Upload
others
View
2
Download
0
Embed Size (px)
Citation preview
Contains Ordnance Survey data © Crown copyright and database right 2017
Predicting CrimeBen O’Hara
Information Specialist
West Yorkshire Police
ben.o’[email protected]
Contains Ordnance Survey data © Crown copyright and database right 2017
Burglary Dwellings in Leeds District
0
200
400
600
800
1000
1200
Jun
07
Au
g 07
Oct
07
Dec
07
Feb
08
Ap
r 0
8
Jun
08
Au
g 08
Oct
08
Dec
08
Feb
09
Ap
r 0
9
Jun
09
Au
g 09
Oct
09
Dec
09
Feb
10
Ap
r 1
0
Jun
10
Au
g 10
Oct
10
Dec
10
Feb
11
Ap
r 1
1
Jun
11
Au
g 11
Oct
11
Dec
11
Feb
12
Crimes
Crimes
Contains Ordnance Survey data © Crown copyright and database right 2017
Predicting Crime
Contains Ordnance Survey data © Crown copyright and database right 2017
Operation Optimal
• Op Trafford in GMP reduced Burglaries by 38%
• Optimal Forager Theory• A criminal will return to a ‘successful’ area, until something disrupts that
feeling of comfort
• “In domestic burglary, for example, the danger of a further crime is greatest at the home of the original victim and spreads out to some 400 metres, but disappears over 6 weeks to 2 months”Ross & Pease (2007)
Contains Ordnance Survey data © Crown copyright and database right 2017
Operation Optimal
Contains Ordnance Survey data © Crown copyright and database right 2017
Patrol Plan
Contains Ordnance Survey data © Crown copyright and database right 2017
Burglary Dwellings in Leeds District
0
200
400
600
800
1000
1200Ju
n 0
7
Sep
07
Dec
07
Mar
08
Jun
08
Sep
08
Dec
08
Mar
09
Jun
09
Sep
09
Dec
09
Mar
10
Jun
10
Sep
10
Dec
10
Mar
11
Jun
11
Sep
11
Dec
11
Mar
12
Jun
12
Sep
12
Dec
12
Mar
13
Jun
13
Sep
13
Dec
13
Mar
14
Jun
14
Sep
14
Dec
14
Mar
15
Jun
15
Sep
15
Dec
15
Mar
16
Jun
16
Sep
16
Dec
16
Crimes Optimal
Contains Ordnance Survey data © Crown copyright and database right 2017
Patrolwise
• Developed by UCL
• Python scripts produce a “patrol plan” identifying which streets are at highest risk of crime
• Maps are accessed on mobile devices indicating the areas that should be patrolled & suggested times.
Contains Ordnance Survey data © Crown copyright and database right 2017
Diffusion
Contains Ordnance Survey data © Crown copyright and database right 2017
Crime Pattern Theory
Home
Work/School
Friends
Entertainment & Shopping
Contains Ordnance Survey data © Crown copyright and database right 2017
Betweenness
• We can’t measure which streets a person will use most often
• Betweenness acts as a proxy to identify which streets have the greatest chance of being used for routine travel
Contains Ordnance Survey data © Crown copyright and database right 2017
Patrol Plans
Contains Ordnance Survey data © Crown copyright and database right 2017
The current process
• UCL Python script produces Esri Shape File
• Automation scripts1. Upload the data into SQL Database
2. Use GeognoSIS WML to generate an image of each patrol area
3. Send the images via email to the Sgts involved
• Sgts forward the emails on to the officers deployed to those areas
Contains Ordnance Survey data © Crown copyright and database right 2017
The future process
• UCL Python script produces Esri Shape File
• Automation scripts1. Upload the data into SQL Database
• Sgts deploy officers to a patrol plan
• Officers access Cadcorp Web Map Layers on their mobile phones showing the patrol areas, along with associated intelligence, local points of interest, etc.
• Should they be deployed to an area with poor coverage, the software on the phones will have cached the image, using GeognoSIS WML, along with available intelligence.
Contains Ordnance Survey data © Crown copyright and database right 2017
Questions?