Group 4 April 23, 2020
#What causes accidents in a specific location?
##Setup
For the setup we loaded text files related directly to Michigan Accidents as well as County data. Additionally, we loaded some of our base packages. As the project went on, we went back and added any additional packages that were needed to dig deeper into the data.
knitr::opts_chunk$set(echo = FALSE) ##
## Attaching package: 'dplyr'
## The following objects are masked from 'package:stats':
##
## filter, lag
## The following objects are masked from 'package:base':
##
## intersect, setdiff, setequal, union
## ── Attaching packages ──────────────────────────── tidyverse 1.3.0 ──
## ✓ tibble 2.1.3 ✓ stringr 1.4.0
## ✓ readr 1.3.1 ✓ forcats 0.5.0
## ✓ purrr 0.3.3
## ── Conflicts ─────────────────────────────── tidyverse_conflicts() ──
## x dplyr::filter() masks stats::filter()
## x dplyr::lag() masks stats::lag()
## x purrr::map() masks maps::map()
##
## Attaching package: 'lubridate'
## The following object is masked from 'package:base':
##
## date
## Loading required package: RColorBrewer
#Part 1 - Variables
For the first phase in looking at specific locations and causes, we took a deep dive at some of our variables. As you can see below we took a summary of all whether characteristics on as a whole (state of Michigan.) By doing this we think we were able to see that although we see extremes in whether, nothing specifically jumps out as happening out of the ordinary during accidents. With this we were also able to note, although a majority of accidents aren’t happening under specific conditions, possibly the severity could be changed by this.
## Severity Distance.mi. Temperature.F. Wind_Chill.F.
## Min. :1.00 Min. : 0.0000 Min. :-17.00 Min. :-40.20
## 1st Qu.:2.00 1st Qu.: 0.0000 1st Qu.: 33.10 1st Qu.: 16.20
## Median :2.00 Median : 0.0000 Median : 52.00 Median : 27.90
## Mean :2.46 Mean : 0.2118 Mean : 50.79 Mean : 31.14
## 3rd Qu.:3.00 3rd Qu.: 0.0100 3rd Qu.: 70.00 3rd Qu.: 40.30
## Max. :4.00 Max. :65.5990 Max. : 97.00 Max. : 95.00
## NA's :235 NA's :41797
## Humidity... Pressure.in. Visibility.mi. Wind_Speed.mph.
## Min. : 12.00 Min. :27.78 Min. : 0.000 Min. : 0.000
## 1st Qu.: 56.00 1st Qu.:29.75 1st Qu.: 9.000 1st Qu.: 6.000
## Median : 73.00 Median :29.98 Median :10.000 Median : 9.200
## Mean : 70.34 Mean :29.89 Mean : 8.544 Mean : 9.735
## 3rd Qu.: 86.00 3rd Qu.:30.14 3rd Qu.:10.000 3rd Qu.:12.700
## Max. :100.00 Max. :31.15 Max. :12.000 Max. :43.700
## NA's :278 NA's :212 NA's :212 NA's :6209
## Precipitation.in.
## Min. :0.00
## 1st Qu.:0.00
## Median :0.00
## Mean :0.02
## 3rd Qu.:0.01
## Max. :1.89
## NA's :58199
In Congruence with the in-depth look at the weather, we decided to take a look at the street indicators as variables. By doing this we were able to provide a graph that jumped out to us. Traffic signals, Junctions and Crossings have a significant amount of crashes compared to other traffic indicators. This is helpful to keep in mind during our work.
