1 Crime Mapping and Spatial Analysis PDF

Crime Mapping and
Spatial Analysis
Mostafa Ahmadi
February 2003
Crime Mapping and Spatial Analysis
By
Mostafa Ahmadi
ITC Primary supervisor: Y. Sun.MSc

ITC Secondary supervisor: Dr. A. Sharifi
Iranian supervisor: Dr. M. J. Valadan
Thesis submitted to the International Institute for Geo-information Science and

Earth Observation in
partial fulfilment of the requirements for the degree of Master of Science in
Geoinformatics.
Degree Assessment Board

Prof.Dr. W. Van den Toorn (Chairman) – Social Sciences Division, ITC
Prof. Dr. A. van der Veen (External examiner) – Spatial Economic, University of
Twente
Mrs Y. Sun (1st Supervisor) – Geoinformatics Division, ITC
Dr. M. A. Sharifi (2nd Supervisor) – Social Science Division, ITC
Dr. M. Sharif (Member) – Earth Observation Sciences, ITC
Dr. H. Ebadi (Member) – KNTousi University of Iran
INTERNATIONAL INSTITUTE FOR GEO-INFORMATION SCIENCE AND EARTH OBSERVATION

ENSCHEDE, THE NETHERLANDS
II
Disclaimer
This document describes work undertaken as part of a programme of study at the
International
Institute for Geo-information Science and Earth Observation. All views and opinions
expressed
therein remain the sole responsibility of the author, and do not necessarily
represent those of
the institute.
III
ABSTRACT
Crime maps are becoming significant tools in crime and justice. Advances in the
areas of information technology and Geographic Information Systems (GIS) have
opened new opportunities
for the use of digital mapping in crime control and prevention programs. Crime maps
are also
valuable for the study of the ecology and the locational aspects of crime. Maps
enable areas of
unusually high or low concentration of crime to be visually identified. Maps are
however only
pictorial representations of the results of more or less complex spatial data
analyses.
A hierarchical model dealing with crime analysis is proposed and applied to the
regional analysis of crime in Tehran, the model helps to identify spatial
concentration of crimes in specific
area (area based method). In area-based methods, crime data are aggregated into
geographical
areas such as blocks, precincts, and for each area, the analyst computes a measure
of crime
value. Multicriteria evaluation concept has been used to assess the crime rate in
various blocks a
discrete (part) of Tehran city. In this part we used two methods for crime density
assessment:
•
•
Crime assessment based on crime per block,
Crime assessment based on density of crime per population.

After determination of hot spots based on two methods mentioned above spatial
function is
used to find suitable location to establish new police station or direct patrol to
the hot spots to
reduce of crime.
IV
Acknowledgements
I would like to express my Sincere and heartfelt gratitude to the Law enforcement
of Islamic
Republic of IRAN (NAJA) for sponsoring me to pursue this study without which I
would not
have realized my dream to continue my studies. I am grateful to my former employer,
Commission for Sustainable crime mapping of Tehran who through the head of
educational deputy
Brigadier Mohammad Ghasri and Brigadier Safari complemented my efforts by
supporting me
to fulfil my wish.
My thanks go to all the staff of GFM for the support and guidance throughout the
modules and
thesis preparation. A special thanks goes to my Supervisor Mrs. Yuxian Sun, for the
guidance
and critical comments that made this research a success. My gratitude also goes to
my second
supervisor, Dr. A. Sharifi, for his effort and guidance me.
I would also like to thank the University of Khajeh-Nasir-Tousi in IRAN, and
particularly
grateful to my Iranian supervisor DR .M. J. Valadan, and second Iranian supervisor
A. Mansourian, MSc.
My heartfelt gratitude goes to my wife for her patience, support and encouragement
and those
special words especially through the hardest times gave me the courage to continue.
Special
thanks go to all my dear parents who have been my mentors always supportive and
urging me
to go on.
V
List of content
ABSTRACT ..........................................................................
............................................................. IV
ACKNOWLEDGEMENTS...................................................................
.............................................V
LIST OF
CONTENT ...........................................................................
............................................. VI
LIST OF
TABLES ............................................................................
............................................. VIII
LIST OF
FIGURES ...........................................................................
............................................... IX
1.
INTRODUCTION ......................................................................
..................................................1
1.1.
1.2.
1.3.
1.4.
1.5.
1.6.
1.7.
2.
INTRODUCTION OF TEHRAN
POLICE ............................................................................
...........1
PROBLEM
DEFINITION ........................................................................
.....................................1
RESEARCH
QUESTIONS .........................................................................
..................................2
OBJECTIVES ........................................................................
....................................................2
METHODS ...........................................................................
....................................................2
CASE
STUDY .............................................................................
..............................................3
STRUCTURE OF THE
THESIS.............................................................................
........................4
LITERATURE REVIEW ON CRIME

ANALYSIS .................................................................5
2.1. DEFINITION OF CRIME AND CRIME
ANALYSIS .........................................................................5
2.1.1.
Crime
definition ........................................................................
.....................................5
2.1.2.
Types of crime
analysis ..........................................................................
........................6
2.2. METHODS FOR AUTOMATING THE GEOGRAPHICAL ANALYSIS OF CRIME INCIDENT DATA......7
2.2.1.
The Geographic Analysis
Machine ...........................................................................
.....7
2.2.2.
The Geographical Explanations
Machine .....................................................................8
2.3. SPATIAL PERSPECTIVES ON
CRIME .............................................................................
..........10
2.4. TOOLS IN THE SPATIAL ANALYSIS OF
CRIME ........................................................................11
3.
DATA COLLECTION AND

PREPARATION.......................................................................1
5
3.1. DATA MODELLING IN
GIS ...............................................................................
......................15
3.2. DATA
COLLECTION ........................................................................
.......................................17
3.3. DATA
PREPARATION .......................................................................
......................................19
3.4. DATABASE MAGEMENT
SYSTEM ............................................................................
..............20
3.4.1.
The Entity Relationship (ER)
diagram.........................................................................21
3.4.2.
Description of
entities ..........................................................................
........................22
3.4.3.
Enterprise
Rules .............................................................................
..............................22
3.4.4.
Skeleton
tables.............................................................................
.................................23
3.4.5.
Data
dictionary ........................................................................
....................................23
4.
MULTICRITERIA DECISION
MAKING ............................................................................
.24
4.1.
MULTI CRITERIA EVALUATION TO SUPPORT DECISION
MAKING ........................................25
VI
4.2. CRIME DEFINITION AND
TYPE ..............................................................................
.................26
4.3. MULTICRITERIA EVALUATION APPLIED FOR CRIME
ASSESSMENTS ......................................27
4.3.1.
Crime assessment based on crime per
block................................................................31
4.3.2.
Crime assessment based on density of
crime ...............................................................39
4.4. ANALYSIS OF THE
RESULT ............................................................................
........................44
5.
ANALYSIS...........................................................................
.......................................................45
6.
CONCLUSION AND
RECOMMENDATION.....................................................................
...50
6.1.
6.2.
CONCLOUTION .......................................................................
...............................................50
RECOMMENDATION ....................................................................
..........................................50
REFERENCES ........................................................................
............................................................. I
APPENDICES ........................................................................
........................................................... III
Appendex1: The form of crime incidents in
blotters:................................................................... iii
Appendex2: Trade-off
procedure .........................................................................
........................ iii
Appendix3: Spatial
function...........................................................................
...............................iv
VII
List of tables
TABLE 3.1 COMPARISON OF RASTER AND VECTOR
DATA.....................................................................16
TABLE 3.2 CENSUS TYPES OF CRIMES IN THE POLICE STATION, NUMBER
109.....................................19
TABLE 3.3 COLLECTED DATASET AND THEIR
FORMAT.........................................................................19
TABLE 3.4 DESCRIPTIONS OF ENTITY TYPES AND THEIR
ATTRIBUTES .................................................23
TABLE 4.1 FREQUENCY TABLES OF CRIMES RELATED TO
SECURITY....................................................30
TABLE 4.2 FREQUENCY TABLES OF CRIME RELATED TO
ECONOMIC ....................................................30
TABLE 4.3 FREQUENCY TABLES OF CRIMES RELATED TO
SOCIAL ........................................................30
TABLE 4.4 NORMALIZED TABLE (RATIO SCALE) PRESENTING THE CRIME RELATED TO
SECURITY ......31
ECONOMIC ....31
SOCIAL ..........32
TABLE 4.7 PAIR WISE COMPARISON TABLE OF CRIME RELATED TO SECURITY (BY
EXPERT)...............33
TABLE 4.8 PAIR WISE COMPARISON TABLE OF CRIME RELATED TO ECONOMIC (BY
EXPERT) .............33
TABLE 4.9 PAIR WISE COMPARISON TABLE OF CRIME RELATED TO SOCIAL (BY
EXPERT) ...................33
TABLE 4.10 NORMALIZED OF PAIR WISE COMPARISON OF CRIME RELATED TO
SECURITY ..................34
ECONOMIC .................34
SOCIAL ......................34
TABLE 4.13 RANKING THE
ALTERNATIVES ......................................................................
....................35
TABLE 4.14 EFFECT
TABLE .............................................................................
......................................35
TABLE 4.15 PREFERENCE OF SECURITY
PERSPECTIVE .......................................................................
..36
TABLE 4.16 PREFERENCE OF ECONOMIC
PERSPECTIVE .......................................................................
.36
TABLE 4.17 PREFERENCE OF SOCIAL
PERSPECTIVE .......................................................................
......37
TABLE 4.18 POPULATION PER
BLOCK..............................................................................
.....................39
TABLE 4.19 DENSITY TABLES PRESENTING THE CRIME RELATED TO
SECURITY ..................................40
TABLE 4.20 DENSITY TABLES PRESENTING THE CRIME RELATED TO
ECONOMIC .................................40
TABLE 4.21 DENSITY TABLE PRESENTING THE CRIME RELATED TO
SOCIAL ........................................40
TABLE 4.22 RANKING THE
ALTERNATIVES ......................................................................
....................41
TABLE 4.23 EFFECT
TABLE .............................................................................
......................................41
TABLE 5.1 CRIME VALUE OF
BLOCKS ............................................................................
.......................45
VIII
List of Figures
FIGURE 1.1 CENSUS DISTRICTS OF
TEHRAN ............................................................................
...............3
FIGURE 1.2 DISTRICTS 12 (CASE
STUDY) ............................................................................
...................4
FIGURE 3.1 LEVELS OF GIS DATA MODELLING (LONGLEY
2001) ........................................................15
FIGURE 3.2 POLICE ORGANIZATION IN
TEHRAN ............................................................................
.......18
FIGURE 3.3 A SAMPLE ER
DIAGRAMS. .........................................................................
........................22
FIGURE 4.1 GENERAL MODEL OF MADM, JANKOWSKI,
1995.............................................................26
FIGURE 4.2 GENERAL MODEL OF MCE OF
CRIME .............................................................................
...28
FIGURE 4.3 OVERALL ASSESSMENT FOR
SECURITY ..........................................................................
....36
FIGURE 4.4 OVERALL ASSESSMENT FOR
ECONOMIC ..........................................................................
..37
FIGURE 4.5 OVERALL ASSESSMENTS FOR
SOCIAL ............................................................................
....38
FIGURE 4.6 OVERALL ASSESSMENT OF CRIME DENSITY FOR
SECURITY ...............................................42
ECONOMIC .............................................42
SOCIAL ...................................................43
FIGURE 5.1 METHOD FOR ESTABLISH NEW POLICE
STATION ................................................................46
FIGURE 5.2 SHOW HOTSPOTS AREA IN (A), AND BUFFERS OF POLICE STATIONS
(B) ...........................47
FIGURE 5.3 MAP CALCULATION OF HOT SPOT AND BUFFERS OF POLICE STATION IN (A), AND
OVERLAY
OF MAP CALCULATION WITH BUFFER OF MAIN STREETS IN
(B) ....................................................47
FIGURE 5.4 SUITABLE AREAS FOR NEW POLICE
STATION .....................................................................48
FIGURE 5.5 RELOCATION OF
BOUNDARY ..........................................................................
...................48
FIGURE 5.6 SUGGESTION OF NEW BOUNDARY FOR ESTABLISHING NEW POLICE
STATIONS..................49
IX
CRIME MAPPING AND SPATIAL ANALYSIS
1. Introduction
1.1.
Introduction of Tehran police
Crime mapping and spatial analysis of crime are recognized as powerful tools for
the study and control of crime, because crime maps help police identify problems at
the block. The most powerful
weapon in law enforcement is information technology.
The application areas of GIS in law enforcement are relatively narrow, because:
Although police forces are using GIS technology, the penetration of GIS into every
application is still relatively limited worldwide.
In Iran GIS has not yet been viewed as an obligatory technology within the police
force,
whereas organizations such as the utilities have taken GIS on board, this is due to
high cost of
GIS tools and lack of GIS experts in the police force.
Law enforcement needs information management, especially location information.

