Relationship between Convenience Store Robberies and Road Environment Masahiro Murakami, Kotaro HIGU

1
Relationship between Convenience Store Robberies
and Road Environment  Masahiro Murakami,
Kotaro HIGUCHI and Akihiro SHIBAYAMADepartment
of Architecture, Faculty of Engineering,
Kogakuin University, JapanPresented at 7th
International Conference onDesign Decision
Support Systems in Architecture and Urban
Planning 2 - 5 July 2004
2
Objective
Geographical Information System
the number of convenience store robberies has
rapidly increased.
Road environment around robbed convenience store
Indexes of graphs and networks
The objective of this study is to demonstate the
road characteristics that are conducive to
convenience store robberies.
3
Plan of Presentation

• Background Information
• Data collection
• Characteristics of road networks around robbed
  convenience stores
• 4. Characteristics of formations of road
  networks around robbed convenience stores
• 5. Conclusions

4
Trend of convenience store robberies in Japan
and Tokyo (1994-2003)
The rate of criminal arrests has concurrently
decreased sharply, revealing that the police can
not fully cope with the present situation.
5
Why does this study focus on road environment
around robbed convenience stores ?
Crimes are known to occur in locations with
spatial factors conducive to criminal acts, as
has been explained in earlier studies on CPTED
(Crime Prevention through Environmental Design).
6
Why does this study focus on road environment
around robbed convenience stores ?
There are few previous studies on this topic in
Japan, and to the best of our knowledge, one
study was presented by Kashiwabara et al. (1997).
Qualitative Factors associated with convenience
store robberies - Density of police office -
Land use (residential area or commercial area)
- Population density - Traffic - Road
environment etc.
7
Why does this study focus on road environment
around robbed convenience stores ?
We demonstrate quantitatively relationship
between convenience store robberies and road
environment by using GIS and indexes of graphs
and networks.
8
Data Collection
Partial information on convenience store
robberies - The date and time of the robbery -
The name and address of a robbed convenience
store - The characteristics of the robber - The
amount of money stolen etc.
Digital database on Asahi newspaper report
9
Data Collection
10
Data Collection
Digital maps with a scale of one-twenty five
hundredth (Bureau of City Planning Tokyo
Metropolitan Government)
Road networks within a 1 kilometer radius of
robbed convenience stores are extracted.
11
Data Collection
An example of the database for investigating road
characteristics that are conducive to convenience
store robberies (Case 1)
We customized two GIS software with Visual Basic
and used them to organize data and analyze road
networks.
12
Characteristics of road networks around robbed
convenience stores
Ratio of total road length to road width
There are many narrow roads in the road networks
around the robbed convenience stores.
13
Characteristics of road networks around robbed
convenience stores
14
Characteristics of road networks around robbed
convenience stores
The whole district, as well as the neighborhood
of the robbed convenience stores, has many roads
of widths less than 6 meters.
Narrow roads of widths less than 6 meters are
broadly distributed around the robbed convenience
stores.
15
Characteristics of road networks around robbed
convenience stores
On narrow roads, there is generally little
traffic from midnight to early morning, making
districts with many narrow roads advantageous to
robbers.
In districts with many narrow roads, robbers can
approach and run away from convenience stores
without being witnessed by passers-by from
midnight to early morning.
16
Characteristics of road networks around robbed
convenience stores
It is thought that having many narrow roads
around convenience stores is one of the factors
conducive for robberies.
17
Characteristics of road networks around robbed
convenience stores
Areas around the robbed convenience stores have a
widely and continuously ranging high density of
narrow roads and many intersections.
18
Characteristics of road networks around robbed
convenience stores
these areas have both narrow roads and many
intersections distributed throughout the
district.
19
Characteristics of road networks around robbed
convenience stores
In such areas, robbers may elude their pursuers
and escape without being witnessed by passers-by
due to the density of narrow roads and poor
visibility.
20
Characteristics of road networks around robbed
convenience stores
Based on these results, it is thought that the
road networks around robbed convenience stores
are potentially conducive to these crimes.
21
Characteristics of road network form around
robbed convenience stores
The road networks around robbed convenience
stores are analyzed using indexes of graphs and
networks.
Indexes of graphs and networks
Where, e is the number of roads, v is the number
of intersections, L is total of road length, and
d (i, j) is the shortest distance between
intersections.
22
Characteristics of road network form around
robbed convenience stores
Results calculated on indexes of graphs and
networks
There are many T-intersections in the road
networks.
23
Characteristics of road network form around
robbed convenience stores
Results calculated on indexes of graphs and
networks
The poor visibility around T-intersections allows
robbers to elude their pursuers there.
It is thought that the number of T-intersections
around convenience stores is one of the spatial
factors associated with convenience store
robberies.
24
Characteristics of road network form around
robbed convenience stores
Results Calculated on indexes of graphs and
networks
Takagi, Taniguchi et al. (1991) have indicated
that analyzing synthetically these indexes is so
effective for clarifying structural differences
and common characteristics of road networks.
We found that the road networks around the robbed
convenience stores had many structural
differences, but we failed to demonstrate common
characteristics to all road networks using these
methods.
25
Characteristics of road network form around
robbed convenience stores
Characteristics of Case 1 - road network is
considerably intricate. (high p-value, 38.2)
- traffic efficiency is high. (low average of
Ai, 1108) - the shape and size of block is
regular as a whole. (low standard
deviation of Ai, 165)
Case 1
The road network of Case 1 have the
characteristic that even a robber with a poor
sense of the locality may manage to successfully
escape.
26
Characteristics of road network form around
robbed convenience stores
Characteristics of Case 3 - road network is
intricate. (moderately high p-value, 35.7) -
traffic efficiency is high. (low average of
Ai, 1120) - the size of the block is much
smaller on the average and the shape and
size of block is comparatively irregular. (low
?-value, 38.2) (high standard deviation of Ai,
185)
Case 3
Though this road network has high traffic
efficiency, it is difficult to readily escape
without knowledge of the locality.
27
Characteristics of road network form around
robbed convenience stores