## # A tibble: 48 x 14
## # Groups: Amenity, Bump, Crossing, Give_Way, Junction, No_Exit, Railway,
## # Roundabout, Station, Stop, Traffic_Calming, Traffic_Signal, Turning_Loop
## # [48]
## Amenity Bump Crossing Give_Way Junction No_Exit Railway Roundabout Station
## <fct> <fct> <fct> <fct> <fct> <fct> <fct> <fct> <fct>
## 1 False False False False False False False False False
## 2 False False False False False False False False False
## 3 False False False False False False False False False
## 4 False False False False False False False False False
## 5 False False False False False False False False False
## 6 False False False False False False False False True
## 7 False False False False False False False False True
## 8 False False False False False False True False False
## 9 False False False False False False True False True
## 10 False False False False False True False False False
## # … with 38 more rows, and 5 more variables: Stop <fct>, Traffic_Calming <fct>,
## # Traffic_Signal <fct>, Turning_Loop <fct>, n <int>
## Amenity Bump Crossing Give_Way Junction
## False:87928 False:88693 False:84817 False:88690 False:81067
## True : 766 True : 1 True : 3877 True : 4 True : 7627
## No_Exit Railway Roundabout Station Stop
## False:88594 False:88436 False:88693 False:88295 False:88519
## True : 100 True : 258 True : 1 True : 399 True : 175
## Traffic_Calming Traffic_Signal Turning_Loop
## False:88690 False:78716 False:88694
## True : 4 True : 9978
## Warning: Removed 235 rows containing non-finite values (stat_bin).
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## Warning: Removed 41797 rows containing non-finite values (stat_bin).
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## Warning: Removed 6209 rows containing non-finite values (stat_bin).
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.
## Warning: Removed 278 rows containing non-finite values (stat_bin).
#Phase 2 - Location
Now that we have taken a quick look at a majority of the variables in our data, we were able to focus on location. You will see in multiple of our worksheets that we focused in on Counties. Below we started off by graphing the counts of accidents in certain counties. For specific location we thought it would be a good idea to focus in on Counties with the highest accidents. In a real-life scenario, this could be most useful for safety issues and change. As you can see below Genesee, Wayne and Kent county were our top three locations by count of accidents.
## County n
## Alcona : 1 Min. : 1.00
## Alger : 1 1st Qu.: 13.25
## Allegan: 1 Median : 27.00
## Alpena : 1 Mean : 1081.63
## Antrim : 1 3rd Qu.: 202.75
## Arenac : 1 Max. :27072.00
## (Other):76
###City
More specifically, we wanted to identify specific locations. So we created a dataset specifically with Cities in Genesee County. By doing this we were able to then count the accidents by city and identify the Cities with the top accident counts. Graphing that out, Flint had the highest rate of accidents by far in Gensee County. You will see later in this sheet that we did this for Wayne and Kent County as well.
#Phase 3 - Time
With Flint in Mind, we wanted to look at an additional variable that originally didn’t stand out to us, time. By looking at time we could see if there were specific patterns to when crashes were happening and if that has any specific affect. In looking at this information two things stood out. The hour and the Day that accidents were happening. A large portion of accidents were happening during the hours in which pedestrians would be going to or leaving from work. Additionally, a large portion of accidents were occuring during weekdays in comparison to weekends.
## Warning: Ignoring unknown parameters: binwidth, bins, pad
#Phase 4 - Filter
Below you will see the most important part of identifying specific locations. By filtering on city and grouping by street, we were able to Identify the top locations by street that accidents were occuring at in Flint. Moreover, we were able to create a dataset strictly based around the top location (I-75.) From there we were able to count term frequency on (I-75) street descriptions. This gave us the exact intersections in which accidents occured most frequently. Below, we do this filtering and also look into most variables using Flint speicifically as the location.