Traditional law
enforcement for different types of police applications really deals with data
collection. However, data
collection without data analysis is useless. Not only GIS allows integration and
spatial analysis of
data to identify, apprehend, and prosecute suspects, it also aids more proactive
measures through effective allocation of resources and better policy setting.
Definition of crime mapping, crime analysis and crime prevention:
Crime mapping involves the manipulation and processing of spatially referenced
crime data in order
to display visually in an output that is informative to the particular user (Alex
and Kate 2001). Crime
mapping can provide information concerning the location of hotspots or high level
of reported crime.
Crime analysis is a set of processes applied on relevant information about crime
patterns. Administrative and operational personal can use the result of analysis to
prevent and suppress of criminal activities and also for investigation aims.
Crime prevention seeks to reduce the risks of criminal events and related anti-
social behaviour by intervening in their causes.
1.2.
Problem definition
It is quite obvious that if the police authorities have goals of improving crime
management (crime
prevention) in the city the manual recordings of crime incidents cannot be the way
forward .
1
The various police stations under the Tehran police office have the responsibility
of handling the
various crime incidences in the boundary under their jurisdiction but at the moment
have problems
with their recording system, and aggregation of crimes incidents in all police
stations.
All the recordings that are done in the police station are manually written in
blotters (police blotter is
a record of all arrests, crime events, and charges. see Appendix.1), and it is
difficult to do decision
making for all of police stations. There are not systems for:
•
•
•
•
•
Crime mapping, incident time and place.

To inform the targeting of resources for crime prevention
For police investigations
To evaluate the effectiveness of crime prevention initiatives.
Assistance in the prevention and rapid response to crime.
Research questions
1.3.
What is the best methodology for crime spatial analysis?

What parameters can be useful for crime mapping and spatial analysis?
Where does the crime tend to concentrate (density)?
How to find the suitable location for establishing a new police station?
1.4.
Objectives
General
The main objective of this project is to develop a methodology for crime spatial
analysis for Tehran
police force.
Details
-
1.5.
Find a method to relocate in current boundary of police station for crime

prevention or reduction,
Identify suitable location for new police station,
Analysis of factors, which are effective for crime prevention.
Methods
The procedure used to achieve the aforementioned objectives is as follows:
•
•
•
•
•
Review of the prior works to get knowledge about them.

Search in blotters of the police station.
In order to obtain the crime causes several interviews were done with authorities
of police stations.
Data collection (attribute and spatial), and preparation.
Multi criteria analysis.

Police have learned from experience that there are particular environments that
concentration of
crimes is larger-than-expected. Sometimes these hot spot areas are defined by
particular activities
(e.g., burglary), other times by specific concentrations of land uses (e.g., bus
station), and sometimes
by interactions between activities and land uses, such as shoplifting in commercial
areas or pickpocket in bus or bus stop.
2
Based on the above-mentioned vector representation will be used for hot spots and
multicriteria decision-making is used for detection of hot spots for crime
prevention.
•
•
•
To find Hot Spots (A location where variety of crimes keep occurring over an
extended period of time).
To find Relationship between crime and geographic environment.
To find suitable location for establishing new police station.

For this research needs access to computer hardware, GIS software, GIS operation
for crime analysis,
and crime data.
1.6.
Case study
Tehran, is the largest and fastest growing city in IRAN has grown from a population
of just more than
half a million in 1940 to 5.4 million in 1980 and has currently a population of
about 9 to 10 million.
The physical expansion of the city has also been rapid. Urban growth from the core
has been much
influenced by the environmental contrasts between north and south. The modern
commercial sector
together with many of the suburbs where the wealthier section of the population
live, have developed
northward from the old core, while the poorer suburbs have developed south towards
the desert. Tehran consists of 22 census districts (Figure 1.1):
Figure 1.1 Census districts of Tehran
The dataset used in the case study were crime data from Central County (BAZAR) of
Tehran (district
12). It covers an area of 11 square kilometres and has a population residential
1.1million, and population non-residential over 3 million, an excellent example of
the partitioning of residential, commercial
3
and political urban space. BAZAR of Tehran is a part of the commercial area in
Tehran (Figure 1.2).
This BAZAR was selected because of the relationship between crime and commercial
environment.
Study area
Figure 1.2 Districts 12 (case study)
1.7.
Structure of the thesis
The report on the crime mapping and spatial analysis in Tehran city has been
arranged in a rather
classical set-up. Chapter 1 to 3 provide an introduction to Tehran police, and
problem definitions
(chapter 1), literature review on crime analysis (chapter 2), data collection and
data modelling in GIS
(chapter 3).
Chapter 4 to 6 analysis and subsequently focus on the processes of using
Multicriteria evaluation
based on block and density population (chapter 4), the result of analysis and
determination of suitable
blocks for establishing new police station and discussion about it (chapter 5),
conclusion and some
suggestion for future work (chapter 6).
4
2. Literature review on crime analysis

Unfortunately there is no experience about GIS (crime mapping) in the Tehran police
organization
and it is not possible to find any research or article about this case, not only in
my country but also in
Asia. Because of these reasons, articles written about the United Kingdom, United
state of America,
and Australia will be used in the literature review.
This chapter consists of the following parts:
2.1 – Definition of crime, and crime analysis.
2.2 – Methods for automating the geographical analysis of crime incident data.
2.3 – Spatial perspectives on crime.
2.4 – Tools in the spatial analysis of crime.
2.1.
Definition of crime and crime analysis
Introduction
As an introduction to crime analysis, this section provides the definition of crime
and crime analysis
as a general concept as well as definitions of six types of crime analysis. These
definitions are meant
to enhance the understanding of crime analysis and to help create commonly
understood terminology.
2.1.1. Crime definition
Crime is a multifaceted concept that can be defined in legal and non-legal sense.
From a legal point of
view, it refers to breaches of the criminal laws that govern particular geographic
areas (jurisdictions)
and are aimed at protecting the lives, property and rights of citizens within those
jurisdictions. Most
of the crimes with which the criminal justice system is concerned involve breaches
of State/Territory
legislation that cover most offences relating to persons (for example, murder and
sexual assault),
property (for example, theft and property damage) and regulation (for example,
traffic violations).
Commonwealth legislation relates primarily to matters such as trade and commerce,
importation/exportation, taxation, defence and external affairs (1).
Non-legal point of view would define crime as acts that violate socially accepted
rules of human ethical or moral behaviour. As the moral principles that underpin
the notion of crime are subject to gradual change over time, the types of behaviour
defined by the legal system as criminal may also change.
Examples of behaviours that have been de-criminalized in some jurisdictions include
prostitution,
abortion, attempted suicide and homosexual intercourse. Other behaviours, such as
tax evasion or
credit card fraud, have been criminalized over time (1).
The distinction between these two points of view is important. While criminal
justice agencies' crime
data will be based upon the relevant legal definition of crime, data collected via
victimisation surveys
may be based upon an individual's interpretation of crime, rather than upon
existing legal definitions.
This potential disparity is more likely to affect the less serious end of the
'crime' spectrum.
5
2.1.2. Types of crime analysis

The following are six types of analysis of crime. As you will see, each contains
characteristics of
crime analysis in general, but each is specific in the type of data and analysis
used as well as in its
purpose.
Not all law enforcement analysts do the same type of work. There are several
different types of policing analysis. Crime analysts spend varying percentages of
their time on the following types of analysis
(1):
• Tactical Crime Analysis:

This is day-to-day crime analysis, looking for series, patterns, sprees, hot spots,
and hot dots immediately affecting the jurisdiction. Tactical crime analysis also
focuses on specific information about
each crime such as method of entry, point of entry, suspects actions, type of
victim, type of weapon
used, as well as the date, time, location, and type of location. Field information
such as suspicious
activity calls for service, criminal trespass warnings, and persons with scars,
marks, or tattoos collected by officers is also considered in the analysis. Used
for:
Day to day
For series, patterns, sprees, hot spots
Used for Deployment & Administration
• Strategic Crime Analysis:
The study of crime and law enforcement information integrated with socio-
demographic and spatial
factors to determine long term “patterns” of activity, to assist in problem
solving, as well as to research and evaluate responses and procedures. Used for:
Identify unusual activity levels by time or location.
Forecasting potential crime events/concentration.
• Administrative/Academic Crime Analysis:
The study of crime and law enforcement information integrated with socio-
demographic and spatial
factors to determine long term “patterns” of activity, to assist in problem
solving, as well as to research and evaluate responses and procedures. Used for:
Reports or statistical summaries for grant funding, commanders & public
Policy implications beyond law enforcement agency.
•
Operations Analysis:
The study of policing practices is regards patrol and resource allocation (e.g., a
study of police overtime assignments last year). Operational analysis coupled with
Strategic Crime Analysis, helps patrol
commanders to make changes that use resources more efficiently. Used for:
Assess needs (calls for service, population of data & demographics)
Generate projections for deployment & resource allocation
-
Intelligence Analysis:
The study of criminal organizations and enterprises, how they are linked, who the
key players are.
Helps investigation and prosecution units within police. The purpose of
intelligence analysis is to as-
6
sist sworn personnel in the identification of networks and apprehension of

individuals to subsequently
prevent criminal activity. Used for:
- Linkage between crime organizations & enterprises
- Relate elements such as companies, agencies, people, times, days, to crimes &
places
Investigative Analysis:
Looks at crime scene, psychological, and forensic analysis used in major crimes. It
also helps catch
serial killers, arsonists, and similar criminals. The primary purpose of criminal
investigative analysis
is to develop patterns of serial crimes crossing city, state, and even national
boundaries by linking behavior and evidence within and among incidents in order to
catch the offender and/or clear cases.
Used for:
Crime scene, psychological & forensic information.
Link serial or related events.
-
2.2.
Methods for automating the geographical analysis of crime incident data
Introduction
This part considers two ways of automating the geographical analysis of crime
locations. Almost all
police stations collect crime reports with a form of postcode of the building
nearest the incident, it is
possible for police organization conjunction with automated analysis technique, and
it is possible for
police department to discover crime hotspots in data.
2.2.1. The Geographic Analysis Machine
The Geographical Analysis Machine (GAM) system (Alex and Kate 2001) considers way
in which
crime analysis can be automated. This system allows crime analysts without detailed
geographical
knowledge to perform geographically sound and rigorous analysis of their data. Two
methods are introduced: the Geographical Analysis Machine (GAM) which shows the
user where there are clusters
or hotspots in their data, and the Geographical Explanations Machine (GEM) which
points to geographical data sets that may explain the clusters found by the GAM.
Almost all law enforcement
agencies collect crime reports with a form of geocode attached; in the UK this is
usually the postcode
of the building nearest the incident. By making use of this locational data in
conjunction with automated analysis techniques, it is possible for agencies to
discover crime hotspots in their data, allowing
the targeting of resources to these areas. At first this matter to introduces the
geographical analysis
machine (GAM) and then discusses the geographical explanation machine (GEM), which
makes use
of related geographical variables to attempt to explain the pattern found by GAM,
and GEM can be
used to investigate crime data.
The geographical analysis machine (Openshaw, Charlton et al. 1987) was an early
attempt at automated exploratory spatial data analysis that was easy to understand.
The GAM sought to answer a
simple practical question: namely given some point referenced data of something
interesting where
might there be evidence of localized clustering if you do not know in advance where
to look due to
lack of knowledge of a possible casual mechanism or if prior knowledge of the data
precluded testing
hypotheses on the database? More simply put, here is a geographically referenced
database, now tell
me if there are any crime clusters or hotspots and if so where are they located? It
offers a solution to
those researchers and users of GIS who want to perform a fast exploratory
geographical analysis of
7
their data with minimum of effort. It is an automated procedure that is designed to

