areas
Functional Border Areas
Positive day- /night-time ratio =
more day-time population (workers,
students…) than nighttime
(re...
Functional Border Areas
Functional Border Areas
Positive day- /night-time ratio =
more day-time population (workers,
students…) than nighttime
(re...
Functional Border Areas
4.8
3.8
2.3
4.1
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
5.0
AT AT_border IT_border IT
%
Population...
Functional Border Areas
7.6
24.4
0.0
5.0
10.0
15.0
20.0
25.0
30.0
AT IT
%
Day- / night-time population ratio
Positive day-...
New geospatial data & Territorial Modelling
Role of new data in territorial modelling:
Complement spatial and thematic acc...
Territorial Analysis and Modelling
Territorial Impact Assessment
RHOMOLO: Investments, Infrastructures, Human Capital
Terr...
New analysis: demographic projection by age
class. Aging patterns in urban areas
Population over 65 years old in 2030.
Pop...
Stay in touch
KC TP Knowledge Centre for Territorial Policies:
https://ec.europa.eu/knowledge4policy/territorial_en
Commun...
Thank you
Any questions?
Carlo.Lavalle@ec.europa.eu
Prochain SlideShare
Chargement dans…5
×

Carlo Lavalle - From place of residence to place of activity: Emerging data and tools for territorial analyses

346 vues

Publié le

Presentation by Carlo Lavalle, Joint Research Centre, European Commission at the OECD Workshop on Spatial Dimensions of Productivity, 28-29 March 2019, Bolzano.

More info: https://oe.cd/GFPBolzano2019

  • Soyez le premier à commenter

  • Soyez le premier à aimer ceci

Carlo Lavalle - From place of residence to place of activity: Emerging data and tools for territorial analyses

