A Stratified Traffic Accident Analysis by jbw10297

VIEWS: 48 PAGES: 51

									A Stratified Traffic Accident
           Analysis
        Case Study: City of Richardson, Texas
                         By: Tope Bello
           Masters in Geographic Information Sciences
                  University of Texas at Dallas
                        December 2005
                  Advisor: Michael Tiefelsdorf




                   Tope Bello                           1
Introduction

 The occurrences of traffic accidents are rare
  and random in space and time.
 These incidents can be represented spatially
  as points.
 Point pattern analysis will be used to explore
  the occurrence of traffic accidents in the City
  of Richardson.



                       Tope Bello                   2
Introduction

   This research performs a stratified analysis to
    check if school age children are involved in
    traffic accidents more around schools.
   For this purpose, data on accident locations,
    school locations, accident victims, and roads
    are vital to either support the above
    statement or reject it.



                        Tope Bello                3
Introduction

   Geographic Information Science and spatial
    statistics techniques were used to analyze
    the data for this research.
   Conclusions were drawn using visualization
    and exploratory statistical techniques.




                       Tope Bello                4
Project Objectives

The specific objectives of this project are to
  examine the following questions:
 Do traffic accidents occur to school age kids
  more around schools than elsewhere?
 Do the slower speeds in school speed zones
  prevent accidents to school age children?
 Does the temporal and spatial pattern of
  accidents to school age children differ
  patterns for all accidents?
                      Tope Bello                  5
Research Methodology
 Spatial queries to check if slower speeds in school speed zones
    prevent accidents to school age children.
   Kernel Density is used to explore if the temporal and spatial pattern
    of accidents to school age children differ patterns for all accidents.
     Kernel density is used to estimate how the density of events
       varies across a study area based on a point pattern (White et al
       2000).
   Bivariate K-Function analysis tests if traffic accidents occur to
    school age kids more around schools than elsewhere.
     The bivariate K- function is defined as the expected number of
       points of pattern 1 within a distance D of an arbitrary point of
       pattern 2, divided by the overall density of the points in pattern 1.




                                   Tope Bello                                  6
Literature Review

   Research in epidemiology were used more
    for addressing the research questions.
   The main area of focus was cluster analysis.
   V. Gomez-Rubio et al. 2005 discussed many
    techniques of detecting clusters of disease
    using R, a statistical package.




                       Tope Bello                  7
Literature Review

   Scan statistics for identifying the concentration of
    events in space have been modified over time by
    Openshaw et al. 1987, Besag and Newell 1991,
    Kulldorf and Nagarwalla 1995.
   Bryn Austin et al. 2005 had a research on clustering
    of fast food restaurants around schools. It used 400
    m radius and 800 meter radius buffer around
    schools to assess proximity to schools. The bivariate
    K-function was used to quantify the degree of
    clustering of fast food restaurants around schools.


                           Tope Bello                   8
Literature Review

   Bailey and Gatrell 1996 explained the
    applicability of K- function analysis to identify
    a point source for larynx cancer and lung
    cancer. A case was also stated on elevated
    risk of respiratory disease along busy main
    roads.
   Andrew Jones et al. 1996 also used the K-
    function analysis for the geographic
    distribution of road accidents in Norfolk,
    England.

                         Tope Bello                 9
Literature Review

   Peter Spooner et al 2004 developed a new
    technique for analyzing distribution of points
    on a network, called network K-function. This
    explores the interaction between points on a
    network.
   We have moved from traffic accident
    research to disease mapping, and finally to
    proximity to fast food. What these all have in
    common is that these events are spatial.
    Therefore they can be modeled for spatial
    analysis.
                        Tope Bello               10
DATA




 Tope Bello   11
Data: Overview

   Data used for this research were provided by
    different departments at the City of
    Richardson.
   Base map for city boundary, roads, and
    schools were provided by the GIS
    department. These are all spatial files.
   Traffic department provided data on traffic
    accidents.
   The Police department also provided traffic
    accident data.
                       Tope Bello              12
Data




       Tope Bello   13
Data: Traffic Accident Data

   The accident data provided by the traffic
    department was a database used for
    generating annual reports on traffic accidents
    occurrences.
   The data provided by the police department
    was a spatial file used to analyze traffic
    accidents occurrences by the crime analyst at
    the city of Richardson.


