SlideShare a Scribd company logo
1 of 4
Download to read offline
Reallocating Patrol Zones Based on Previous
Crime Patterns
Kyung Kim
December 10, 2015
Abstract
This report will cover the application of the PPAC model on Georgia Tech Police
Department (GTPD) crime records from 2011-2014. In Section I, we will talk about
our motivation to do this research. In Section II, we will mathematically introduce,
in detail, how we cleaned up the data and formulated the PPAC model. In Section
III, we will present our results obtained by the application of the PPAC model. Fi-
nally in Section IV, we will entertain the idea of moving forward with our research by
introducing future directions.
1 Introduction
It is time for high school students to apply to colleges, and campus safety is one aspect
that should not be overlooked. Although Georgia Tech is one of the safest schools, to further
relieve students from having these types of concerns, we worked with the Georgia Tech Police
Department to come up with a more efficient way of allocating police officers. We measured
efficiency as how close police officers were to the crime location. In our semester-long project,
we cleaned up data, statistically analyzed them, and came up with an optimization model
that gives patrol location suggestions.
2 Methods – The PPAC Model
2.1 Data Cleaning
To be efficient in our data analysis, we had to process our data. First, we merged many
csv files with different information into one master file. Then, we read that master file with
R, converted its data type to dataframe and removed all null values. This resulted in a
significant number of data point loss, but we figured that only chose to go with quality over
quantity because we still had more than enough data. Then, we used built-in functions to
perform easy, but important statistical test to finalize our dataset. The resulting csv file
contained information (location, type, etc) of data-wise significant crime points.
2.2 Model Assumptions
The PPAC Model may be applied to data only if this assumption is followed: an acceptable
level of service from the police patrol units has been agreed upon relative to the desired level
of citizen safety[1]
.
2.3 General formulation
Our goal is to maximize the coverage of police officers. We will set a boundary which one
police officer can cover, and we will try to cover as many areas as possible with limited
number of police officers. On top of that, we would like to determine hot spots (areas of
concentrated crime) and prioritize, or, if possible, put more police offers to guard that area.
Mathematically, this can be represented as follows:
max
i∈I
aiyi
s.t.
j∈N
xi ≤ yi∀i ∈ I
xj ∈ {0, 1}
yi ∈ {0, 1}
2.4 Applied formulation
We used Python to write a program that can run this optimization function. After we ran
this code, we represented centers of patrol circles with dots. We present to you a sample of
our code.
3 Results & Analysis
We checked for optimality, and it turns out that our optimization model was able to generate
optimal patrol zones.
Further, we plotted these points, and we observed that we should re-zone Georgia Tech police
zones so that each zone has similar number of police officers patrolling at any given time.
That way, we can ensure better distribution of work force, and thus can increase our chance
of spotting crimes.
4 FUTURE WORK
Now that we were able to locate optimal patrol zones, we can work on re-zoning Georgia
Tech police patrol zones. Another problem to work on would be locating optimal patrol
zones depending on time. To do so, however, we will need a more comprehensive dataset. If
we have such a dataset, it would not be hard to repeat our work because we can simply add
time series analysis in our studies.
ACKNOWLEDGMENT
This work is supported by the PURA program, Georgia Tech Police Department, Atlanta
Police Department, and The H. Milton Stewart School of Industrial Engineering. I would
like to thank my faculty advisor David Goldberg, who was extremely supportive. Further,
I would like to thank my teammates Bingyi Bao, Hojin Lee, Davd Wang, and Yuanheng
Wang for helping me get through the work.
References
[1] K. M. Curtin et al., Integrating GIS and Maximal Covering Models to Determine Op-
timal Police Patrol Areas

More Related Content

What's hot

Using Data Mining Techniques to Analyze Crime Pattern
Using Data Mining Techniques to Analyze Crime PatternUsing Data Mining Techniques to Analyze Crime Pattern
Using Data Mining Techniques to Analyze Crime PatternZakaria Zubi
 
Analytics-Based Crime Prediction
Analytics-Based Crime PredictionAnalytics-Based Crime Prediction
Analytics-Based Crime PredictionProdapt Solutions
 
Predictive Policing - How Emerging Technologies Are Helping Prevent Crimes?
Predictive Policing - How Emerging Technologies Are Helping Prevent Crimes?Predictive Policing - How Emerging Technologies Are Helping Prevent Crimes?
Predictive Policing - How Emerging Technologies Are Helping Prevent Crimes?Sunil Jagani
 
Crime Pattern Detection using K-Means Clustering
Crime Pattern Detection using K-Means ClusteringCrime Pattern Detection using K-Means Clustering
Crime Pattern Detection using K-Means ClusteringReuben George
 