yield safe results
that are largely self-evident (Alex and Kate 2001).
GAM reflects the view that useful spatial tools have to be able to cope with both
the spatial data and
end users who do not have degrees in statistics. The results have also to be easily
understood and selfevident so that they can be readily communicated to other non-
experts.
How does GAM work? The GAM algorithm involves the following steps:
1. Read in X, Y data for population at risk and a variable of interest (crime) from
a GIS.
2. Identify the rectangle containing the data; identify starting circle radius and
degree of overlap.
3. Generate a grid covering this rectangle so those circles of current radius
overlap by the desired amount.
4. For each grid intersection generate a circle of radius.
5. Retrieve two counts for the population at risk and the variable of interest.
6. Apply some ‘significance’ test procedure.
7. Keep the result if significant.
8. Repeat steps 5 to 7 until all circle have been processed.
9. Increase circle radius and return to step 3 or else go step 10.
10. Create a smoothed density surface of excess incidence for the significant
circle using a kernel
smoothing procedure and aggregating the result for all circles.
11. Maps this surface.
The choice of significance test is not considered as being too critical. The aim is
not to test conventional hypotheses, but merely to determine whether or not an
observed positive excess incidence is
sufficiently large to be unusual and hence of interest. It is more a measure of
unusualness or surprise
than a formal statistical significance test. A number of different measures of
‘unusualness’ can be applied depending on the rarity of incidence of interest, e.g.
Poisson, bootstrapped z-scores, and Monte
Carlo test based on rates. The aim here is not a formal test of significance,
instead; significance’ is
being used only as a descriptive filter employed to reject circles. It is map
created by the overall distribution of significant circles that is of most
interest.
2.2.2. The Geographical Explanations Machine
GAM is purely a pattern detector and no explanation is provided that can help the
user understand
what variables may be geographically correlated with any clusters it may find. The
search for explanation using a GIS reduces to multiple spatial queries to find out
what appears to be geographically associated with the clusters. There is nothing
wrong with this approach but it would be better if the procedure could be automated
so that it could be made far more comprehensive and thus rigorous.
It is also possible that the apparent patterns being uncovered by GAM merely
reflect confounding
variables and those they are, therefore, of no real consequence. There is also a
risk that a good understanding of what the patterns of cluster mean will be greatly
hindered by missing information. Indeed,
there are many different possible variables that may be considered important. Some
of these could
well be related more to individual behaviour, others to geographical location, and
some to both (Alex
and Kate 2001).
8
How does GEM (Alex and Kate 2001) work? The GEM algorithm involves the following
steps:
1. Read X, Y data for population at risk and a variable of interest from a GIS.
Also read in the
X, Y data classified by value for each of overages.
2. Identify the rectangle containing the data set a starting circle radius, and the
degree of circle
overlap.
3. Generate a grid covering this rectangular map area so that circles of current
radius overlap by
the desired amount.
4. For each grid intersection generate a circle of radius.
5. Retrieve two counts for the population at risk and the variable of interest.
6. Apply some significance test procedure.
7. If the result is not significant go to step 5 and investigate another location,
or store summary
of results to permit later identification of recurrent or important coverage
permutations.
8. Repeat step 5 to 7 until all circle have been processed.
9. Increase circle radius and return to step 3 or else go to step 10.
10. Create smoothed density surface of excess incidence for the significant circle
for selected
combinations of coverage permutations using a kernel smoothing procedure and
aggregating
the result for all circle.
11. Map this surface.
This system has demonstrated that with a little effort and very little GIS know-how
it is possible for
an analyst to produce maps of crime hotspots in their area using the geographical
analysis machine.
The discussion of GEM has show how with a little more data from their GIS it is
possible to start to
look for explanatory geographic variables that may assist police officers on the
ground to deal with
the hotspots found.
Several standard types of crime analyses have proven helpful to patrol officers:
automated pin mapping, hotspot analysis, and radial analysis are a few of the most
extensively used. Automated pin
mapping is the most rudimentary and in some ways the most flexible of these uses.
Law enforcement
practitioners have begun to realize the advantages of maps over printouts in
enabling officers to visualize spatial patterns of criminal events.
Mapping provides the capability of displaying any subset of events on a map. Not
only can use the
user specify the time period they want to examine, they can also display events of
a certain type or
that meet specific criteria. By enabling the visualization of subsets of
information, mapping provides
an invaluable tool for revealing clusters and patterns of crime that are not
readily apparent from a list
of crime events in a report. Another important function that mapping enable, is the
visualization of the
concentration of events at a single address. This is accomplished by tying the size
of the symbol at a
location to the number of events occurring there: the more events, the larger the
symbol. This method
for identifying report events at a single address supports problem-oriented
policing efforts by making
locations with several calls easily identifiable.
Conclusion
This part has demonstrated with a little effort and very little GIS Know-how it is
possible for an analyst to produce maps of crime hotspots in their area using the
GAM.
9
GEM has shown how with a little more data from their GIS it is possible to start to
look for explanatory geographic variables that may well assist police officers on
the ground to deal with the hotspots
found.
The GAM looks for evidence of clustering in crime data, whereas the GEM goes a step
further and
investigates information, such as census data. The advantage of using these tools
is that they include
statistical testing which can assess the significance of any clusters of crime or
relationships with other
geographical information.
2.3.
Spatial perspectives on crime
“Although criminological research reaches back over some two centuries,

professional geographers
have not been directly involved until the recent past”(Herbert 1982).
The study of crime has, not surprisingly, been dominated by research in
criminology, sociology and
law, but spatial and ecological perspectives on crime by mostly criminologists, did
precede the ‘recent
past’ after which professional geographers entered the crime research arena. Lawman
(1986) identifies four major stages, which the spatial and ecological perspectives
on crime underwent during the
past 150 years. First, there was the cartographic school of criminology, which
started in France, and
spread to other European countries, most notably England, from 1830 t0 1880. The
second stage was
the Chicago ecological school of the 1920s and 1930s, followed by a third stage,
the factor analytic
school of the 1950s. The final and current stage is the geography of crime and
environmental criminology (Lawman 1982).
The cartographic school was called that because of the frequent use of maps to
depict regional and
seasonal variations in pattern of crime. One of the French exponents of the
cartographic school was
Guerry, cited in Herbert (1982), who mostly concentrated on the urban-rural
differences in crime occurrence. Guerry also identified different seasonal
patterns, concluding that crimes against property
were more frequent during winter in the north of France, whilst the south had more
incidence of crime
s against the person during summer. Followers of the nineteenth century
cartographic school in England, such as Tobias and Mayhew, were the first to
isolate cities as the focus of their studies. They
concluded that there were different source areas of criminals, and that areas exist
within cities where
more crime occur than in other areas (Herbert 1982).
The spatial ecology of crime was developed at the Chicago school of criminology
through the work of
Shaw and McKay ((Herbert 1982); Dun 1980). Shaw and McKay used data collected on
crime in Chicago to map the homes of juvenile offenders through the use of dot
maps, to show actual distributions,
and crime rate maps, to show aerial variations in delinquency residences (Herbert
1982). These maps
indicated a regular change from the centre to the periphery of the city, in terms
of the number of offender residences, prompting the conclusion that delinquents are
concentrated near the city centre.
These “delinquency areas” (Dun 1980) were correlated with variables related to its
physical and social environments such as “substandard housing, poverty, foreign-
born population and mobility”
(Herbert 1982). This ecological analysis of areas with a large number of
delinquents, in terms of the
above variables, showed that the communities within these areas “tended to function
least efficiently
as an agent of social control” (Dunn 1980). This explanation led to the social
disorganization theory
10
(Bottoms and Wiles 1995), which proved to become either the source of constant
debate or central
principle within all disciplines, geography included, dealing with crime.
Both Herbert (1982) and Lawman (1986) argued that the social disorganization theory
was not based
on a biotic model of ecology. Instead, it interprets delinquency only in terms of
social, cultural and
economic factors, and not an ecological adaptation to different areas (Herbert
1982). This severe criticism of the social disorganization theory caused the
ecological approach to crime to wane within
criminology.
Lawman (1986) cited the factor analytic school as the third stage in the
development of spatial perspective on crime, but Herbert (1982) viewed factor
analysis only as an extension to the methodology
of the spatial ecologists. The factor analytic stage represented the resource of
spatial perspective on
crime after sever criticism of ecological theory, and therefore is an important
contribution to the study
of crime. Herbert (1982) cites the work of Lander and Schmid who used factor
analysis in the study of
crime in Baltimore and Seattle, respectively.
Both studies produced dimensions of crime, but in fact drew similar conclusions to
that of the ecological approach. High proportion of ‘non whites’ that rent their
accommodation, for example, provides a dimension of anomie to crime, according to
Lander. Broken families with a low economic
status, on the other hand, is also a dimension, according to Schmid (Herbert 1982).
The final, and current stage of spatial perspective on crime according to Lawman
(1986) is that of the
complementary fields of the geography of crime and environmental criminology. The
geography of
crime represented the first attempts by geographers to enter the crime research in
arena. This mainly
involved engaging in a “modern form research in aerial and ecological traditions”
which forms the
initial links with the geography of crime (Herbert 1982). The majority of this work
emerged in the
USA, although British research is also well documented.
2.4.
Tools in the spatial analysis of crime
This part examines four statistical mapping techniques that are useful for the
spatial analysis of crime:
block aggregation, voronoi diagrams, kernel smoothing and animation. Each technique
is discussed in
terms of its application to crime analysis, its utility for displaying crime
patterns, and its ability to
monitor changes in crime patterns over time. Used by analysts at the New York City
Police Department (NYPD), and illustrates the application of the methods in a GIS
environment and the strengths
and limitations of the methods when linked with broader crime reduction policies.
Crime Mapping and Analysis Application (CMAA) that incorporates both area and point
based methods in a user-friendly tool for mapping and analysis crime patterns. The
application allows users to
query crime data and perform four different types of mapping and spatial analysis.
Specifically, the
application:
1. Performs ‘block aggregation’ to generate chloropleth maps (maps with colour
varying according to the scale of the problem) of incidents aggregated to several
types of geographic units
(census blocks, police sectors, police precincts, etc.);
2. Creates smoothed density maps of crime incidence using kernel estimation;
11
3. Prepares voronoi diagrams (Theissen polygon) of individual crimes and performs a

coverage
analysis based on the areas of voronoi regions; and
4. Creates map animations to show changes in crime patterns over time.
Although the CMAA incorporates several complex methods, the goal was to create
tools for rigorous
display of distribution and change in crime patterns within an easy to use system
aimed at no specialists. The CMAA was designed to facilitate the query, mapping and
exploration of spatial and temporal
crime data(Alex and Kate 2001).
The first tool, block aggregation, involves aggregating crime incidents into
geographical areas and
generating a chloropleth map in which those areal units are shaded based on the
number of incidents
within them. This type of analysis enables the user to determine quickly which
areas have a high incidence of crime and allows them to ‘zero in’ on those areas
and perform further analysis. It is also useful for creating tables that show
counts of crime by area and change in crime incidence over time.
The block aggregation technique has several advantages. First, it is easy to
calculate. Other than acquisition of a digital census block map, there is no
additional cost to employing the method. In addition, the output is easy to
explain. Whether the audience is executive staff, beat cops, or community
groups, block aggregation maps are easily understood, and do not require technical
expertise to interpret. The technique is precise yet flexible. While many cluster
identification techniques incorporate
non-hotspot areas into their hotspots, the block aggregation technique precisely
identifies which
blocks have large amounts of crime and which do not.
Despite its ease of use and intuitive appeal, the block aggregation technique has
several limitations.
First, the technique does not handle small amounts of data (Diehr 1984). The data
for each block have
a large random variability from month to month, so that the map depicting those
values may be misleading and inaccurate. On the block level chloropleth maps,
increasing the cut-off point by one additional crime may drastically reduce the
number of blocks identified in the highest crime category and
vice versa.
The block aggregation technique was included in the CMAA because the advantages
outweighed the
shortcomings. Block aggregation is a useful tool for studying historic crime
patterns, and provides an
easy first cut at the data before additional analysis using more sophisticated
techniques. It also creates
a direct link to demographic and housing data from the census and thus provides a
valuable foundation for exploring the associations between crime and socio-economic
conditions (Alex and Kate
2001).
The second tool, Kernel density estimation, is a statistical method for determining
the density of
crime or other point events at different locations (Bailey and Gutrell 1995). The
method is used to
generate a continuous crime density surface from crime point data. The analyst
begins with a dot map
of crime events. Kernel smoothing results in a continuous weather map that shows
geographic variation in the density or intensity of crime. Peaks on the map
represent areas of high crime (crime hotspots) and valleys represent areas of low
crime. Increasingly, crime analysts are employing kernel
smoothing to visualize and analyse crime patterns (Williamson, McLafferty et al.
1999);(Brown and
Dalton 1998). A particular benefit of the method is that unlike in block
aggregation, the analysis is not
limited to some patterns than on a complex point map.
12
Geographical variation in crime density is clearly visible on the smooth density