  1. 1. The European Commission’s science and knowledge service Joint Research Centre
  2. 2. From place of residence to place of activity Emerging data and tools for territorial analyses Carlo LAVALLE, Filipe BATISTA, Ricardo BARRANCO, Claudia BARANZELLI, Mert KOMPIL , Chris JACOBS-CRISIONI, Andrius KUCAS JRC.B.3 – Territorial Development unit Workshop on Spatial Dimensions of Productivity, 28-29 March 2019 Bolzano, Italy
  3. 3. Outline  Emerging geospatial data  Applications  Localisation of productive activities  Cities’ attractiveness  Day/Night Time population mapping  Functional Border Areas  New insights in Territorial Modelling
  4. 4. Emerging geospatial data sources Data generated as a by-product Unintentional crowd-sourced data.  Mobile network operator (MNO) data  Web activity (content, traffic, searches…)  Social media (tweets, check-ins, photos…)  Transactions (costumer, financial…) Data generated on purpose Intentionally produced as a core component of ICT-based service. Aspects in common with of Big Data.  Navigation/mapping data (e.g. POI, road networks), but volunteered/collaborative or private initiatives  Sensors (count of vehicles, pedestrians, air-borne, satellite)
  5. 5. Points of Interest (POI)  Physical structures on the Earth’s surface with a functionality relevant to human or societal activities.  Mapped as a precise points on a (digital) map.  Many sources:  OpenStreetMap (VGI, free and open source)  Navigation /mapping / sector data (proprietary) (e.g. TomTom)  Derived from mining web services (e.g. Booking.com, TripAdvisor)  Different levels of quality, completeness, overlap  Different classification systems  Different quality (completeness, accuracy…)
  6. 6. Density of Points of Interest in Paris per 500 m cells Commerce Food Commerce Other Education School Education University Health General Health Hospital Parks Recreation Restaurant Source: TomTom Points of Interest Elaboration: European Commission JRC B.3 LUISA Territorial Modelling Platform, 2018
  7. 7. POI data – Application: Land Use Land use characterization using POI data Part of a wider project to refine the thematic and spatial detail of CORINE Land Cover and map spatiotemporal population densities (ENACT). Main objective: To break down CLC class 121 (“Industrial and Commercial Sites”) into 3 more detailed land uses.
  8. 8. CORINE Land Cover LUISA Base Map Localisation of industry-commerce-service clusters using POI data CORINE Land Cover LUISA Base Map
  9. 9. 1211 Production facilities (ABCDE) 1212 Commercial/service facilities (GHIJKLMN) 1213 Public facilities (OPQ) 121 Industrial and commercial units Localisation of industry-commerce-service clusters using POI data CORINE Land Cover LUISA Base Map
  10. 10. Original Land Use Map (CLC 2012) Bologna Parma
  11. 11. LUISA base map Bologna Parma
  12. 12. Torino LUISA base map - Population Bologna Parma
  13. 13. Identification of clusters of mixed services as element for walkable neighbourhoods  Input data:  Open Street Map points of interest (POI)  Filtered tags include: amenity, leisure, shop and tourism (values excluded if related to cars and, in general, vehicular mobility)  Methodology:  Optimised hotspot analysis (index of agglomeration) POI – Application: Walkability
  14. 14. High Low Agglomeration index Service centres - Milan
  15. 15. High Low Agglomeration index Service centres - Milan
  16. 16. Brussels Wien Vilnius Athens Copenhagen Madrid Milan Warsaw Service centres – various capitals
  17. 17. Service centres – various capitals
  18. 18. New set of POI data from Google Maps enriched with Popular Hours data.  Fine spatial resolution and 24/7 temporal detail.  Multiple activity types. POI: Spatiotemporal population
  19. 19. Web mining Applied to extract useful information from websites. Many applications: Non directly geospatial-oriented  Media monitoring (e.g. EMM).  Mining of prices for price indexes, inflation rates.  Citizen and costumer sentiment (widely used by private sector to optimize business). Geospatial-oriented When information can be linked to a geographical location by means of coordinates or place names.
  20. 20. FDI data from www.fdimarkets.com (FT) EU28 investors: most common value (frequency) Period Source Country Sub Sector Market Motive 2003-2018 United States (14576) Software publishers, except video games (5279) Regional (8028) Proximity to markets or customers (1357) 0 5000 10000 15000 20000 25000 30000 35000 40000 45000 50000 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 $Millions EU28 FDI: Top 10 investors (2003-2017) USA Germany France UK Netherlands Switzerland Spain Japan Austria Sweden Multidimension dataset on FDI: Spatial: Source – Destination (country/region/city). Temporal: Monthly for 2003-2018. Thematic: Sector, activity, type, market, motive... Capital expenditure and Jobs created.
  21. 21. Mapping hot spots of FDI origin and destination Source (2003-2018) Destinations
  22. 22. African FDI flows (2003-2018)Egypt investitors analysis: most common value per period (frequency) Period Source Cntr Sub Sector Market Motive 2003 - 2007 United States (40) Oil & gas extraction (63) Regional (6) Domestic Market Growth Potential (15) 2008 - 2012 United States (60) Retail banking (91) Regional (6) Domestic Market Growth Potential (16) 2013 - 2018 UAE (39) Retail banking (88) Domestic (5) Domestic Market Growth Potential (11) All (2003-2018) United States (57) Retail banking (137) Regional (7) Domestic Market Growth Potential (33) FDI Analysis: Flows in Africa