Crime prediction-using-data-mining
Crime prediction-using-data-miningCrime prediction-using-data-mining
Crime prediction-using-data-miningmohammed albash
 
Fundamentalsof Crime Mapping Tactical Analysis Concepts
Fundamentalsof Crime Mapping Tactical Analysis ConceptsFundamentalsof Crime Mapping Tactical Analysis Concepts
Fundamentalsof Crime Mapping Tactical Analysis ConceptsOsokop
 
A Comparative Study of Data Mining Methods to Analyzing Libyan National Crime...
A Comparative Study of Data Mining Methods to Analyzing Libyan National Crime...A Comparative Study of Data Mining Methods to Analyzing Libyan National Crime...
A Comparative Study of Data Mining Methods to Analyzing Libyan National Crime...Zakaria Zubi
 
Crime analysis of different situations
Crime analysis of different situationsCrime analysis of different situations
Crime analysis of different situationsKanukulaAkhil
 
Chicago Crime Dataset Project Proposal
Chicago Crime Dataset Project ProposalChicago Crime Dataset Project Proposal
Chicago Crime Dataset Project ProposalAashri Tandon
 
Crime Analysis & Prediction System
Crime Analysis & Prediction SystemCrime Analysis & Prediction System
Crime Analysis & Prediction SystemBigDataCloud
 
Discovery of ranking fraud for mobile apps
Discovery of ranking fraud for mobile appsDiscovery of ranking fraud for mobile apps
Discovery of ranking fraud for mobile appsNexgen Technology
 
Crime rate analysis using k nn in python
Crime rate analysis using k nn in python Crime rate analysis using k nn in python
Crime rate analysis using k nn in python CloudTechnologies
 
Machine Learning Approaches for Crime Pattern Detection
Machine Learning Approaches for Crime Pattern DetectionMachine Learning Approaches for Crime Pattern Detection
Machine Learning Approaches for Crime Pattern DetectionAPNIC
 
Us Pennsylvania State Police
Us Pennsylvania State PoliceUs Pennsylvania State Police
Us Pennsylvania State PoliceDawnStarling
 
Propose Data Mining AR-GA Model to Advance Crime analysis
Propose Data Mining AR-GA Model to Advance Crime analysisPropose Data Mining AR-GA Model to Advance Crime analysis
Propose Data Mining AR-GA Model to Advance Crime analysisIOSR Journals
 

What's hot (20)

Using Data Mining Techniques to Analyze Crime Pattern
Using Data Mining Techniques to Analyze Crime PatternUsing Data Mining Techniques to Analyze Crime Pattern
Using Data Mining Techniques to Analyze Crime Pattern
 
Analytics-Based Crime Prediction
Analytics-Based Crime PredictionAnalytics-Based Crime Prediction
Analytics-Based Crime Prediction
 
Predictive Policing - How Emerging Technologies Are Helping Prevent Crimes?
Predictive Policing - How Emerging Technologies Are Helping Prevent Crimes?Predictive Policing - How Emerging Technologies Are Helping Prevent Crimes?
Predictive Policing - How Emerging Technologies Are Helping Prevent Crimes?
 
Crime
CrimeCrime
Crime
 
Crime Pattern Detection using K-Means Clustering
Crime Pattern Detection using K-Means ClusteringCrime Pattern Detection using K-Means Clustering
Crime Pattern Detection using K-Means Clustering
 
Crime analysis
Crime analysisCrime analysis
Crime analysis
 
Crime prediction-using-data-mining
Crime prediction-using-data-miningCrime prediction-using-data-mining
Crime prediction-using-data-mining
 
Application of GIS in Criminology and Defence Intelligence
Application of GIS in Criminology and Defence IntelligenceApplication of GIS in Criminology and Defence Intelligence
Application of GIS in Criminology and Defence Intelligence
 
Crime analysis
Crime analysisCrime analysis
Crime analysis
 
Fundamentalsof Crime Mapping Tactical Analysis Concepts
Fundamentalsof Crime Mapping Tactical Analysis ConceptsFundamentalsof Crime Mapping Tactical Analysis Concepts
Fundamentalsof Crime Mapping Tactical Analysis Concepts
 
A Comparative Study of Data Mining Methods to Analyzing Libyan National Crime...
A Comparative Study of Data Mining Methods to Analyzing Libyan National Crime...A Comparative Study of Data Mining Methods to Analyzing Libyan National Crime...
A Comparative Study of Data Mining Methods to Analyzing Libyan National Crime...
 