map. In addition to its
ease of interpretation, kernel smoothing, unlike other traditional methods for
identifying crime hotspots, such as STAC (Spatial and Temporal Analysis of Crime),
is that the hotspots can be irregularly
shaped and need not follow regular geometric shapes like circles or ellipses (Block
1995).
The third tool, voronoi diagram, is technique of spatial analysis that has seen
wide application across
many disciplines. Voronoi diagrams or Theissen polygons, divide a mapped area, into
a number of
polygons. The voronoi diagram associated with a crime incident map may be quite
helpful to police
analysts and managers who are confronted with the problem of choosing the
geographical areas to be
assigned some special patrol resources. Specifically we present a method for
performing what we
have termed coverage analysis. Coverage analysis relates the percentage of the
total map area covered
by the sum of the corresponding voronoi polygons.
The relationship between the percentage of incidents and percentage of area is
based upon two factors. First, each voronoi polygon is associated with one and only
one crime incident, and second the
voronoi polygons are associated with clustered crime incidents. Given this, we can
create a voronoi
diagram that shows the trade-off between the number of crime incidents and the
total area that lies
within the voronoi polygons associated with those incidents.
The final tool incorporated in the CMAA is an animated map, characterized by
continuous or dynamic
change (Slocum 1999). The maps are displayed dynamically in sequence, forming a
constantly changing image or animation. The field of animation has advanced rapidly
in recent years, stimulated by
developments in computer hardware and graphics. These advances are fuelling changes
in mapmaking as cartographers gain access to one of the first effective tools for
representing continuous
change through space and time. Analysts are just beginning to explore the use of
animation in crime
mapping (Openshaw, Waugh et al. 1994).
The animation component of the CMAA is designed to provide flexibility in defining
duration and
rate of change for a particular animated sequence. The crime analyst first selects
the time interval for
maps in the sequence and then defines the degree of overlap among the maps. In this
way one can
control the smoothness or roughness of change in the animation and the time periods
to be displayed.
The animation works with crime density maps generated by kernel smoothing, because
they are visually appealing and easy to comprehend in the short time interval of
the animation (Alex and Kate
2001).
Animation is an effective way of displaying changes in crime through time and
space; however, it is
primarily a visual tool. Animated maps clearly show regular patterns, but if events
move or vary in
intensity unpredictably over time, the animation will be difficult to comprehend.
Viewers often have
trouble comprehending information on animated maps. The images move by quickly and
are difficult
to compare. Therefore we expect that the animation tool in the CMAA will primarily
be used for
visualization and display, for ‘gee-whiz’ presentations, and to give the viewer a
sense of general
space-time trends (Alex and Kate 2001).
Conclusion
The Crime Mapping and Analysis Application moves beyond the creation of
computerized pin maps
to assist decision makers with visualizing, analysing and uncovering patterns in
the large spatial data13
base that law enforcement organizations work with on a daily basis. The application
offers a series of
technically sophisticated in space, how those clusters change over time, and how a
given level of resources might be allocated to cover the largest amount of crime.
Although the CMAA was designed for using by decision makers with little or no
training in spatial
analysis, it is unclear how these non-specialists will respond to the system and
which of its components they will find of greatest value.
14
3. Data collection and preparation

3.1.
Data modelling in GIS
A data model is a set of constructs for describing and representing selected

aspects of the real world
in a computer. According to Longley (2001), when representing the real world in
computer, it is helpful to think in terms of four different levels of abstraction
(figure 3.1).
Human-oriented
Reality
Increasing
Abstraction
Conceptual model
Logical model
Physical model
Computer-oriented
Figure 3.1 Levels of GIS data modelling (Longley 2001)
First, reality is made up of real-world phenomena, and includes all aspects that
may not perceive by
individuals, or deemed relevant to particular application. Second, the conceptual
model is a human
oriented, often partially structured, model of selected objects and processes that
are thought relevant
to a particular problem domain. Third, the logical model is an implementation-
oriented representation
of reality that is often expressed in the form of diagrams and lists. Forth, the
physical model portrays
the actual application in a GIS, and often comprises tables stored as files or
databases.
There are two important components of geographic data:
The spatial data:

Geographic position specifics the location of a feature or phenomenon by using a
coordinate system, which are used for representing geographical places: cities,
rivers, lakes, etc.
The attribute data

Attribute data refer to the properties of spatial entities, which describe the
characteristics of the
geographic features: population, length, and area.
Spatial features in a GIS database are stored in either vector or raster form:
15
Vector data
GIS data structure adhering to a vector format store the position of map features
as pairs X, Y (and
sometimes Z) coordinates. A point is described by a single X-Y coordinate pair and
by its name or
label. A line is described by a set of coordinate pairs and by its name or label.
Therefore, a line is built
up of straight-line segments. An area, also called polygon, is described by a set
of coordinate pair and
by its name or label with the difference that the coordinate pairs at the beginning
and end are the
same.
A vector format represents the location and shape of features and boundaries
precisely. Only the accuracy and scale of the map compilation process, the
resolution of input devices and the skill of the
data-in putter limit precision.
Raster data
The raster or grid-based format generalizes map features as cell or pixels in a
grid matrix. The space
is defined by matrix of points or cells organized into rows and columns. If the
rows and columns are
numbered, using column number and row number can specify the position of each
element. There are
advantages to the raster format for storing and processing some types of data in
GIS.
Comparison of raster and vector data formats is depicted in Table 3.1. In this
project vector data are
used.
Table 3.1 Comparison of raster and vector data
Raster data
Vector data
Shape description
Not precise
Precise
Data volume
Large
Small
Processing speed
Fast
Slow
Complexity
Simple
Complex
Construction cost
Low
High
Data updating
Easy
Not easy
Data analysis
Easy
Difficult
In this project object-oriented relational and Geodatabase are used. More recently
United Modelling
Language (UML) has been adopted in more and more object-oriented data modelling
activities. Object oriented GIS data model, is the most significant advancement in
GIS technology in 10 years.
Unlike traditional GIS data models, which involve a myriad of linkages between data
tables and
16
graphic and attribute data stored in different locations, the object oriented GIS
data model can accommodate vector and raster data.
Advantages of object oriented:
The object oriented design more closely mirrors the real world, because it features
a series of
intelligent objects.
The object oriented GIS data model handles these intelligent objects with precise
rules and relationships embedded up front in the database design, because these
business rules are inside
the object oriented GIS data model itself, the number and complexity of custom
programming
routines that need to be written and run in the background are reduced.
An object oriented GIS data model can be planned and designed using computer-aided
software engineering tools and Unified Modelling Language (UML).
Object oriented GIS data model is seamless data, is not separated into tiles or
grids and any
user can edit any feature in the object oriented at any time. Since areas and
features are not
locked out during editing, multiple users can complete updates faster.
The object oriented GIS data model can exist inside a standard corporate relational
database,
and that is good news for IT departments in local governments that might have
struggled with
operating and maintaining proprietary in the past.
Geodatabase model supports an object-oriented vector, data model that is a kind of

physical data
model in ArcGIS 8.2. The defining purpose of this new data model is to make the
features in GIS
dataset smart by endowing them with natural behaviours, and to allow any sort of
relationship to be
defined among features. In this model, entities are represented as objects with
properties, behaviour,
and relationships. Support for a variety of different geographic object types is
built into the system.
These object types include simple objects, geographic features (objects with
location), network features (objects with geometric integration with other
features), annotation features, and other more
specialized feature types. The model allows the user to define relationships
between objects, besides
rules for maintaining the referential integrity between objects.
3.2.
Data collection
Data used in GIS comes from many sources, are of many types and are stored in
different ways. A
GIS provides tools and methods for the integration of data into a format so that
data can be compared
and analysed. A small part of Tehran was selected and relative attribute data and
spatial data were
collected.
The police organization in Tehran has the following hierarchy: one police command
at the top that
includes a number of peace officer, each peace officer includes several police
stations, each police
station contains one precinct (jurisdiction), which is divided into five blocks for
better administration
and response to the area (figure 3.1).
17
Command
Peace officer
Peace officer
Peace officer
Peace officer
Peace officer
Precinct (jurisdiction)
Block
Block
Block
Block
Block
Figure 3.2 Police organization in Tehran

The needed of information can be extracted based on the following procedure:
Required attribute data for this research was collected from central district
(peace officer) of Tehran,
for duration of one season from Farvardin (April) to Khordad (June) 2002, which
were based on the
offences statistic at the various blocks that was reported and recorded by police
service. The following are some of the crime incidents:
-
Burglary
Shoplifting
Auto theft
Fraud
Theft from vehicle
Robbery
Addiction
Pick pocket
Hold up
Motor theft
Abuse of confidence
An example:
Table 3.2 depicts type of crime and number of crime in different blocks on the
police station number
109.
18
Table 3.2 Census types of crimes in the Police station, number 109
1
2
3
4
5
6
7
8
9
10
3.3.
Type of crime
Burglary
Shoplifting
Auto theft
Fraud
Theft from vehicle
Robbery
Addiction
Pick pocket
Hold up
Abuse of confidence
Block1
8
1
6
25
27
3
17
20
1
30
Block2
3
3
5
18
16
0
15
18
0
32
Block3
0
7
5
20
10
1
14
14
1
55
Block4
4
6
2
22
18
0
23
18
0
50
Block5
2
1
6
15
7
1
12
10
0
37
Total
17
18
24
100
78
5
81
80
2
204
Spatial data contains 31 digital map sheets (DWG files) of study area at scale
1:2000 that
were collected from the municipality of Tehran. Paper maps about police stations,
precinct, and
block were collected from police stations.
Data preparation
This step includes the following tasks:
Joining digital map sheets (merged maps) was done when the study area spread across
more
than one map. Most spatial data available in IRAN use a rectangular lattice as
basic spatial
units. They are provided as separate map sheets. Therefore, it is often necessary
to join separate spatial data even if the existing data are used.
•
•
31 digital map sheets covering the area were joined in AutoCAD 2000.
Creation of different required layers using CAD, because the lack of different
layers only one
layer for all of features.
Spatial data acquisition of this research depicted in Table 3.3.
Table 3.3 Collected dataset and their format
1
2
3
4
5
6
7
Features
Boundary (Precinct)
Block
Population
Network road (main streets)
Bus line
Police station
Bank
19
Spatial type
Polygon
Polygon
Polygon (block)
Line
Line
Point
Point
Format
Shape file
Shape file
Shape file
Shape file
Shape file
Shape file
Shape file
8
9
10
11
12
13
14
15
16
Military (restriction area)

Embassy
Government offices
Car parking
Educational centre
Entertainment cultural
Gas station
Fire station
Public health centre
Point
Point
Point
Point
Point
Point
Point
Point
Point
Shape file
Shape file
Shape file
Shape file
Shape file
Shape file
Shape file
Shape file
Shape file
Conversion of data format in AutoCAD 2000 to the Arc View (ESRI) shape file for
using and
adding database.
•
•
Georefrencing; used WGS_1984_UTM_Zone-39N projected coordinate systems (metric).

Geocoding; if the location is given only by its address, the data cannot be
directly handled in
GIS. In such a case we have to convert the address into its XY coordinates such as
longitude
and latitude coordinates, this process is called geocoding.
For example:
Address of Police station number 109:
No.10, alley Shahid TAGHAVI, street JOMHURI-YE-ESLAMI, Tehran-IRAN.
Geocoding Process
Police station no109, 537986X, and 3948851Y
Conversion of shape file to coverage in Arc Info (ESRI) Arc Catalog in order to use
in Arc
map.
Exporting shape files using custom values to the coverage file, using different
grid level for
features, and WGS_1984_UTM_Zone-39N projected coordinate systems.
•
3.4.
Importing coverage files to Geodatabase using custom value.
Database Magement System
A database management system (DBMS), or simply a database system (DBS), consists

of:
•
•
A collection of interrelated and persistent data (usually referred to as the

database).
•
•
Retrieving information from the database.
A set of application programs used to access, update and manage that data (which
form the
data management system).
The goal of a DBMS is to provide an environment that is both convenient and
efficient for:
Storing information into the database.
Why database management systems are necessary?
It’s necessary for the following reasons:
20
1- Development of the system

New application programs must be written as the need arises.
New permanent files are created as required.
But over a long period of time files may be in different formats, and
Application programs may be in different languages.
2- There is problems with the straight file-processing approach
Data redundancy and inconsistency
Multiple users
Difficulty in accessing data
Data isolation
Security problems
Integrity problems
3.4.1. The Entity Relationship (ER) diagram
The entity-relationship models use diagrams to depict the entities and their
relationships. In this diagrams rectangles represent entity types and diamonds
represent relationships. Relationships are linked
to their constituent entity types by arcs, and the degree of the relationship is
indicated on the arc.
For example:
The entity types POLICE STATION, PRECINCT, and BLOCK might be represented
schematically as
shown in figure 3.3. Each police station has many blocks but only one precinct. A
block belongs to
only one police station.
The entity types for this situation (with primary key attributes underlined) are as
follows:
1) Entity type POLICE STATION with attribute POL-ID (a unique police station ID),
X, and Y.
2) Entity type PRECINCT with attribute PR-ID (a unique precinct ID), Non-RD, Area,
Perimeter, RD, and pol-ID.
3) Entity type BLOCK with attribute block-ID (a unique block ID), Non-RD, Area,
Perimeter,
RD, Non-rd, BU, AU, MO, TH, RO, AD, PI, and PR-ID.
The relationships among the entity types are as follows:
The 1:1 relationship HAS_TO between POLICE STATION and PRECINCT.
The 1:N relationship CONTAINS between PRECINCT and BLOCK.
21
Non-RD
Area
Perimeter
RD
PPOLICE
STATION
Has
PRECINCT
Pol_ID
Contain
BU
……….
SH
BLOCK
AU
MO
Perimeter
AD
TH
Figure 3.3 A sample ER diagrams.