## Street n
## Buick Fwy N:1 Min. : 242.0
## Corunna Rd :1 1st Qu.: 262.2
## I-69 E :1 Median : 338.5
## I-69 W :1 Mean : 400.5
## I-75 N :1 3rd Qu.: 421.0
## N Dort Hwy :1 Max. :1009.0
## (Other) :4
## [1] 1009 49
## Severity Distance.mi. Temperature.F. Wind_Chill.F.
## Min. :2.000 Min. :0.0000 Min. :-15.00 Min. :-30.90
## 1st Qu.:3.000 1st Qu.:0.0000 1st Qu.: 28.90 1st Qu.: 9.30
## Median :3.000 Median :0.0000 Median : 46.00 Median : 22.60
## Mean :2.852 Mean :0.1662 Mean : 47.38 Mean : 24.12
## 3rd Qu.:3.000 3rd Qu.:0.0100 3rd Qu.: 68.00 3rd Qu.: 35.20
## Max. :4.000 Max. :8.2300 Max. : 95.00 Max. : 91.00
## NA's :5 NA's :452
## Humidity... Pressure.in. Visibility.mi. Wind_Speed.mph.
## Min. : 16.00 Min. :28.55 Min. : 0.200 Min. : 0.00
## 1st Qu.: 57.00 1st Qu.:29.82 1st Qu.: 9.000 1st Qu.: 6.90
## Median : 74.00 Median :30.01 Median :10.000 Median : 9.20
## Mean : 70.85 Mean :29.95 Mean : 8.553 Mean :10.19
## 3rd Qu.: 86.00 3rd Qu.:30.18 3rd Qu.:10.000 3rd Qu.:12.70
## Max. :100.00 Max. :30.76 Max. :10.000 Max. :42.60
## NA's :5 NA's :2 NA's :3 NA's :64
## Precipitation.in.
## Min. :0.0000
## 1st Qu.:0.0000
## Median :0.0000
## Mean :0.0182
## 3rd Qu.:0.0100
## Max. :0.7400
## NA's :654
###Flint Specific weekdays and time Crash data
## Warning: Ignoring unknown parameters: binwidth, bins, pad
## Warning: Use of `Specificlocation$Start_Time` is discouraged. Use `Start_Time`
## instead.
## Warning: Ignoring unknown parameters: binwidth, bins, pad
## [1] 388 49
## # A tibble: 1 x 14
## # Groups: Amenity, Bump, Crossing, Give_Way, Junction, No_Exit, Railway,
## # Roundabout, Station, Stop, Traffic_Calming, Traffic_Signal, Turning_Loop
## # [1]
## Amenity Bump Crossing Give_Way Junction No_Exit Railway Roundabout Station
## <fct> <fct> <fct> <fct> <fct> <fct> <fct> <fct> <fct>
## 1 False False False False False False False False False
## # … with 5 more variables: Stop <fct>, Traffic_Calming <fct>,
## # Traffic_Signal <fct>, Turning_Loop <fct>, n <int>
## Amenity Bump Crossing Give_Way Junction No_Exit
## False:388 False:388 False:388 False:388 False:388 False:388
## True : 0 True : 0 True : 0 True : 0 True : 0 True : 0
## Railway Roundabout Station Stop Traffic_Calming Traffic_Signal
## False:388 False:388 False:388 False:388 False:388 False:388
## True : 0 True : 0 True : 0 True : 0 True : 0 True : 0
## Turning_Loop
## False:388
##
## # A tibble: 2 x 14
## # Groups: Amenity, Bump, Crossing, Give_Way, Junction, No_Exit, Railway,
## # Roundabout, Station, Stop, Traffic_Calming, Traffic_Signal, Turning_Loop
## # [2]
## Amenity Bump Crossing Give_Way Junction No_Exit Railway Roundabout Station
## <fct> <fct> <fct> <fct> <fct> <fct> <fct> <fct> <fct>
## 1 False False False False False False False False False
## 2 False False False False True False False False False
## # … with 5 more variables: Stop <fct>, Traffic_Calming <fct>,
## # Traffic_Signal <fct>, Turning_Loop <fct>, n <int>
## Amenity Bump Crossing Give_Way Junction No_Exit
## False:1009 False:1009 False:1009 False:1009 False:965 False:1009
## True : 0 True : 0 True : 0 True : 0 True : 44 True : 0
## Railway Roundabout Station Stop Traffic_Calming
## False:1009 False:1009 False:1009 False:1009 False:1009
## True : 0 True : 0 True : 0 True : 0 True : 0
## Traffic_Signal Turning_Loop
## False:1009 False:1009
## True : 0
## # A tibble: 46,338 x 2
## Description n
## <fct> <int>
## 1 Accident on I-75 Northbound at Exit 118 M-21 Corunna Rd. 112
## 2 Accident on Hill Rd at Fenton Rd. 106
## 3 At I-75/Exit 18 - Accident. 96
## 4 At 12 Mile Rd/Exit 63 - Accident. 84
## 5 Accident on I-75 Southbound at Exit 118 M-21 Corunna Rd. 79
## 6 At Davison/I-96 Local/Exit 186 - Accident. 77
## 7 Accident on I-75 Northbound at Exit 131 M-57 Vienna Rd. 76
## 8 Accident on Court St at Center Rd. 75
## 9 Accident on US-23 Northbound at Exit 88 Grand Blanc Rd. 75
## 10 Accident on M-11 28th St at Eastern Ave. 74
## # … with 46,328 more rows
#Phase4 - Continued
We then went through this same process on Wayne Count and Kent County to see if there was a large difference from Genesee. In our project writeup we go indepth on what we found.
