@gregbonnette

linkedin.com/in/gbonnette

Greg Bonnette Vice President, Strategy & Innovation

This is NOTMinority Report

POPULATION

110,000SQUARE MILES

35SWORN OFFICERS

237CRIME (2014)

649 Violent4743 Property

Rise in Crime (5 Year)

• Robbery

• Burglary

• Theft from MVs

Lack of Resources

• Reduced Manpower

• Increased Demand

• Shrinking Budgets

It is not uncommon at many police departments,

according to the National Alliance on Mental Illness,

for more than 20% of daily calls

for service to involve people who

are emotionally disturbed.

Mental illness cases swamp criminal justice system, Kevin Johnson, USA Today, 2014

“We have replaced the hospital bed with the jail cell, the homeless shelter and the coffin”

“Some are looking to the police department to do things that you never imagined. Dealing with the mentally ill is part of that.”

“Police should not be involved this process at all, but nobody else can or wants to do it.”

“Law enforcement did not ask for this additional challenge.”

They end up here (the criminal justice system), because we are the only system that can’t say no.”

Mental illness cases swamp criminal justice system, Kevin Johnson, USA Today, 2014

491 OVERDOSE CALLS 59 DEATHSManchester, NH Police Department 2015 Year To Date

-10000

-8000

-6000

-4000

-2000

0

2000

4000

6000

Jan

2008

Apr

2008

Jul

2008

Oct

2008

Jan

2009

Apr

2009

Jul

2009

Oct

2009

Jan

2010

Apr

2010

Jul

2010

Oct

2010

Jan

2011

Apr

2011

Jul

2011

Oct

2011

Jan

2012

Apr

2012

Jul

2012

Oct

2012

Jan

2013

Apr

2013

Jul

2013

Oct

2013

Jan

2014

Apr

2014

Jul

2014

Oct

2014

Jan

2015

Apr

2015

Jul

2015

Na

t C

ha

ng

e I

n J

ob

s

Govt Growth Private Growth

2015 US Bureau of Labor Statistics (Government, Private)

Reduce Crime

Increase ActionableIntelligence for Officers

Improve Agency Efficiency & Effectiveness

Descriptive Hot Spot Policing

2014: Descriptive

Hot Spots

• Data clean up

• Refine policing tactics

• Organizational preparedness

Predictive Hot Spot Policing

Results

Suitable Targets

(Victims)

Motivated Offenders

Lack of Capable

Guardians

Cohen and Felson (1979)

Crime will cluster in certain sectors of urban zones, while others remain crime free¹

Over 50% of crimes occur in 3% of places²

1) Sherman 1988,1989), 2) Pierce et al. 1988, Sherman et al. 1989a,b

Initial results exposed data quality issues

Process issues caused incident clustering at HQ and Local Hospitals

Internal campaign to improve data quality

Koper (1995)

12-16 Minutes = 2 Hours

Mitchell (2011)

Violent & Property (Part 1) Crimes Down 25%

Descriptive Hot Spot Policing

2014: Descriptive

Hot Spots

• Data clean up

• Refine policing tactics

• Organizational preparedness

Predictive Hot Spot Policing

Q1 2015: Ironside

• Initial Proof of Concept

• Acquired IBM SPSS Modeler

• Data Science Mentoring & Development

Results

Data Collection

Crime Analysis

Police Operations

Criminal Response

AlteredEnvironment

Data Management

PredictionIntervention

Record Management

System

Parole & Probation

Liquor Licensing Traffic

Weather

Event Detail

Integration & Automation

Crime Analysis Data Hub

Crime Analysis Workbench

Community Specific Models

Data Management

Predictive Hot Spots

FullAutomation

Reporting & Visualization

Ad-hocAnalysis

Descriptive Hot Spots

Heat Maps

Other Offender, Perpetrator, Spree, Victim Analysis

Data Collection Crime Analysis Police Operations

IBM SPSSModeler 17

GIS

IBM SPSS Modeler

Not a black box - we wanted to know what was going on underneath

Graphical - Light on coding, easy to learn

Not limited to hot spots only

Easy to merge multiple sources

Ironside

POC demonstrated significant improvement over retroactive hot spots

Brought strong data science mentors who knew the technology

High-touch side-by-side model build to transfer all knowledge

End result was acquiring a department capability to do all things predictive policing, not just a hot spot scoring engine.