3.4.2. Description of entities
-
Police station: national law and enforcement office, which is in charge of

boundary.
Boundary: The political unit under jurisdiction of police station.
Block: The smallest political unit under jurisdiction of police station.
3.4.3. Enterprise Rules

The enterprise rules based on which all constraints between the entity types are
constructed can be
listed as follows:
A police station must contain a precinct,
Each precinct includes several blocks, and
22
Many blocks are under jurisdiction of a police station.
3.4.4. Skeleton tables

Pol_ID*
Police station
Boundary
PR_ID*
Area
Perimeter
RD
Non-RD
NonBU
RD
# * Refers to primary key, F and refers to foreign key.
Block
B_ID*
Area
Perimeter
RD
SH
Pol_IDF
AU
FR
TH
RO
AD
PI
HO
3.4.5. Data dictionary

Data catalogue (dictionary) is a repository of information about the data in the
database (data about
data); it can perhaps be the best view as the yellow page of the database. It
contains information as the
following about data:
Table 3.4 Descriptions of entity types and their attributes
Abbreviation
Pol_ID
X
Y
PR_ID
RD
Non RD
B_ID
BU
SH
AU
FU
TH
RO
AD
PI
HO
Description
Police station identify (ID)
Longitude
Latitude
Precinct ID
Residential Population
Non Residential Population
Block ID
Burglary
Shoplifting
Auto theft
Fraud
Theft from vehicle
Robbery
Addiction
Pick pocket
Hold up
23
PR_I
DF
4. Multicriteria Decision Making

Introduction
Police have learned from experience that there are particular environments that
concentration of
crimes is larger-than-expected. Sometimes these hot spot areas are defined by
particular activities
(e.g., burglary), other times by specific concentrations of land uses (e.g., bus
station), and sometimes
by interactions between activities and land uses, such as shoplifting in commercial
areas or pickpocket in bus or bus stop. While there are some theoretical concerns
about what links disparate crime
incidents together into a cluster, nonetheless, the concept is very useful. Police
officers patrolling a
precinct can focus their attention on particular environments because they know
that crime incidents
will continually reappear in these places. Crime prevention units can target their
efforts knowing that
they will achieve a positive effect in reducing crime with limited resources.
Unfortunately, measuring a hot spot is also a complicated problem. There are
different statistical
techniques designed to identify ‘hot spots’. Many, but not all, of the techniques
are typically known
under the general statistical label of cluster analysis. These are statistical
techniques aimed at grouping cases together into relatively coherent clusters. All
of the techniques depend on optimising various
statistical criteria, but the techniques differ among themselves in their
methodology as well as in the
criteria used for identification. Because ‘hot spots are perceptual constructs, any
technique that is
used must approximate how someone would perceive an area. The techniques do this
through various
mathematical criteria.
Types of cluster analysis (Hot Spot) methods
Several typologies of cluster analysis have been developed as cluster routines
typically fall into several general categories:
Point locations: This is the most intuitive type of cluster involving the number of
incidents
occurring at different locations. Locations with the most number of incidents are
defined as ‘hot
spots’.
Partitioning techniques: frequently called the K-means technique, partitioning

technique
where all points are assigned to clusters and are displayed as ellipses.
Density techniques: Identify clusters by searching for dense concentrations of

incidents, one
type of density search algorithm using the Kernel Density method.
•
Risk -based techniques: Identify clusters in relation to an underlying base ‘at
risk ’ variable,
such as population, employment, or active targets.
Miscellaneous techniques: other methods that are less commonly used including
techniques
applied to zones, not incidents.
Hierarchical techniques: are like an inverted tree diagram in which two or more
incidents are
first grouped on the basis of some criteria.
Based on the above-mentioned reasons and the available data (general and only
vector data) hierarchical methods (multicriteria decision-making) were used for
detection of hot spots for crime prevention.
24
4.1.
Multi Criteria Evaluation to Support Decision Making
Colson and Bruyn (1989) as cited by Sharifi (2001) defines Multicriteria Decision
Making (MCDM)
as a world of concepts, approaches and methods to help decision makers to describe,
evaluate, sort,
rank and select or reject on the basis of evaluation (expressed by score, value,
preference intensities)
based on several criteria. To what extent an alternative course of action attains
the objectives can be
measured by a set of criteria. Evaluation aims at rationalizing the planning and
the Decision Making
Process (DMP) by structuring the relevant aspect of the problem.
Colson and Bruyn (1989) classify MCDM into the two methods, namely, the Multiobject
Decision
Making Methods (MODM) and the Multi Attribute Decision Making Methods (MADM). The
MADM is concerned with choice out of finite set of options. Attributes are
evaluated, the alternatives
are ranked, the best-chosen or bad rejected decision space to only good
alternatives. The distinction
between MADM and MODM is based on the classification of evaluation criteria into
attributes or
objectives. ‘Criteria’ is a generic term that includes both attributes and
objectives. An objective is a
statement about the desired state of the system under consideration. For any given
objective, one or
more different attributes are used to measure the performance in relation to that
objective(Sharifi
2001).
Multi objective methods (MODM) attempt to help DM to make a good (efficient)
decision that would
be satisfied. MODM problems involve designing the alternatives and searching for
the best decision
among an infinite or very large set of feasible alternatives. The MODM, which are
sometimes viewed
as a natural extension of mathematical programming, consider several objective
functions simultaneously.
Multiattribute methods (MADS) a good solution is selected based on various
assumptions concerning
DM. Usually preference of DM in the form of trade-off, pair wise comparisons,
included in the analysis. MADM problems involve obtaining the performances of
options on each attribute in order to analyse and rank a finite, moderate, small
and discrete set of alternatives. A common goal to both is an
attempt to help the DMs make good (efficient) decisions that makes them satisfied.
The components of a MADM are: Alternatives, criteria (attribute), criterion score,
criterion priorities
(rank order), and multicriteria evaluation method and sensitivity analysis of the
results, are shown in
Fig 4.1.
25
Set of alternatives
Set of criteria
Criterion
Score
Decision Table
A1 A2 . . . Ai
C1
C2
DM Preferences
Multi Criteria Elevation method
Selected alternatives
Sensitivity analysis
Final recommendation
Figure 4.1 General model of MADM, Jankowski, 1995
4.2.
Crime definition and type
It is important to understand the critical features of crime and the criminal

justice system. Crime is a
multifaceted concept that can be defined in both a legal and non-legal sense. From
a legal point of
view it refers to breaches of the criminal laws that govern particular geographic
areas (jurisdictions),
and are aimed at protecting the lives, property and rights of citizens within those
jurisdiction. A nonlegal point of view would define crime as acts that violate
socially accepted rules of human ethical or
moral behaviour.
Burglary: The unlawful entry of a structure to commit a felony or theft. The use of
force to
gain entry is not required to classify an offence as burglary, originally under
English common
law burglary was limited to entry in residences at night, but it has been expanded
to all criminal entries into any building.
Shoplifting: The unlawful entry of a shop to commit a theft, usually occur when the
shops
closed.
26
Auto theft: the crime of stealing automobile.

Motor theft: the crime of stealing motorcycle.
Theft from vehicle: Defined as the theft or attempted theft of a motor vehicle,
this offence
category includes the stealing of automobiles, trucks, buses, motorcycles, motor
scooters,
snowmobiles, etc.
Robbery: Robbery is the taking or attempting to take anything of value from the
care, custody,
or control of a person or persons by force or threat of force or violence and/or by
putting the
victim in fear. Uses force on any person or puts or seek to put any person in fear
of being then
and there subjected to force.
Addiction: the need to have something regularly because you are addicted to it:
drug addiction.
Abuse of confidence: the crime of stealing on the cheap things.
Pickpocket: someone who steals things from people's pockets, especially in a crowd
places.
Fraud: Fraudulent conversion and obtaining money or property by false pretenses.
Crime classification
There are several types of crime classification. In this research the following
classification is used:
Crime I: security
A group of crime focus on security, and important crime as viewpoint of security
against of person.
Crimes in this division are classified into the following groups: Hold up,
Burglary, shoplifting, robbery, and motor theft.
Crime II: Economic
A group of crime focus on economic, and important crime as viewpoint of economic
against of money
or property of person. Crimes in this division are classified into the following
groups: auto theft, shoplifting, burglary, and Fraud.
Crime III: social
A group of crime focus on social, and important crime as viewpoint of social.
Crimes in this division
are classified into the following groups: Addiction, robbery, pick pocket, and
Theft from vehicle.
4.3.
Multicriteria evaluation applied for crime assessments
Multicriteria evaluation concept has been used to assess the crime rate in various
blocks of Tehran
city. To achieve that the general model of MCDM is adapted for crime assessment as
presented in
figure 4.2.
27
Administration block
Type of crime
Crime event
Frequency table
A1
A2 . . . Ai
C1
C2
Comparative value
of crime
Aggregation of crime indicator
Classification
Figure 4.2 General model of MCE of crime
Components of this method

A set of alternatives and criteria for a decision problem is established. Then
together with the criterion
score are placed in a frequency table (effected table, decision table), the
alternatives forming the columns and the criteria forming the rows. Alternatives’
score on each criterion are entered into the table.
For this study, the method for ranking is weighted summation. In this method as a
first step all effect
scores are standardized. An appraisal score is then calculated for each alternative
first by multiplying
these standardized effect scores by their appropriate weight, followed by summing
the weighted
scores of all effects. The final ranking of the alternatives is assessed based on
the resulting appraisal
scores for each alternative (Sharifi 2001). Scores from the various effects can
only be compared if the
28
measurement units are the same. Through the standardization procedure the
measurement units are
made uniform, and the scores lose their dimension along with their measurement unit
(Voogd 1983).
Based on Maximum method of standardization these criteria have been standardized
(because of different unit). The maximum method is the best choice whenever we want
to criterion is measured on a
ratio scale, and standardized values are proportional to the original value
(advantages), but it will be a
disadvantage is that small differences between the alternatives do not become
totally visible. In this
method, the scores are normalized with a linear function between 0 and the highest
absolute score.
In this study AHP method has been applied, the following brief description of
method will be introduced.
Analytic Hierarchy Process (AHP) introduced by Saaty to overcome the difficulties
of the weights
and scores method and the trade off method (Appendix 2). It focuses on the
achievement of objectives, and assumes a rational decision-making process with
means: rational decision is one that best
achieves the multitude of objectives of the decision maker. The key will be to
focus on objectives,
rather than alternatives, criteria or attributes. It contains three main components
as follows:
Analytic
Analytic is a form of the world analysis, which means the separation of any
material or abstract entity
into its constituent elements. Analysis is the opposite of synthesis; AHP should
really be called the
Synthesis Hierarchy Process because at its core, AHP helps us measure and
synthesise the multitude
of factors involved in complex decisions.
Hierarchy
Herbert Simon, father of field of artificial Intelligence and Nobel laureate,
writes: “Large organisations are almost universally hierarchical in structure. That
is to say, they are divided into units, which
are subdivision into smaller units, which are, in turn, subdivision and so on.
Hierarchical subdivision
is not a characteristic that is peculiar to human organizations. It is common to
virtually all-complex
systems of which we have knowledge.
Process
A process is a series of actions, changes, or functions that bring about an end or
result. The AHP is
not magic formula or model that finds the right answer. Rather it is a process that
helps decision makers to find the best answer.
AHP is based on criteria that are measured on a ratio or ordinal scale. The
decision maker has to make
a comparison for every pair of criteria: first qualitative which is the transformed
into quantitative
scale from 1 to 9 as follows:
Equally preferred = EQUAL
1
Weak preference = MODERATE
3
Strong preference = STRONG
5
Demonstrable preference = VERY STRONG 7
Absolute preference = EXTREME
9
29
Frequency table
In this research the effects table contains most of the information necessary to
evaluate the problem
and is the basis for multicriteria methods. An effects table contains the
alternatives, the criteria and
the effects of the alternatives for the criteria. The names of the alternatives are
listed in the first row
include of block numbers: (1091,1092, …) for more details refer to chapter 3.2 in
page 17, the names
of the criteria in the first column include the type of crime (Burglary, …….) above
some of crimes are
defined, and the other part of table shows event numbers of crime (Per block and
per type of crime).
Showed different type of crime in fifteen blocks in table 4.1, 4.2, and 4.3 quality
appraisals depend on
the number of the criteria, the used priorities and the arithmetic technique.
Table 4.1 Frequency tables of crimes related to security
Police station No 109 109 109 109 109 113 113 113 113 113 116 116 116 116 116
Block Numbers
HOLDUP
2
1
3
0
4
1
5
0
1
0
2
0
5
0
2
0
3
0
4
0
5
0
BURGLARY
12
10
SHOPLIFTIN
1
3
12
ROBBERY
20
28
11
29
15
13
17
24
MOTOR_THEF
Table 4.2 Frequency tables of crime related to economic