Crime analysis of different situations
Crime analysis of different situationsCrime analysis of different situations
Crime analysis of different situations
 
Chicago Crime Dataset Project Proposal
Chicago Crime Dataset Project ProposalChicago Crime Dataset Project Proposal
Chicago Crime Dataset Project Proposal
 
Crime Analysis & Prediction System
Crime Analysis & Prediction SystemCrime Analysis & Prediction System
Crime Analysis & Prediction System
 
Discovery of ranking fraud for mobile apps
Discovery of ranking fraud for mobile appsDiscovery of ranking fraud for mobile apps
Discovery of ranking fraud for mobile apps
 
Crime rate analysis using k nn in python
Crime rate analysis using k nn in python Crime rate analysis using k nn in python
Crime rate analysis using k nn in python
 
Machine Learning Approaches for Crime Pattern Detection
Machine Learning Approaches for Crime Pattern DetectionMachine Learning Approaches for Crime Pattern Detection
Machine Learning Approaches for Crime Pattern Detection
 
Us Pennsylvania State Police
Us Pennsylvania State PoliceUs Pennsylvania State Police
Us Pennsylvania State Police
 
U24149153
U24149153U24149153
U24149153
 
Propose Data Mining AR-GA Model to Advance Crime analysis
Propose Data Mining AR-GA Model to Advance Crime analysisPropose Data Mining AR-GA Model to Advance Crime analysis
Propose Data Mining AR-GA Model to Advance Crime analysis
 

Similar to kyung_kim_pura

A predictive model for mapping crime using big data analytics
A predictive model for mapping crime using big data analyticsA predictive model for mapping crime using big data analytics
A predictive model for mapping crime using big data analyticseSAT Journals
 
Predictive Modeling for Topographical Analysis of Crime Rate
Predictive Modeling for Topographical Analysis of Crime RatePredictive Modeling for Topographical Analysis of Crime Rate
Predictive Modeling for Topographical Analysis of Crime RateIRJET Journal
 
Crime Data Analysis, Visualization and Prediction using Data Mining
Crime Data Analysis, Visualization and Prediction using Data MiningCrime Data Analysis, Visualization and Prediction using Data Mining
Crime Data Analysis, Visualization and Prediction using Data MiningAnavadya Shibu
 
Secure crime identification system
Secure crime identification systemSecure crime identification system
Secure crime identification systemSameer Telikicherla
 
Certain Analysis on Traffic Dataset based on Data Mining Algorithms
Certain Analysis on Traffic Dataset based on Data Mining AlgorithmsCertain Analysis on Traffic Dataset based on Data Mining Algorithms
Certain Analysis on Traffic Dataset based on Data Mining AlgorithmsIRJET Journal
 
Inspection of Certain RNN-ELM Algorithms for Societal Applications
Inspection of Certain RNN-ELM Algorithms for Societal ApplicationsInspection of Certain RNN-ELM Algorithms for Societal Applications
Inspection of Certain RNN-ELM Algorithms for Societal ApplicationsIRJET Journal
 
Richard Smith: Addressing the Problems of Addressing at British Transport Police
Richard Smith: Addressing the Problems of Addressing at British Transport PoliceRichard Smith: Addressing the Problems of Addressing at British Transport Police
Richard Smith: Addressing the Problems of Addressing at British Transport PoliceAGI Geocommunity
 
A Survey on Data Mining Techniques for Crime Hotspots Prediction
A Survey on Data Mining Techniques for Crime Hotspots PredictionA Survey on Data Mining Techniques for Crime Hotspots Prediction
A Survey on Data Mining Techniques for Crime Hotspots PredictionIJSRD
 
Fake News Detection using Passive Aggressive and Naïve Bayes
Fake News Detection using Passive Aggressive and Naïve BayesFake News Detection using Passive Aggressive and Naïve Bayes
Fake News Detection using Passive Aggressive and Naïve BayesIRJET Journal
 
San Francisco Crime Prediction Report
San Francisco Crime Prediction ReportSan Francisco Crime Prediction Report
San Francisco Crime Prediction ReportRohit Dandona
 
Story Tellers: Hartford Crime Analysis
Story Tellers: Hartford Crime AnalysisStory Tellers: Hartford Crime Analysis
Story Tellers: Hartford Crime AnalysisNeil Ryan
 
Determining Ideal Number of Police Patrols to Meet Reference Response Time Us...
Determining Ideal Number of Police Patrols to Meet Reference Response Time Us...Determining Ideal Number of Police Patrols to Meet Reference Response Time Us...
Determining Ideal Number of Police Patrols to Meet Reference Response Time Us...inventionjournals
 