                        Tope Bello               14
Characteristics of the Data

   Police department accident data
       Geocoded
       Up to date
       Not detailed
       Restrictions on usage

   Traffic department accident database
       Not spatial
       Detailed
       Structured database



                                Tope Bello   15
Police Department Accident Data




                Tope Bello        16
Police Department Accident Data

   Address
   Speed Limit
   Intersection
   Distance from Intersection
   Direction
   Accident Date
   Accident Time
   Day of the Week
   Level of Damage
                       Tope Bello   17
Traffic Department Accident Data

   Year                                  Direction from
   Accident Number                        intersection
   STREET CODE                           Date
   Intersecting Street                   DOW (Day of the week)
    Code                                  Time
   Miles from intersection               Passenger age for 2
   Feet From Intersection                 units (potential of 4
                                           passengers)
                                          Pedestrian Age (for 2
                                           potential pedestrians)

                              Tope Bello                        18
PROCESSES




   Tope Bello   19
Processes




            Tope Bello   20
Data Preparation: Geocoding

   There were a total of
    10,883 accidents in
    Richardson between
    2000 and 2004.
   Address field was
    generated by
    concatenating the two
    field as an intersection.




                            Tope Bello   21
Data Preparation: Geocoding
                                          Geocode




   Create address locator
    using the road data.




                             Tope Bello             22
Data Preparation: Geocoding
                                      Geocode




   Geocode results show
    the number of records
    matched were low. This
    was due to wrong
    spellings and
    abbreviations in the
    data.
    Records were matched
    interactively.



                         Tope Bello             23
Data Preparation: Geocoding
                                       Geocode




   The statistics shows the
    final geocode results.
   The unmatched records
    were a result of
    incomplete records.




                          Tope Bello             24
Data Preparation: Queries
                                                Spatial
                                                Queries




   SQL queries were used
    to extract traffic
    accidents involving
    school age kids (TASK)
       TASK is defined as
        accidents involving kids
        between 5 and 15 years
        old, accidents that
        happened during the
        weekdays and during
        daytime.


                                   Tope Bello             25
Data Preparation: Accuracy

   Not all accidents
    happened at
    intersections
   Data contained
    information on distance
    away from the
    intersection.
   602 TASK and 8402
    other accidents from
    2000 – 2004 were used
    for analysis.

                          Tope Bello   26
Data




       Tope Bello   27
Data




       Tope Bello   28
Data Preparation: Speed zones
                                           Pictometry




   School speed zone data was also created for
    this project.
   GPS and Pictometry were considered for
    identifying speed zones around schools.
   A six inch spatial resolution image used for
    pictometry by the city of Richardson GIS
    department was used to identify the speed
    zones.


                       Tope Bello                       29
Data Preparation: Speed zones
                                School
                                Speed
                                Zones




                  Tope Bello             30
Data
                    School
                    Speed
                    Zones




       Tope Bello            31
ANALYSIS




  Tope Bello   32
Analysis: Spatial Queries
                                                       Spatial
                                                       Queries




   Do the slower speeds           Spatial query to identify
    in school speed                 TASK that fall in the
    zones prevent                   speed zones.
    accidents to school
    age children?




                       Tope Bello                                33
Analysis




           Tope Bello   34
Analysis: Spatial Queries
                                                     Spatial
                                                     Queries




   Perform basic GIS analysis to check for clustering.
   TASK that fall within various buffer distances from
    schools.



Distance 150 ft 250 ft          500 ft   1000 ft   1500 ft

No of       0       8      22            61      150
TASK        (0%)    (1.3%) (3.6%)        (10.1%) (24.9%)


                           Tope Bello                          35
Analysis: Kernel Densities
                                           Kernel
                                          Densities




   Does the temporal and spatial pattern of
    accidents to school age children differ
    patterns for all accidents?
   Kernel density explores the spatial
    distribution of TASK and other accidents.
   Selection of bandwidth was a challenge, and
    various bandwidths were explored to
    generate an heterogeneous pattern.


                       Tope Bello                     36
Analysis
                         Kernel
                        Densities




           Tope Bello               37
Analysis
                         Kernel
                        Densities




           Tope Bello               38
Analysis
                         Kernel
                        Densities




           Tope Bello               39
Analysis: Kernel Densities
                                               Kernel
                                              Densities




   Spatial analyst - raster calculator was used to
    estimate ratio of kernels, using the other
    accidents as a control population.
   Raster calculator – (task - others) / (task +
    others) = Kernel ratio
   This normalizes the distribution of TASK.




                        Tope Bello                        40
Analysis
                         Kernel
                        Densities




           Tope Bello               41
Analysis: Bivariate K-Function
                                                   K-Function
                                                    Analysis




   Do traffic accidents occur to school age kids more
    around schools than elsewhere?
   Bivariate K-function identifies clustering of point
    pattern A around point pattern B.
   This was done using a statistics software called R.
   It has various packages that are recommended for
    different purposes.
   SPLANCS (Spatial Point Pattern Analysis Code)
    package was used for analyzing clustering of TASK
    around schools.