Block Numbers
5
AUTO_THEFT
SHOPLIFTIN
12
1
3
BURGLARY
12
10
FRAUD
25
18
20
22
15
25
22
20
23
35
5
10
Table 4.3 Frequency tables of crimes related to social

Block Numbers
ADDICTION
17
15
14
23
12
17
13
19
28
29
30
60
20
80 120
PICK_POCKE
20
18
14
18
10
15
20
THEFT_FROM
27
16
10
18
7
11
15
ABUSE_OF_CO
30
50
50
32
37
39
50
45
62
88
18
14
23
20
12
30
Crime density assessment

In this part we used two methods for crime density assessment:
Crime assessment based on crime per block (an absolute number of incidents reported
in each
block).
Crime assessment based on density of crime (the block count expressed as a rate
per, 1000
resident population).
4.3.1. Crime assessment based on crime per block

Normalization (Standardization):
This falls under the MCE process. Because of the different nature of the criteria,
the criterion scores
are usually incompatible (unrelated) due to differences in the measuring units and
scales. Normalization transforms the score into one common measurement scale. Voogd
(1983) says it’s a mechanical
way of rescaling value. He identifies three types of normalization:
•
•
•
Transform to 0-1 with additivity constraint (N=raw (i)/sum raw).

Transform to 0-1 with ratio scale properties, referred to as Maximum (N=raw (i)
/Max raw).
Transform to 0-1 with interval scale properties, refer to as Interval (N=(raw (i)-
Min raw
(i))/(Max raw-Min raw)(Voogd 1983).
In this research used Ratio scale methods for normalized (Transform to 0-1 with
ratio scale properties,
referred to as Maximum (N=raw (i) /Max raw), or each raw score in effected table
(4.1, 4.2, and 4.3)
is divided by the highest raw score in the effects table of the criterion depicted
in table (4.4, 4.5, 4.6)
is very useful in standardizing an evaluation matrix that will be analysed by a
weighted summation
technique or any other technique which utilizes the magnitude of the individual
scores. This range is
very important as a reference for weighting the effects.
Table 4.4 Normalized table (ratio scale) presenting the crime related to security
Block Numbers
HOLDUP
2
1
3
0
4
1
5
0
1
0
2
0
3
0
4
0
5
0
BURGLARY
0.67 0.25
SHOPLIFTIN
0.08 0.25 0.58 0.5 0.08 0.25 0.58 0.08 0.33
ROBBERY
0.75
0.33 0.17 0.42 0.17
0.25
1
0
3
0
4
0
5
0
0.25 0.08 0.83 0.33 0.25 0.17 0.42

1
0.25 0.25 0.5 0.75 0.5 0.25
MOTOR_THEF 0.28 0.69 0.97 0.38
0.17 0.08 0.25 0.08
0.75 0.5
0.25
0.52 0.45 0.31 0.59 0.83 0.17 0.14 0.1 0.17 0.1
Table 4.5 Normalized table (ratio scale) presenting the crime related to economic
Block Numbers
AUTO_THEFT
2
1
0.83 0.83 0.33
1
1
0.17
3
0
4
0.08 0.25 0.58 0.5 0.08 0.25 0.58 0.08 0.33
BURGLARY
0.67 0.25
FRAUD
0.71 0.51 0.57 0.63 0.43 0.71 0.63 0.57 0.66
0.33 0.17 0.42 0.17
31
0.17 0.17 0.33
SHOPLIFTIN
4
0
0.17
0.17
0.17 0.08 0.25 0.08
0.25 0.08 0.83 0.33 0.25 0.17 0.42

1
0.14 0.23 0.29 0.2 0.17

Table 4.6 Normalized table (ratio scale) presenting the crime related to social
Block Numbers
ADDICTION
0.14 0.13 0.12 0.19 0.1 0.14 0.11 0.16 0.23 0.24 0.25 0.5 0.17 0.67
5
1
PICK_POCKE
0.9
THEFT_FROM
0.59 0.37 0.67 0.26 0.41 0.22 0.3 0.33 0.56 0.15 0.19 0.07 0.19 0.11
0.7
0.9
0.5 0.75 0.45

1
0.05 0.2
ABUSE_OF_CO 0.34 0.57 0.57 0.36 0.42 0.44 0.57 0.51 0.7
0.15 0.05 0.1
0.1
0.2 0.16 0.26 0.23 0.14
BUILDING CRIME HIRARCHY

Crime Assessment
Step 3
…………………………………………………………………………………………………………...
Step 2
Security
Economic
Social
…………...
Step 1
Abuse of confidence
Theft from vehicle
Pick-pocket
Addiction
Fraud
Burglary
Shoplifting
Auto theft
Motor theft
Robbery
Shoplifting
Burglary
Hold up
Relative Importance of various crimes
The aim of this part is to find the relative importance of various types of crimes,
as well as different
category of crimes:
Security
Economic
32
Social
There are several methods to drive the relative importance of crimes among which
Pair wise comparison has been selected for this study.
-
Weighted derivation
There are several methods for derivation weights, but in this research pair wise
comparison is used
which, are basic to the AHP methodology. When comparing a pair of factors (may be
objectives, subobjectives, scenarios, players, or alternatives), a ratio of
relative importance, preference of the factors
can be established. This ratio need not be based on some standard scale such as
feet or meters but
merely represents the relationship of the two factors being compared. This method
creates a matrix
containing the pair wise comparison judgments for the criteria (table 4.7, 4.8, and
4.9), from which a
priority vector is derived of relative weights for these elements (This procedure
can be used to assess
effect values).
Weights or criteria priorities allow the decision maker to specify the perceived
importance of individual factors relative to the others included in the evaluation.
Assigning weights to criteria scores is a
crucial step into the process.
Table 4.7 Pair wise comparison table of crime related to security (by Expert)
Hold up
Burglary
Shoplifting
Robbery
Motor theft
1
0.5
0.33
0.33
0.25
2
1
0.5
0.5
0.33
3
2
1
1
0.5
3
2
1
1
0.5
4
3
2
2
1
Hold up
Burglary
Shoplifting
Robbery
Motor theft
Table 4.8 Pair wise comparison table of crime related to economic (by Expert)
Auto theft
Shoplifting
Burglary
Fraud
Auto theft
Shoplifting
Burglary
Fraud
1
0.5
0.5
0.33
2
1
1
0.5
2
1
1
0.5
3
2
2
1
Table 4.9 Pair wise comparison table of crime related to social (by Expert)
Addiction
Pickpocket
Theft from vehicle
Abuse of confidence
Addiction
Pickpocket
1
0.5
0.33
0.25
2
1
0.5
0.33
Theft from vehicle
3
2
1
0.5
33
Abuse of confidence
4
3
2
1
Next step normalized of pair wise comparison tables; In this research used
Additivity constraint
methods (Transform to 0-1 with additivity constraint properties, referred to
(N=column (i) /sum
column (i)), or each column score in pair wise comparison table (4.7, 4.8, 4.9) is
divided by the
sum of column scores in pair wise comparison table. This kind of transformation is
especially in
normalizing various sets of different criterion weights are shown in table 4.10,
4.11, and 4.12.
Table 4.10 Normalized of pair wise comparison of crime related to security
Hold up
Burglary
Shoplifting
Robbery
Motor theft
Weight
0.5
0.2
0.1
0.1
0.1
0.4
0.3
0.1
0.1
0.1
0.4
0.3
0.1
0.1
0.1
0.33
0.25
0.17
0.17
0.08
0.4
0.24
0.14
0.14
0.08
0.42
0.21
0.14
0.14
0.1
Hold up
Burglary
Shoplifting
Robbery
Motor theft
Table 4.11 Normalized of pair wise comparison of crime related to economic
Auto theft
Shoplifting
Burglary
Fraud
Auto theft
Shoplifting
Burglary
Fraud
Weight
0.4
0.2
0.2
0.1
0.4
0.2
0.2
0.1
0.4
0.2
0.2
0.1
0.4
0.3
0.3
0.1
0.42
0.23
0.23
0.12
Table 4.12 Normalized of pair wise comparison of crime related to social
Addiction
Pickpocket
Theft from vehicle
Abuse of confidence
Addiction
Pickpocket
Theft from vehicle
Abuse of confidence
Weight
0.5
0.3
0.2
0.1
0.571
0.286
0.143
0.086
0.5
0.333
0.167
0.077
0.4
0.3
0.2
0.1
0.46
0.28
0.16
0.1
Aggregation method
Weight summation is a simple and frequently used evaluation method. This procedure
generates a
ranking of the effectives based on the weighted sum of the alternative scores. As a
first step all effect
score are normalized, and then sum of each raw and divided by type of crime (number
of columns)
showed in table (4.5, 4.6, and 4.7). Alternatives rank orders were obtained by
totalling the products of
the normalized value (see table 4.4, 4.4, and 4.6) with the weights of each
criterion (see table 4.10,
4.11, and 4.12) use this formula(Sharifi and Marijnen 2002):
OVERALL ASSESMENT (i)=
Wi* Ci .
i =1
i: indices of each criterion.

W: weight of each criterion.
C: normalized score of each criterion.
34
Overall assessment (i) is equal value (degree) of crime in each block (zero is
equal no crime event and
one is equal (100%crime) or critical area depicted in table 4.13.
Table 4.13 Ranking the alternatives
Security
Block numbers
1161
1091
1093
1133
1135
1132
1134
1162
1131
1094
1095
1165
1163
1092
1164
Economic
Block numbers
Rank
1161
0.82
1135
0.79
1091
0.68
1132
0.49
1094
0.48
1093
0.45
1133
0.35
1134
0.35
1131
0.34
1092
0.28
1162
0.22
1164
0.22
1165
0.17
1163
0.15
1095
0.14
Rank
0.75
0.70
0.59
0.38
0.26
0.23
0.22
0.22
0.21
0.18
0.17
0.16
0.15
0.15
0.09
Social
Block numbers
1091
1165
1094
1092
1133
1131
1093
1135
1164
1162
1095
1132
1134
1161
1163
Rank
0.54
0.52
0.48
0.46
0.45
0.38
0.37
0.36
0.36
0.32
0.27
0.27
0.25
0.16
0.13
Overall assessment of crime rate

This part creates the final ranks instead of crime classification and preference of
crime classification
based on which one importance than others, and changes this preference. .
Table 4.14 Effect table
Block Numbers
1
Security
0.70 0.15 0.59 0.18 0.17 0.21 0.23 0.38 0.22 0.26 0.75 0.22 0.15 0.09 0.16
Economic
0.68 0.28 0.45 0.48 0.14 0.34 0.49 0.35 0.35 0.79 0.21 0.21 0.15 0.22 0.17
Social
0.54 0.46 0.37 0.48 0.27 0.38 0.27 0.45 0.25 0.36 0.16 0.32 0.13 0.36 0.52
This research used three-perspective form of crimes for ranking:
Security perspectives: this vision security is important than others, and shows
preference and
weight in table 4.15. When comparing a pair of crime classification, a ratio of
relative importance, preference of the factors can be established. This ratio need
not be based on some standard scale such as feet or meters but merely represents
the relationship of the two factors being
compared.
35
Table 4.15 Preference of security perspective