  23. 23. Malta: 1230 houses Density: 3.89 point/km2 Housing (ask) prices from Remax (Italy) Features (Total 34): lat - Latitude lon - Longitude keys - ID price - Selling Price (€) totalRooms - Total Rooms bedrooms - Bed Rooms bathrooms - Bathr totalsqm - Total Square Meters (m2) lotsize - Lot Size (m2) year - Construction Year builtArea - Building Area (m2) parkingspaces - Number of parking spaces floors - Number of floors floorlevel - Floor of the house toiletRooms - Number of toilets energyClass - Energy class *** energyEff - Energy Efficiency (kWh/m2 per year) *** Features (Total 17 - yes or no): garage - Garage (yes/no) pool - Swimming Pool (yes/no) renovated - Renovated (yes/no) fireplace - Fireplace (yes/no) terrace - Terrace (yes/no) balcony - Balcony (yes/no) garden - Garden (yes/no) liftelev - Lift or Elevator (yes/no) parking - Parking places (yes/no) heating - Heating System (yes/no) solar - Solar panels (yes/no) oil - Oil heating (yes/no) ac - Air Conditioner (yes/no) sewer - Connected to sewer (yes/no) pool - Swimming Pool(yes/no) security - Alarm or security system (yes/no) kitchen - Kitchen (yes/no)
  24. 24. Convergenc e Housing Cost: Milan time-series (Sep 2016 – Aug 2018)
  25. 25. Conventional + Big Data: Mapping tourism Goal  Tourism density maps at the highest possible temporal and spatial resolutions, for the whole of the EU28. Tourism density defined as the number of tourists present in a given location at a given time. Methodology  Downscale monthly nights-spent to local level. Input data  Conventional statistical data: Nights-spent (ESTAT) & Seasonal curves (NSIs)  Big data: location of accommodation establishments (Booking.com and TripAdvisor)
  26. 26. Mapping tourism
  27. 27. Summer Summer Easter Skiing Summer Summer Santa Claus Aurora borealis Seasonality: Very low Seasonality: Very high Seasonality: medium Seasonality: high Tiroler Unterland, Austria
  28. 28. Application: Spatiotemporal population Day & night-time 24 grids Integrating conventional and non-conventional data in a coherent framework
  29. 29. Difference in population between day- and night-time + in night-time + in daytime
  30. 30. Spatiotemporal population Ljubljana Night-time population Day-time population
  31. 31. Functional Border Areas Modelling travel time to land border crossings ‘Origin’ points  Regularly spaced points at 10 Km intervals, +  Centroids of "urban clusters" (continuously populated areas of 300 inhab./Km2 and at least 5k inhabitants). ‘Destination’ points  Land border crossings, including: Paved roads/bridges (this includes the Copenhagen- Malmo bridge) and river ferries.
  32. 32. Functional Border Areas Dashed polygon = Polish NUTS3 of Szczecinecko-pyrzycki (PL427). Officially considered a cross-border region, but… …only a very small portion of its territory is within 30 or 60 minutes from border crossings.
  33. 33. Functional Border Areas
  34. 34. Functional Border Areas
  35. 35. Functional Border Areas
  36. 36. Functional Border Areas Positive day- /night-time ratio = more day-time population (workers, students…) than nighttime (residents).
  37. 37. Functional Border Areas
  38. 38. Functional Border Areas Positive day- /night-time ratio = more day-time population (workers, students…) than nighttime (residents).
  39. 39. Functional Border Areas 4.8 3.8 2.3 4.1 0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5 4.0 4.5 5.0 AT AT_border IT_border IT % Population change (2001-2011) 0 50 100 150 200 250 300 350 400 1961 1971 1981 1991 2001 2011 Inhabitants(thousands) Austria - Italy AT IT
  40. 40. Functional Border Areas 7.6 24.4 0.0 5.0 10.0 15.0 20.0 25.0 30.0 AT IT % Day- / night-time population ratio Positive day- /night-time ratio = more day-time population (workers, students…) than nighttime (residents).
  41. 41. New geospatial data & Territorial Modelling Role of new data in territorial modelling: Complement spatial and thematic accuracy of statistical data Estimate supply and cost of services (e.g. childcare, education etc.) at sub-national level Evaluate attractiveness and suitabilities for investments Perfom ad-hoc spatial analysis (spill-over effects, super-linearity growth of cities)
  42. 42. Territorial Analysis and Modelling Territorial Impact Assessment RHOMOLO: Investments, Infrastructures, Human Capital Territorial Trends LUISA: GDP and Demographic trends (Reference)
  43. 43. New analysis: demographic projection by age class. Aging patterns in urban areas Population over 65 years old in 2030. Population in the core city (including those that over 65 years old) will decrease. Population in the Functional Urban Ares (including those that over 65 years old) in most cases will increase.
  44. 44. Stay in touch KC TP Knowledge Centre for Territorial Policies: https://ec.europa.eu/knowledge4policy/territorial_en Community of Practice on Cities (CoP-CITIES): https://ec.europa.eu/knowledge4policy/territorial/topic/urban_e n#CoP_CITIES UDP Urban Data Platform: https://urban.jrc.ec.Europa.eu STRAT-Board Territorial and Urban Development Strategies: https://urban.jrc.ec.europa.eu/strat-board GHSL - Global Human Settlement Layer: http://ghsl.jrc.ec.europa.eu
  45. 45. Thank you Any questions? Carlo.Lavalle@ec.europa.eu

×