History in this square and 8 adjacent

Crime History & Incident Category

Time of Day

Day of Week

Weather Conditions

12

3

1-3

1-2

1-1

1-4

2-2

2-1

2-5

2-34

3-1

3-4 3-3

3-2

FR

ON

TST

LOND

ON

DERR

YTPK

ROUTE

101- W

EST

ROU TE 101 - EAST

RAYMOND WIE

CZOREK

DR

LO

NDONDERR

YTPK

UN

IO

N S

T

SRI

VERRD

ELM

ST

PIN

E S

T

BE

EC

H S

T

RI VER

RD

CANDIA RD

HANOVER ST

HU

SE

RD

BODWELL

R

D

SM

AM

MO

TH

RD

LAKE AVE

HA

RVEY

RD

SM

YTH

RD

RO

CKI

NGHAM

R D

CILLEY RD

VALLEY ST

S B

EE

CH

ST

DU

N

BARTON

RD

CA

LEF

RD

WEL

LINGTON RD

BRIDGE ST

MA

STRD

SM

AIN

ST

GOLD ST

CA

NA

LS

T

COHA

SA

VE

S

WILL

O

WST

GOFFSTOWN

RD

BOYNTON S

T

MA

MM

OTH

R

D

WEBSTER ST

BREMER ST

KELLEY ST

SOMERVILLE ST

PER

IMETE

RRD

GO

FFS

FALL

SRD

VINTON STISLAND POND RD

FR

ON

T S

T

HO

OK

SETT R

D

ED

DY

RD

WESTON RD

MA

IN

ST

MASSABESIC

ST

SECO

ND

ST

HA

ZELT

ON

AV

E

SP

OR

TE

RS

T

JO

BINDR

AIR PORT RD

COHAS AVE

SM

YTH

RD

BRIDGE ST

ST101

ST101

§̈¦293

§̈¦93

§̈¦293

INTERSTATE

93

-SOUTH

INTER

STA

TE

93

-NO

RTH

INTERSTATE 293 - NORTH

F.E

.E

VER

ET

TT

UR

NP

IK

ES

OU

TH

ROUTE

101-EA

ST

12

3

1-3

1-2

1-1

1-4

2-2

2-1

2-5

2-34

3-1

3-4 3-3

3-2FR

ON

TST

LOND

ON

DERR

YTPK

ROUTE 101 - W

EST

ROU TE 101 - EAST

RAYMOND WIE

CZO

REK

DR

LO

NDONDERR

YTPK

UN

IO

N S

T

SRI

VERRD

ELM

ST

PIN

E S

T

BE

EC

H S

T

R I VER

RD

CANDIA RD

HANOVER ST

HU

SE

RD

BODWELL

R

D

SM

AM

MO

TH

RD

LAKE AVE

HA

RVEY

RD

SM

YTH

RD

RO

CKI

NGHAM

R D

CILLEY RD

VALLEY ST

S B

EE

CH

ST

DU

N

BARTON

RD

CA

LE

FR

D

WEL

LINGTON RD

BRIDGE ST

MA

ST

RD

SM

AIN

ST

GOLD ST

CA

NA

LS

T

COHA

SA

VE

S

WIL

LO

WST

GOFFSTOWN

RD

BOYNTON S

T

MA

MM

OTH

R

D

WEBSTER ST

BREMER ST

KELLEY ST

SOMERVILLE ST

PER

IM

ETE

RRD

GO

FFS

FALLS

RD

VINTON STISLAND POND RD

FR

ON

T S

T

HO

OK

SETT R

D

ED

DY

RD

WESTON RD

MA

IN

ST

MASSABESIC

ST

SECO

ND

ST

HA

ZELT

ON

AV

E

SP

OR

TER

ST

JO

BINDR

AIR PORT RD

COHAS AVE

SM

YTH

RD

BRIDGE ST

ST101

ST101

§̈¦293

§̈¦93

§̈¦293

INTERSTATE

93

-SOUTH

INTE

RSTA

TE

93

-N

OR

TH

INTERSTATE 293 - NORTH

F.E

.EV

ER

ET

TT

UR

NP

IK

ES

OU

TH

ROUTE

101

-EA

ST

1024000.000000

1024000.000000

1028000.000000

1028000.000000

1032000.000000

1032000.000000

1036000.000000

1036000.000000

1040000.000000

1040000.000000

1044000.000000

1044000.000000

1048000.000000

1048000.000000

1052000.000000

1052000.000000

1056000.000000

1056000.000000

1060000.000000

1060000.000000 15

00

00

.00

00

00

15

10

00

.00

00

00

15

30

00

.00

00

00

15

40

00

.00

00

00

15

60

00

.00

00

00

15

70

00

.00

00

00

15

90

00

.00

00

00

16

00

00

.00

00

00

16

20

00

.00

00

00

16

30

00

.00

00

00

16

50

00

.00

00

00

16

60

00

.00

00

00

16

80

00

.00

00

00

16

90

00

.00

00

00

17

10

00

.00

00

00

17

20

00

.00

00

00

17

40

00

.00

00

00

17

50

00

.00

00

00

17