Security
Economic
Social
Weight
0.54
Economic
1/2
0.30
Social
1/3
1/2
0.16
Security
Overall assessment for security perspectives and maps of ranking showed in figure
4.3.
Block
. No
1091
1161
1093
1135
1133
1094
1132
1131
1162
1134
1165
1092
1164
1163
1095
Rank
0.70
0.53
0.50
0.39
0.38
0.31
0.29
0.26
0.24
0.23
0.23
0.21
0.18
0.15
0.13
Figure 4.3 Overall assessment for security
Economic perspectives: economic perspectives: this vision economic is important

than others,
and preference and weight are shown in table 4.16.
Table 4.16 Preference of economic perspective
Economic
Security
0.43
Security
1/2
0.35
Social
1/2
1/2
0.22
Economic
36
Social
Weight
Overall assessment for economic perspectives and maps of ranking showed in figure
4.4.
Block
. No
1091
1093
1161
1135
1133
1094
1132
1131
1165
1092
1134
1162
1164
1095
1163
Rank
0.68
0.48
0.46
0.42
0.38
0.34
0.31
0.28
0.26
0.24
0.24
0.24
0.20
0.14
0.14
Figure 4.4 Overall assessment for economic
Social perspectives: social perspectives: IN this vision, social is important than

others, and
preference and weight are shown in table 4.17.
Table 4.17 Preference of social perspective
Social
Security
Economic
Weight
Social
1
1
0.43
Security
0.43
1/3
1/2
0.14
Economic
37
Overall assessment for security perspectives and maps of ranking are shown in
figure 4.5.
Block
. No
Rank
1091
0.67
1093
1135
1161
1133
1094
1131
1132
1134
1092
1162
1163
1164
1165
1095
0.50
0.50
0.44
0.38
0.35
0.29
0.35
0.28
0.25
0.23
0.15
0.19
0.21
0.17
Figure 4.5 Overall assessments for social

The MCE approach was implemented on the types of crimes. The results are summarized
in graph 4.1.
Block
. No
1091
1092
Block
1
2
1093
3
1094
1095
1131
1132
1133
1134
1135
1161
1162
1163
1164
1165
4
5
6
7
8
9
10
11
12
13
14
15
Graph 4-1 Classify of crime value depend on three category
38
Based on the above-mentioned reasons and comparison of three figures, the following
shows the results:
The reason for sudden crime increasing in block 1161 from security viewpoint can be
follows:
High population.
Immigration.
Accommodation of offences.
Accessibility to suburbs.
The reason for sudden crime increasing in block 1135 from social viewpoint can be
follows:
Frequent trade centres.
High floating population.
Time limit of traffic.
The reason for sudden crime increasing in block 1132 and 1134 from social viewpoint
can be follows:
Closeness of trade centre.
Lack of entertainment and cultural centres.
4.3.2. Crime assessment based on density of crime
In the most general sense, measurement is the foundation of scientific analysis,
and it lies behind any
quantitative analytical statement. To calculate this density we must know how many
crime incidents
have occurred, and, if we are calculating a population-based density, how many
persons there are per
block. Density value is calculated for each block (Harries 2001).
Density =
Number of incidents per bloc

Population per block
These density values can be used to create a chloropleth map, which uses colour to
represent different
values among blocks within the study area. An application of density analysis is
shown in figures 4.6,
4.7 and 4.8 for three vision of crime.
Table 4.18 Population per block

Block Numbers
Population
80 75
1 indicated 1000 people.
4
60
5
40
1
45
130 150 230
5
15
1
20
110
3
50
4
80
5
50
60
For calculation of density use frequency tables 4.1, 4.2, and 4.3 divided by
population per block
showed in tables 4.19, 4.20 and 4.21.
39
Table 4.19 Density tables presenting the crime related to security

Block Numbers
HOLDUP
0.01
BURGLARY
0.1 0.04
SHOPLIFTIN
0.01 0.04 0.12 0.15 0.02 0.02 0.05
ROBBERY
0.04
MOTOR_THEF
0.1 0.27 0.47 0.28 0.64 0.12 0.12 0.23 1.13 1.2 0.05 0.08 0.04 0.1 0.05
0.02
5
0
1
0
2
0
3
0
4
0
5
0
1
0
0.01
3
0
4
0
5
0
0.1 0.04 0.04 0.01 0.05 0.2 0.05 0.09 0.08 0.04 0.04 0.08
0.02
0.27 0.6
0.04 0.01 0.06 0.02
0.02 0.01 0.01 0.01 0.13 0.05 0.04 0.06 0.03
0.02
Table 4.20 Density tables presenting the crime related to economic

Block Numbers
AUTO_THEFT
0.08 0.07 0.08 0.05 0.13 0.01
SHOPLIFTIN
0.01 0.04 0.12 0.15 0.02 0.02 0.05

BURGLARY
0.1 0.04
FRAUD
0.1 0.24 0.33 0.55 0.33 0.19 0.15 0.09 1.53 1.75 0.05 0.16 0.13 0.14 0.1
3
0
4
0
0.07 0.1
0.02
0.02
0.27 0.6
0.04 0.01 0.06 0.02
0.1 0.04 0.04 0.01 0.05 0.2 0.05 0.09 0.08 0.04 0.04 0.08
Table 4.21 Density table presenting the crime related to social
Block Numbers
ADDICTION
0.21 0.2 0.23 0.58 0.27 0.13 0.09 0.08 1.87 1.45 0.27 1.2 0.25 1.6
PICK_POCKE
0.25 0.24 0.23 0.45 0.22 0.12 0.06 0.09 0.07 0.2
THEFT_FROM
0.34 0.21 0.17 0.45 0.16 0.08 0.04 0.03 0.6 0.75 0.04 0.1 0.03 0.1 0.05
5
2
0.06 0.01 0.04 0.03
ABUSE_OF_CO 0.38 0.67 0.83 0.8 0.82 0.3 0.33 0.2 4.13 4.4 0.16 0.28 0.29 0.4 0.2
Then aggregated method (weight summation is a simple and frequently used evaluation
method. This
procedure generates a ranking of the effectives based on the weighted sum of the
alternative scores.
Alternatives rank orders were obtained by totalling the products of the density
value (see table 4.19,
4.20, and 4.21) with the weights of each criterion (see table 4.10, 4.11, and 4.12)
used in this formula(Sharifi and Marijnen 2002):
OVERALL ASSESMENT (i)=
Wi* Ci .
i =1
Also, ranks are shown in table 4.22.
40
Table 4.22 Ranking the alternatives

Security
Block numbers
1161
1091
1093
1133
1135
1132
1134
1162
1131
1094
1095
1165
1163
1092
1164
Economic
Block numbers
Rank
1161
0.03
1135
0.40
1091
0.07
1132
0.03
1094
0.14
1093
0.10
1133
0.03
1134
0.32
1131
0.04
1092
0.08
1162
0.06
1164
0.04
1165
0.04
1163
0.03
1095
0.11
Rank
0.03
0.04
0.06
0.03
0.20
0.02
0.19
0.04
0.02
0.07
0.07
0.03
0.02
0.04
0.03
Social
Block numbers
1091
1165
1094
1092
1133
1131
1093
1135
1164
1162
1095
1132
1134
1161
1163
Rank
0.26
0.96
0.54
0.26
0.09
0.14
0.28
1.28
0.80
0.61
0.29
0.10
1.39
0.15
0.15
Overall assessment of crime density

This part creates the final ranks instead of crime classification and preference of
crime classification
based on which one important than others, and changes this preference.
Table 4.23 Effect table
Block Numbers
1
2
Security
0.04 0.04 0.06 0.07 0.07 0.02 0.02 0.03 0.19 0.2 0.03 0.04 0.02 0.03 0.03
Economic
0.07 0.08 0.1 0.14 0.11 0.04 0.03 0.03 0.32 0.4 0.03 0.06 0.03 0.04 0.04
Social
0.26 0.26 0.28 0.54 0.29 0.14 0.1 0.09 1.39 1.28 0.15 0.61 0.15 0.80 0.96
This research used three-perspective form of crimes for ranking:

Security perspectives: In this vision security is important than others, and shows
preference and
weight in table 4.15. When comparing a pair of crime classification, a ratio of
relative importance,
preference of the factors can be established. Overall assessment for security
perspectives and maps of
ranking, density maps offer the map user a broader look at where crimes occur
without his having to
interpret a large number of individual locations shows in figure 4.
41
Block
. No
Rank
1135
0.43
1134
1165
0.42
0.18
1094
0.17
1164
1162
1095
1093
1092
1091
1161
1131
1132
1133
1163
0.15
0.14
0.12
0.11
0.09
0.08
0.05
0.04
0.04
0.04
0.04
Figure 4.6 Overall assessment of crime density for security
Economic perspectives: economic perspectives: this vision economic is important

than others,
and preference and weight are shown in table 4.16. Overall assessment for security
perspectives
and maps of ranking showed in figure 4.5.
Block. No
Rank
1135
0.51
1134
1165
1094
0.50
0.24
0.20
1164
1162
1095
1093
1091
1092
1131
1161
1163
1132
1133
0.20
0.17
0.13
0.12
0.10
0.10
0.05
0.05
0.05
0.04
0.04
Figure 4.7 Overall assessment of crime density for economic
Social perspectives: social perspectives: this vision social is important than

others, and preference and weight are shown in table 4.17. Overall assessment for
security perspectives and
maps of ranking are shown in figure 4.5.
42
Block. No
Rank
1135
0.44
1134
1094
1165
1164
1162
1095
1093
1092
1091
1131
1132
1133
1161
1163
0.41
0.17
0.16
0.14
0.13
0.12
0.11
0.09
0.08
0.05
0.04
0.04
0.04
0.04
Figure 4.8 Overall assessment of crime density for social
The MCE approach was implemented on the types of crimes. The results are summarized
in graph 4.2.
Block
. No
1091
1092
1093
1094
1095
1131
1132
1133
1134
1135
1161
1162
1163
1164
1165
Block
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
Graph 4-2 Classify of crime value depend on three category
43
Based on the above mentioned reasons and compared three figures, the following
shows the results:
The reason for sudden crime increasing in block 1134 and 1135 from security,
economic and social
viewpoint can be follows:
Frequent trade centres.
High floating population.
Time limit of traffic.
-
4.4.
Analysis of the result
The most suitable locations for the police stations are selected based on closeness
to the crime area.
After interviewing a few police officers the value and effects of each crime was
examined, depending
on the behaviour of criminals, some might prefer quiet and some might prefer busy
areas for their activities, therefore, if we transfer the different criminals
behaviour on a chart, problem solving could be
impossible due to complications.
In crime assessment, based on crime per block, mostly there is direct relationship
between crime values and distance from police stations. Therefore, by increasing
the number of new police stations, it is
possible to reduce crime rate in the area located between the stations. But, in
crime assessment based
on density of crime, peaks occurred near the central business district and in the
high population area.
Therefore, by increasing the number of police officers, it is possible to reduce
crime rate in these areas.
44
5. Analysis
The data used in this research is from a part of Tehran city police department.
Purpose of this research
is crime prevention by using different methods, focused on hold up, burglary,
shoplifting, auto theft,
fraud, robbery, pick pocket, theft from vehicle, abuse of confidence and addiction
data, since these
types of crimes are the major types of interest to the command structure of the
police department.
Crime mapping can help law enforcement and protects citizens more effectively in
the areas they
serve. Simple maps display the locations where crimes or concentrations of crimes
have occurred, and
they can be used to help direct patrols to the places where they are most needed.
Policy makers in police departments might use more complex maps to observe trends
in criminal activities, and maps may
have invaluable roles in solving criminal cases.
Result of multicriteria evaluation (crime assessment based on crime per block) for
determined hot
spots is shown as follow:
Table 5.1 Crime value of blocks
Block.
No
Preference
Crime
value
1091
1093
1161
1133
1135
1094
1131
1132
1162
1092
1134
1164
1165
1163
1095
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
7
5
5
4
4
3
3
3
2.5
2
2.5
2
2
1.5
2
Classification of crime value:

1. Hot spots with crime value >= 4.
2. Important spots with crime value=3.
3. Negligible spots with crime value<3.
For crime prevention the three following approaches were used:
Concentration of police force in the hot spot areas.

45
•
•
Finding suitable location for establishing new police stations.