• We have identified the following reasons
• Only a small number of convenience store
  robberies were included in this study.
• 2. We did not have access to the means of escape
  (on foot, or by bicycle, motorcycle or car) for
  the robbers and could not analyze this important
  point.
• 3. The level of knowledge of the locality was
  not available.

28
Characteristics of road network form around
robbed convenience stores
From this frequency distribution of Ai, we can
analyze how the areas of high intersection
density are distributed within road networks
(Toi and Yoshitake, 1992).
Frequency distribution of Ai
29
Characteristics of road network form around
robbed convenience stores
we can understand precisely the distribution of
high density areas.
Kernel Density of intersections and Contour of Ai
(Case 3)
30
Characteristics of road network form around
robbed convenience stores
Not only did we find many peaks in all districts,
but we also found several peaks with the near Ai
value.
31
Case 1
Case 2
Case 3
Kernel Density of intersections and Contour of Ai
Case 4
Case 5
- We succeeded to precisely identify the
distribution in high-density areas. - It was
found that the distribution of those areas tended
to be similar.
32
Conclusions

• Results can be summarized as follows
• We succeeded to demonstrate factors that
  influence the incidence of convenience store
  robberies.
• Based on this result, we found that the road
  networks around the robbed convenience stores
  potentially had common characteristics.
• GIS and indexes of graphs and networks are
  effective techniques for explaining road
  characteristics that are conducive to convenience
  store robberies.

33
Thank you1-24-2, Nishi-Shinjuku,
Shinjuku-ku, Tokyo, Japan 163-8677 Department of
Architecture, Faculty of Engineering, Kogakuin
UniversityTel 81-3-3340-3116 Fax
81-3-3340-3116e-mail murakami_at_cc.kogakuin.ac.jp