Chicago Crime Analysis
Chicago Crime AnalysisChicago Crime Analysis
Chicago Crime AnalysisTom Donoghue
 
EGR Expo 2016 - EGR 402 - 38x48 poster - Easter (1)
EGR Expo 2016 - EGR 402 - 38x48 poster - Easter (1)EGR Expo 2016 - EGR 402 - 38x48 poster - Easter (1)
EGR Expo 2016 - EGR 402 - 38x48 poster - Easter (1)Jonathan Easter
 
San Francisco Crime Analysis Classification Kaggle contest
San Francisco Crime Analysis Classification Kaggle contestSan Francisco Crime Analysis Classification Kaggle contest
San Francisco Crime Analysis Classification Kaggle contestSameer Darekar
 
Criminal identification system
Criminal identification systemCriminal identification system
Criminal identification systemAkash Kumar Singh
 
Database and Analytics Programming - Project report
Database and Analytics Programming - Project reportDatabase and Analytics Programming - Project report
Database and Analytics Programming - Project reportsarthakkhare3
 
ŠVOČ: Design and architecture of a web applications for interactive display o...
ŠVOČ: Design and architecture of a web applications for interactive display o...ŠVOČ: Design and architecture of a web applications for interactive display o...
ŠVOČ: Design and architecture of a web applications for interactive display o...Martin Puškáč
 

Similar to kyung_kim_pura (20)

Cis second draft
Cis second draftCis second draft
Cis second draft
 
A predictive model for mapping crime using big data analytics
A predictive model for mapping crime using big data analyticsA predictive model for mapping crime using big data analytics
A predictive model for mapping crime using big data analytics
 
Predictive Modeling for Topographical Analysis of Crime Rate
Predictive Modeling for Topographical Analysis of Crime RatePredictive Modeling for Topographical Analysis of Crime Rate
Predictive Modeling for Topographical Analysis of Crime Rate
 
Crime Data Analysis, Visualization and Prediction using Data Mining
Crime Data Analysis, Visualization and Prediction using Data MiningCrime Data Analysis, Visualization and Prediction using Data Mining
Crime Data Analysis, Visualization and Prediction using Data Mining
 
Secure crime identification system
Secure crime identification systemSecure crime identification system
Secure crime identification system
 
Certain Analysis on Traffic Dataset based on Data Mining Algorithms
Certain Analysis on Traffic Dataset based on Data Mining AlgorithmsCertain Analysis on Traffic Dataset based on Data Mining Algorithms
Certain Analysis on Traffic Dataset based on Data Mining Algorithms
 
Inspection of Certain RNN-ELM Algorithms for Societal Applications
Inspection of Certain RNN-ELM Algorithms for Societal ApplicationsInspection of Certain RNN-ELM Algorithms for Societal Applications
Inspection of Certain RNN-ELM Algorithms for Societal Applications
 
Richard Smith: Addressing the Problems of Addressing at British Transport Police
Richard Smith: Addressing the Problems of Addressing at British Transport PoliceRichard Smith: Addressing the Problems of Addressing at British Transport Police
Richard Smith: Addressing the Problems of Addressing at British Transport Police
 
A Survey on Data Mining Techniques for Crime Hotspots Prediction
A Survey on Data Mining Techniques for Crime Hotspots PredictionA Survey on Data Mining Techniques for Crime Hotspots Prediction
A Survey on Data Mining Techniques for Crime Hotspots Prediction
 
Fake News Detection using Passive Aggressive and Naïve Bayes
Fake News Detection using Passive Aggressive and Naïve BayesFake News Detection using Passive Aggressive and Naïve Bayes
Fake News Detection using Passive Aggressive and Naïve Bayes
 
San Francisco Crime Prediction Report
San Francisco Crime Prediction ReportSan Francisco Crime Prediction Report
San Francisco Crime Prediction Report
 
Story Tellers: Hartford Crime Analysis
Story Tellers: Hartford Crime AnalysisStory Tellers: Hartford Crime Analysis
Story Tellers: Hartford Crime Analysis
 
Determining Ideal Number of Police Patrols to Meet Reference Response Time Us...
Determining Ideal Number of Police Patrols to Meet Reference Response Time Us...Determining Ideal Number of Police Patrols to Meet Reference Response Time Us...
Determining Ideal Number of Police Patrols to Meet Reference Response Time Us...
 