Relocation of current boundary of police stations, after established new police
station..
Multicriteria evaluation was used to determine hot spots or areas with high
concentration of crime.
Highlighting such areas is the simplest way to help police to direct patrols to
where they are most
needed. Presence of police has positive effect on the reduction of crime (recent
experiments in USA
have good results as showed in BBC).
The following criteria will be used to find the suitable location for establishment
of police stations.
There are two methods to find suitable locations; one method is to query your data
to identify locations that meet your criteria; the other method is to produce a
suitability map through the combination
of datasets to find out the suitability of each location in the area.
Criteria for establishing new police stations:
•
•
•
Closeness to places with hotspot values

Appropriate distance from existing police station
Closeness to the main streets.
Suitability model
The aim of the model is to find optimum locations. A suitability model might
identify suitable locations for a new police station.
Therefore the suitable locations can be founded for a new police station as follow:
Suitable site for a
new police station
Closeness of high
value of crime
Distance from existing

police station
Closeness to the
main streets
Cross three layers

and Create new layer
Figure 5.1 Method for establish new police station
46
Producing a suitability map

Figure 5.2(A) is a chloropleth map. It shows Crime rate based on block, and Figure
5.2(B) shows suitable distance around each police station, to find out areas beyond
50-meter buffer along main streets,
suitable distance from existing police station, and within hot spot areas.
(A)
(B)
Figure 5.2 Show hotspots area in (A), and buffers of police stations (B)
The map shown in figure 5.3(A) is the result of map calculation of two layers
(figures 5.2(A), 5.2(B)),
and figure 5.3(B) shows result of map calculation and overlay with buffer of main-
Street.
(A)
(B)
0
Figure 5.3 Map calculation of hot spot and buffers of police station in (A), and
overlay of map calculation with buffer of main streets in (B)
The final map shows that blocks no.1091, 1133, and 1161 have the highest potential
for establishing
new police station.
47
We should now assess these locations to see which might be the suitable location.
This should be
done in the field, as well as by examining the data we have on each potential area.
Figure 5.4 Suitable areas for new police station

From the above view, it is clear that the outer area represents the suitable areas
for new police station.
After identifying the suitable location for new police station, a method is
required to relocate the police station boundaries, which in this study theissen
polygon was used as shown in figure 5.5(A),
which shows the relocation of boundary for establishing one new police station and
in figure 5.5(B),
which shows the relocation of boundary for establishing two new police station
(used assign proximity).
(A)
(B)
Figure 5.5 Relocation of boundary
48
A technique for spatial analysis is Voronoi diagrams, which has wide application in
many disciplines.
Voronoi diagrams or theissen polygons divide a mapped area into a number of
polygons. Each polygon is constructed around a generating point. Almost all of the
boundaries might be changed.
The main reason could be that, it is impossible for police department to relocate
all of the boundaries
of police station, therefore buffering method is used around suitable area as shown
in figure 5.6, the
result of buffering shows only relocation of boundaries in neighbouring blocks.
.
Figure 5.6 Suggestion of new boundary for establishing new police stations
49
6. Conclusion and Recommendation

6.1.
concloution
Crime mapping and spatial analysis are important tools for mapping, analysis, and
visualisation of
crime data. Adapting spatial clustering (block) analysis to support crime analysis
and decisionmaking. Law enforcement has been using these methods to examine the
associations between crime
and environment features, to allocate resources for crime prevention in areas where
they are most
needed.
This technique (crime includes of blocks) has several advantages:
•
•
•
It is easy to calculate and the output is easy to explain.

The block technique is precise and also flexible (precisely identifies which blocks
have large
amounts of crime and which one do not).
Analysis at block level can still be very useful, by ranking the blocks based on
crime value.
Limitation
•
•
This technique does not handle small amount of data well.

The size of the blocks may skew the data. Some blocks incorporate parks, while
others refer
only to small underpass areas. Large blocks, not surprisingly, often have a lot of
crime, but may
not be especially hotspots comparing to the areas with same size, that are divided
into two or
three blocks.
This technique is used because the advantages outweighed the shortcomings, useful
tool for studying
historic crime patterns, and provides a valuable foundation for exploring the
associations between
crime and socio-economic conditions. Regional crime analysis is powerful tool, and
also easy-to-use
in geographic information system because it contains mapping, analysis, and
reporting tools for police
officers, crime analysts, and police managers.
Good crime analysis should precede good crime prevention policy and planning. I
believe that crime
rate can be reduced through good management and it is as essential to 21st century
policy, as it follows
new information technology. If crime is not simply a function of individual
pathology or inevitable
cycles, then it can be affected by how and where resources are deployed, and by
what strategies and
tactics are employed. Goals must be set, and decisions must be based on good
information.
Many of the most successful law enforcement agencies in other countries are using
crime mapping,
crime pattern analysis, and data-driven management to support community policing,
problem solving,
crime prevention and crime rate reduction.
6.2.
•
Recommendation
A system for recording crime data based on points is recommended.
50
It is recommended to kernel density estimation, this method is used to generate a

continuous
crime density surface from crime point data. The analyst begins with a dot map of
crime event.
A particular benefit of this method is that unlike in block aggregation, the
analysis is not limited
to some arbitrary geographic boundary, and it is much easier to discern spatial
patterns than a
complex point map.
It would be beautiful to use animation, changes in crime distributions over time

and to relate
these changes to crime prevention interventions in area. The animation works with
crime density
maps generated by kernel density, because they are visually appealing and easy to
comprehend
in the short time interval of the animation.
51
References
Alex, H. and B. Kate (2001). MAPPING AND ANALYSING CRIME DATA. London, Taylor and
Francis, 9-22.
Alex, H. and B. Kate (2001). MAPPING AND ANALYSING CRIME DATA. London, Taylor and
Francis, 185-202.
Bailey, T. and A. Gutrell (1995). Interactive Spatial Data Analysis. New York
Wiley.
Block, C. R. (1995). Crime Analysis Through Computer Mapping, Washington, D.C:
Police Executive Research forum, 15-32.
Bottoms, A. E. and P. Wiles (1995). "Crime and insecurity in the city, in C.
Fijnaut et al. (ed) Changes
in security, Crime and CriminalJustic in Europe, two vols." The Hague: Kluwer.
Brown, D. and J. Dalton (1998). Regional Crime Analysis Program (RECAP), presented
at Crime
Mapping Specialists Meeting, National Institute of Justice, Washington, DC,
February.
Diehr, P. (1984). "small area statistics, large statistical problems." American
Journal of public Health
74: 313-14.
Harries, K. (2001). "Mapping Crime principle and practice."
http://www.ncjrs.org/html/niji/mapping:
105.
Herbert, D. T. (1982). The Geography of urban crime New York, Longman.
Lawman, J. (1982). "Conceptual issues in geography of crime." Toward a Geography of
Control Annals, Association of American Geographers 76: 81-94.
Longley, P. A. (2001). Geographic Information System and Science.
Openshaw, S., M. Charlton, et al. (1987). "A mark 1 geographical analysis machine
for the automated
analysis of point data sets." International Journal of GIS 1(GIS): 335-58.
Openshaw, S., D. Waugh, et al. (1994). some ideas about the use of map animation as
a spatial analysis tool', in R.A. Earn Shaw and D. Watson (eds) Animation and
Scientific Visualization: Tools and
Applications. London: Academic press.
Sharifi, M. A. (2001). Introduction to Decision Support Systems and Multicriteria
Evaluation Techniques. Lecture Notes, ITC.
Sharifi, M. A. and V. H. Marijnen (2002). Spatial Decision Support Systems. Lecture
Note, ITC.
Slocum, T. (1999). "Thematic Cartography and Visualization, Upper Saddle River,
NJ." Prentic Hall.
Voogd, H. (1983). Multicriteria evaluation for urban and regional planning.
I
Williamson, D., S. McLafferty, et al. (1999). A better method to smooth crime

incidence data, Arc
user, January.
1- the national criminal justice statistical framework, Australian Bureau of
statistics
http://www.abs.gov.au/Ausstats/abs
%40.nsf/66f306f503e529a5ca25697e0017661f/8b467f3b44b2
32ecca256ab8007fee8e(accessed, January 2003)
2- Portland Police Bureau, Crime definition.
http://www.portlandpolicebureau.com/crimes.html (accessed, December 2002).
II
Appendices
Appendex1: The form of crime incidents in blotters:
Code of crime
Number
66 / 1 / 50
13 − 10 − 2002
60 / 5 / 150
13 − 10 − 2002
44 / 2 / 5
13 − 10 − 2002
2
3
Type of crime
Time
Rate of damage
Address
Theft
16:40
1000$
No32, street Azadi
Robbery
17
100$
No12, street Behboudi
Fraud
17:25
100000$
No11, street Jomhori
Appendex2: Trade-off procedure

A trade-off procedure (keeney and Raiffa, 1979) can be used for the assessment of
weights within a
linear additive value function or as an approximation for a more complex value
function. A trade-off
procedure is based on trade-offs between the criteria. A decision maker has to
answer questions of
type:
•
•
One unit of criterion C1 is equally attractive as how much of criterion C2.

How much are we ready to spend (sacrifice) from C1 in order to gain one unit of C2.
By repeating these kind of question for all pairs of criteria the weights of all
criteria can be determined.
Example
To clarify the calculations that have to be made assume a problem with three
criteria (C1, C2, and C3).
The first question to be answered is: one unit of C1 is equivalent to how many
units of C2? Assume the
answer is 2. This results in the following question:
C1/C 2=2
The second question to be answered is: one unit of C2 is equivalent to how many
units of C3? Assume
the answer is 1.5. This results in the following equation:
C2/C3=3/2
Now the two questions have been answered three equations leads us to the weights of
the three criteria.
III
C1 = C 2
2C 2 = 3C 3
C1 + C 2 + C 3 = 1
2C 2 + C 2 + 2 / 3C 2 = 1 ⇔ 11C 2 = 3 ⇔ C 2 = 3 / 11
C1 = 2.3 / 11 = 6 / 11
C 3 = 2 / 3.3 / 11 = 2 / 11
Appendix3: Spatial function
Spatial function is an operation for performing spatial analysis. All spatial
operations on the spatial
analyst user interface are classified as spatial functions, for example: distance,
slope, or density.
IV

1 Crime Mapping and Spatial Analysis PDF

Uploaded by

Document Informationclick to expand document informationcrime mapping criminology

Document Informationclick to expand document information

Copyright:

Available Formats

1 Crime Mapping and Spatial Analysis PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

1 Crime Mapping and Spatial Analysis PDF

Uploaded by

Copyright:

Available Formats

Crime Mapping and

ITC Primary supervisor: Y. Sun.MSc

Thesis submitted to the International Institute for Geo-information Science and

Degree Assessment Board

INTERNATIONAL INSTITUTE FOR GEO-INFORMATION SCIENCE AND EARTH OBSERVATION

Crime assessment based on crime per block,

Crime assessment based on density of crime per population.

LITERATURE REVIEW ON CRIME

DATA COLLECTION AND

Introduction of Tehran police

Law enforcement needs information management, especially location information.

Crime mapping, incident time and place.

What is the best methodology for crime spatial analysis?

Find a method to relocate in current boundary of police station for crime

The procedure used to achieve the aforementioned objectives is as follows:

Review of the prior works to get knowledge about them.

Multi criteria analysis.

To find suitable location for establishing new police station.

Figure 1.1 Census districts of Tehran

Figure 1.2 Districts 12 (case study)

Structure of the thesis

2. Literature review on crime analysis

Definition of crime and crime analysis

2.1.2. Types of crime analysis

• Tactical Crime Analysis:

sist sworn personnel in the identification of networks and apprehension of

Methods for automating the geographical analysis of crime incident data

their data with minimum of effort. It is an automated procedure that is designed to

Spatial perspectives on crime

“Although criminological research reaches back over some two centuries,

Tools in the spatial analysis of crime

3. Prepares voronoi diagrams (Theissen polygon) of individual crimes and performs a

Geographical variation in crime density is clearly visible on the smooth density

3. Data collection and preparation

Data modelling in GIS

A data model is a set of constructs for describing and representing selected

The spatial data:

The attribute data

Geodatabase model supports an object-oriented vector, data model that is a kind of

Figure 3.2 Police organization in Tehran

This step includes the following tasks:

Military (restriction area)

Georefrencing; used WGS_1984_UTM_Zone-39N projected coordinate systems (metric).

Police station no109, 537986X, and 3948851Y

Importing coverage files to Geodatabase using custom value.

Database Magement System

A database management system (DBMS), or simply a database system (DBS), consists

A collection of interrelated and persistent data (usually referred to as the

Retrieving information from the database.

1- Development of the system

Figure 3.3 A sample ER diagrams.

Police station: national law and enforcement office, which is in charge of

3.4.3. Enterprise Rules

Many blocks are under jurisdiction of a police station.

3.4.4. Skeleton tables

3.4.5. Data dictionary

4. Multicriteria Decision Making