## Street n
## Chrysler Expy S :1 Min. : 279.0
## Edsel Ford Fwy E :1 1st Qu.: 362.0
## Edsel Ford Fwy W :1 Median : 433.0
## Edward J Jeffries Fwy:1 Mean : 657.7
## Fisher Fwy N :1 3rd Qu.: 996.0
## Grand River Ave :1 Max. :1825.0
## (Other) :7
## [1] 1009 49
## Severity Distance.mi. Temperature.F. Wind_Chill.F.
## Min. :2.000 Min. :0.0000 Min. :-11.0 Min. :-32.70
## 1st Qu.:3.000 1st Qu.:0.0000 1st Qu.: 36.0 1st Qu.: 19.30
## Median :3.000 Median :0.0000 Median : 54.0 Median : 29.00
## Mean :2.785 Mean :0.1221 Mean : 52.5 Mean : 30.06
## 3rd Qu.:3.000 3rd Qu.:0.0790 3rd Qu.: 70.0 3rd Qu.: 37.70
## Max. :4.000 Max. :5.0500 Max. : 93.0 Max. : 85.00
## NA's :5 NA's :1027
## Humidity... Pressure.in. Visibility.mi. Wind_Speed.mph.
## Min. : 19.00 Min. :28.41 Min. : 0.200 Min. : 0.00
## 1st Qu.: 56.00 1st Qu.:29.83 1st Qu.:10.000 1st Qu.: 5.80
## Median : 69.00 Median :30.01 Median :10.000 Median : 9.00
## Mean : 69.46 Mean :29.96 Mean : 8.895 Mean : 9.36
## 3rd Qu.: 86.00 3rd Qu.:30.15 3rd Qu.:10.000 3rd Qu.:11.50
## Max. :100.00 Max. :31.03 Max. :10.000 Max. :33.40
## NA's :5 NA's :6 NA's :5 NA's :119
## Precipitation.in.
## Min. :0.0000
## 1st Qu.:0.0000
## Median :0.0000
## Mean :0.0314
## 3rd Qu.:0.0300
## Max. :0.8400
## NA's :1329
## # A tibble: 3 x 14
## # Groups: Amenity, Bump, Crossing, Give_Way, Junction, No_Exit, Railway,
## # Roundabout, Station, Stop, Traffic_Calming, Traffic_Signal, Turning_Loop
## # [3]
## Amenity Bump Crossing Give_Way Junction No_Exit Railway Roundabout Station
## <fct> <fct> <fct> <fct> <fct> <fct> <fct> <fct> <fct>
## 1 False False False False False False False False False
## 2 False False False False False False False False False
## 3 False False False False True False False False False
## # … with 5 more variables: Stop <fct>, Traffic_Calming <fct>,
## # Traffic_Signal <fct>, Turning_Loop <fct>, n <int>
## Amenity Bump Crossing Give_Way Junction No_Exit
## False:1825 False:1825 False:1825 False:1825 False:1450 False:1825
## True : 0 True : 0 True : 0 True : 0 True : 375 True : 0
## Railway Roundabout Station Stop Traffic_Calming
## False:1825 False:1825 False:1825 False:1825 False:1825
## True : 0 True : 0 True : 0 True : 0 True : 0
## Traffic_Signal Turning_Loop
## False:1821 False:1825
## True : 4
Kent County