70

00

.00

00

00

17

80

00

.00

00

00

18

00

00

.00

00

00

18

10

00

.00

00

00

18

30

00

.00

00

00

18

40

00

.00

00

00

18

60

00

.00

00

00

18

70

00

.00

00

00

18

90

00

.00

00

00

19

00

00

.00

00

00

19

20

00

.00

00

00

19

30

00

.00

00

00

19

50

00

.00

00

00

19

60

00

.00

00

00

19

80

00

.00

00

00

19

90

00

.00

00

00

20

10

00

.00

00

00

Predictive Hot Spot Maps

15 Minute Hot Spot Patrols

=CRIME

+

Descriptive Hot Spot Policing

2014: Descriptive

Hot Spots

• Data clean up

• Refine policing tactics

• Organizational preparedness

Predictive Hot Spot Policing

Q1 2015: Ironside

• Initial Proof of Concept

• Acquired IBM SPSS Modeler

• Data Science Mentoring & Development

Results

Q3 2015: Deployment

• 15 Minute Hot Spot Patrols

• Hot Spot Radio Call-Ins

Robbery Burglary Theft from MV

Vs. Prior 15 Week Period 32% 7% 40%

Vs. SamePeriod Last

Year43% 11% 29%

Manchester NH, Police Department is rated “Cutting Edge” with their Predictive Hot Spots Policing Initiative

Mayor Ted Gatsas & Board of Alderman

•Approval & Funding

Chief Nick Willard

•Primary Sponsor

•Drove Alignment

•Created Accountability

Officer Matthew Barter, S.W.A.T.

•Champion / Evangelist

•Crime Analyst

•Subject Matter Expert

Krista Comrie

•Crime Analyst

•Subject Matter Expert

Chi Shu

•Ironside Data Scientist

•Statistics and Machine Learning Domain Expert

Tony Schaffer

•Manchester , NH IS

•City GIS and Data Systems Domain Expert

Top down support (Police Driven)

Willingness to Change

Focus on Interested Personnel

Collaboration with IT Staff Early & Often

Organizational Preparedness Create System of Accountability (e.g. CompStat) Align on Data Quality

Keep Deployment Simple

Less is More

Be Prescriptive to a Point

Set Performance Goals

Make it Competitive

Make it Accountable

Highlight Wins Publicly

Offender Based Modeling

Predictive Hot Spot methodology for traffic accidents Areas to increase traffic enforcement

Secure grant funding

Gun Crime Analysis

Everyday Data Prep Tasks Faster ability to run queries

Create streams so information can be routinely run

Dash Boards, Improved KPI tracking, etc.

https://youtu.be/BNeMoIbEgI8

http://nh1.com/news/predictive-analytics-help-manchester-police-solve-crimes-with-computers-not-guns/

http://www.usatoday.com/story/news/nation/2014/05/12/mental-health-system-crisis/7746535/

http://www.usatoday.com/longform/news/nation/2014/07/21/mental-illness-law-enforcement-cost-of-not-caring/9951239/

http://data.bls.gov/timeseries/CES0500000001

http://data.bls.gov/timeseries/CES9000000001

http://cops.usdoj.gov/html/dispatch/06-2012/hot-spots-and-sacramento-pd.asp

http://www.dhhs.nh.gov/dcbcs/bdas/documents/issue-brief-heroin.pdf

http://www.rand.org/pubs/research_reports/RR233.html

http://www.rand.org/pubs/research_briefs/RB9735.html

http://www.rand.org/blog/2013/11/predictive-policing-an-effective-tool-but-not-a-crystal.html

http://www.rand.org/pubs/research_reports/RR531.html

https://www.ncjrs.gov/App/Publications/abstract.aspx?ID=167664