Chicago Crime Analysis
Chicago Crime AnalysisChicago Crime Analysis
Chicago Crime Analysis
 
EGR Expo 2016 - EGR 402 - 38x48 poster - Easter (1)
EGR Expo 2016 - EGR 402 - 38x48 poster - Easter (1)EGR Expo 2016 - EGR 402 - 38x48 poster - Easter (1)
EGR Expo 2016 - EGR 402 - 38x48 poster - Easter (1)
 
San Francisco Crime Analysis Classification Kaggle contest
San Francisco Crime Analysis Classification Kaggle contestSan Francisco Crime Analysis Classification Kaggle contest
San Francisco Crime Analysis Classification Kaggle contest
 
Criminal identification system
Criminal identification systemCriminal identification system
Criminal identification system
 
sdReport
sdReportsdReport
sdReport
 
Database and Analytics Programming - Project report
Database and Analytics Programming - Project reportDatabase and Analytics Programming - Project report
Database and Analytics Programming - Project report
 
ŠVOČ: Design and architecture of a web applications for interactive display o...
ŠVOČ: Design and architecture of a web applications for interactive display o...ŠVOČ: Design and architecture of a web applications for interactive display o...
ŠVOČ: Design and architecture of a web applications for interactive display o...
 

kyung_kim_pura

  • 1. Reallocating Patrol Zones Based on Previous Crime Patterns Kyung Kim December 10, 2015 Abstract This report will cover the application of the PPAC model on Georgia Tech Police Department (GTPD) crime records from 2011-2014. In Section I, we will talk about our motivation to do this research. In Section II, we will mathematically introduce, in detail, how we cleaned up the data and formulated the PPAC model. In Section III, we will present our results obtained by the application of the PPAC model. Fi- nally in Section IV, we will entertain the idea of moving forward with our research by introducing future directions. 1 Introduction It is time for high school students to apply to colleges, and campus safety is one aspect that should not be overlooked. Although Georgia Tech is one of the safest schools, to further relieve students from having these types of concerns, we worked with the Georgia Tech Police Department to come up with a more efficient way of allocating police officers. We measured efficiency as how close police officers were to the crime location. In our semester-long project, we cleaned up data, statistically analyzed them, and came up with an optimization model that gives patrol location suggestions. 2 Methods – The PPAC Model 2.1 Data Cleaning To be efficient in our data analysis, we had to process our data. First, we merged many csv files with different information into one master file. Then, we read that master file with R, converted its data type to dataframe and removed all null values. This resulted in a significant number of data point loss, but we figured that only chose to go with quality over quantity because we still had more than enough data. Then, we used built-in functions to perform easy, but important statistical test to finalize our dataset. The resulting csv file contained information (location, type, etc) of data-wise significant crime points.
  • 2. 2.2 Model Assumptions The PPAC Model may be applied to data only if this assumption is followed: an acceptable level of service from the police patrol units has been agreed upon relative to the desired level of citizen safety[1] . 2.3 General formulation Our goal is to maximize the coverage of police officers. We will set a boundary which one police officer can cover, and we will try to cover as many areas as possible with limited number of police officers. On top of that, we would like to determine hot spots (areas of concentrated crime) and prioritize, or, if possible, put more police offers to guard that area. Mathematically, this can be represented as follows: max i∈I aiyi s.t. j∈N xi ≤ yi∀i ∈ I xj ∈ {0, 1} yi ∈ {0, 1} 2.4 Applied formulation We used Python to write a program that can run this optimization function. After we ran this code, we represented centers of patrol circles with dots. We present to you a sample of our code.
  • 3. 3 Results & Analysis We checked for optimality, and it turns out that our optimization model was able to generate optimal patrol zones. Further, we plotted these points, and we observed that we should re-zone Georgia Tech police zones so that each zone has similar number of police officers patrolling at any given time. That way, we can ensure better distribution of work force, and thus can increase our chance of spotting crimes.
  • 4. 4 FUTURE WORK Now that we were able to locate optimal patrol zones, we can work on re-zoning Georgia Tech police patrol zones. Another problem to work on would be locating optimal patrol zones depending on time. To do so, however, we will need a more comprehensive dataset. If we have such a dataset, it would not be hard to repeat our work because we can simply add time series analysis in our studies. ACKNOWLEDGMENT This work is supported by the PURA program, Georgia Tech Police Department, Atlanta Police Department, and The H. Milton Stewart School of Industrial Engineering. I would like to thank my faculty advisor David Goldberg, who was extremely supportive. Further, I would like to thank my teammates Bingyi Bao, Hojin Lee, Davd Wang, and Yuanheng Wang for helping me get through the work. References [1] K. M. Curtin et al., Integrating GIS and Maximal Covering Models to Determine Op- timal Police Patrol Areas