## Street n
## 28th St SE :1 Min. :245.0
## Burton St SE :1 1st Qu.:328.5
## Division Ave S :1 Median :360.0
## Gerald R Ford Fwy W:1 Mean :411.9
## I-96 E :1 3rd Qu.:494.2
## Leonard St NE :1 Max. :661.0
## (Other) :2
## [1] 661 49
## Severity Distance.mi. Temperature.F. Wind_Chill.F.
## Min. :2.000 Min. :0.0000 Min. :-9.90 Min. :-28.40
## 1st Qu.:2.000 1st Qu.:0.0000 1st Qu.:33.00 1st Qu.: 15.00
## Median :2.000 Median :0.0100 Median :48.90 Median : 27.20
## Mean :2.186 Mean :0.2345 Mean :48.09 Mean : 30.90
## 3rd Qu.:2.000 3rd Qu.:0.3600 3rd Qu.:67.00 3rd Qu.: 41.77
## Max. :4.000 Max. :3.8600 Max. :91.90 Max. : 90.00
## NA's :2 NA's :263
## Humidity... Pressure.in. Visibility.mi. Wind_Speed.mph.
## Min. : 17.00 Min. :28.47 Min. : 0.100 Min. : 0.00
## 1st Qu.: 62.00 1st Qu.:29.65 1st Qu.: 6.000 1st Qu.: 6.90
## Median : 78.00 Median :29.96 Median :10.000 Median : 9.20
## Mean : 74.28 Mean :29.85 Mean : 7.733 Mean :10.41
## 3rd Qu.: 89.00 3rd Qu.:30.14 3rd Qu.:10.000 3rd Qu.:13.80
## Max. :100.00 Max. :30.79 Max. :10.000 Max. :35.70
## NA's :2 NA's :2 NA's :2 NA's :25
## Precipitation.in.
## Min. :0.0000
## 1st Qu.:0.0000
## Median :0.0000
## Mean :0.0196
## 3rd Qu.:0.0100
## Max. :0.4200
## NA's :325
## # A tibble: 2 x 14
## # Groups: Amenity, Bump, Crossing, Give_Way, Junction, No_Exit, Railway,
## # Roundabout, Station, Stop, Traffic_Calming, Traffic_Signal, Turning_Loop
## # [2]
## Amenity Bump Crossing Give_Way Junction No_Exit Railway Roundabout Station
## <fct> <fct> <fct> <fct> <fct> <fct> <fct> <fct> <fct>
## 1 False False False False False False False False False
## 2 False False False False True False False False False
## # … with 5 more variables: Stop <fct>, Traffic_Calming <fct>,
## # Traffic_Signal <fct>, Turning_Loop <fct>, n <int>
## Amenity Bump Crossing Give_Way Junction No_Exit
## False:661 False:661 False:661 False:661 False:573 False:661
## True : 0 True : 0 True : 0 True : 0 True : 88 True : 0
## Railway Roundabout Station Stop Traffic_Calming Traffic_Signal
## False:661 False:661 False:661 False:661 False:661 False:661
## True : 0 True : 0 True : 0 True : 0 True : 0 True : 0
## Turning_Loop
## False:661
##