                           Tope Bello                           42
Analysis: Bivariate K-Function
                                               K-Function
                                                Analysis




   Foreign package was also used. This enables the
    use of d-base files in the R environment.
   Data used in R contained information on X and Y
    coordinates of the TASK and schools.
   The average shortest distance was relevant to
    identify the catchments area for each school.
   The average shortest distance (2375.38ft) was
    calculated using Point distance in Arc GIS.



                         Tope Bello                         43
Analysis: Bivariate K-Function
                                                                                                                                       K-Function
                                                                                                                                        Analysis




                                                                                          K-Function for short distance With Toroidal Shifts
                  K-Function With random toroidal shifts




                                                                                  2000
      4000




                                                                                  1000
      2000




                                                                            L12
L12




                                                                                  0
      0




                                                                            ^
^




                                                                                  -1000
      -2000




                                                                                  -2000
      -4000




              0      2000     4000       6000   8000       10000                              200   400    600       800     1000   1200   1400

                                distance (ft)                                                                    distance (ft)




             The Bivariate K-function checks the degree of
              clustering of TASK around schools.
             It uses a toroidal shift edge effect correction.

                                                                   Tope Bello                                                                       44
RESULTS AND DISCUSSIONS




          Tope Bello      45
Results and Discussions

   Do the slower speeds in school speed zones
    prevent accidents to school age children?
       No, Spatial queries show that TASK occur within the speed
        zones.
   Does the temporal and spatial pattern of accidents
    to school age children differ patterns for all
    accidents?
       Yes, Kernel densities shows a variation in pattern of TASK
        and other accidents.
   Do traffic accidents occur to school age kids more
    around schools than elsewhere?
       No, Bivariate K-function shows no significant deviation from
        a random pattern.


                                Tope Bello                           46
Results and Discussions

 Further studies will consider a heterogeneous
  background.
 Edge effect correction for future research.

 The event distance was calculated as the
  crow flies, network distances should be
  considered for future studies.
 Combination of the datasets from both the
  police and traffic department could have been
  more efficient.
                     Tope Bello               47
Conclusion

   Rowlingson and Diggle noted that GIS lacks
    the basic statistical functionality. Meaning
    there is limit to the statistical analysis that can
    be performed by a GIS.
   However, my research has shown that a
    combination of various GIS techniques can
    test complex statistical hypothesis.
   Hence, accidents to school age kids do not
    happen more around schools in the city of
    Richardson.
                          Tope Bello                  48
Reference
   Austin B, Melly S.J, Sanchez B.N, Patel A, Buka S, Gortmaker S.L
    (2005) “Clustering of Fast-food restaurant around schools: A novel
    application of spatial statistics to the study of food environments”
    American Journal of public health Vol. 95, No 9: 1575 – 1581
   Bailey T, Gatrell T (1995) “Interactive Spatial data analysis”
    Longman Scientific and technical, Essex, UK
   Besag J, Newell J (1991) “The detection of cluster in rare diseases”.
    Journal of the Royal Statistical Society A 154: 143 – 155
   Gomez-Rubio V, Ferrandiz-Ferragud J, Lopez- Quilez A (2005)
    “Detecting clusters of disease with R” Journal of Geographic
    Systems 7: 189-206
   Jones A.P, Langford I.H and Bentham G (1996) “The application of
    K-Function analysis to the geographic distribution of road traffic
    accident outcomes in Norfolk, England” Social Science and
    Medicine Vol. 42, No. 6: 879-885



                                  Tope Bello                            49
Reference
   Kulldorf M, Nagarwalla N (1995) “Spatial disease clusters: detection
    and inference” Statistics in Medicine 14: 799-810
   Openshaw S, Charlton M, Wymer C, Craft A.W (1987) “A mark I
    geographical analysis machine for the automated analysis of point
    data sets” International Journal of Geographical Information
    Systems 1: 335-358
   Rowlingson B.S and Diggle P.J (1993) “Splancs supplement –
    spatial and spatial- temporal analysis” Unpublished report,
    Lancaster University. U.K
   Spooner P.G, Lunt I.D, Okabe A, Shiode S (2004) “Spatial analysis
    of roadside Acacia populations on a road network using the network
    K-function” Landscape ecology 19: 491-499
   White S.J, Ashby D, and Brown P.J (2000) “An introduction to
    statistical methods for health technology assessment” Health
    Technology Assessment 2000, 4(8).

                                 Tope Bello                            50
        Thank you!
             Questions?




“I not only use all the brains that I have,
but all that I can borrow.”
Woodrow Wilson



               Tope Bello                     51

								
To top