-
Notifications
You must be signed in to change notification settings - Fork 2
Expand file tree
/
Copy pathindex.html
More file actions
427 lines (426 loc) · 18.5 KB
/
index.html
File metadata and controls
427 lines (426 loc) · 18.5 KB
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
<!DOCTYPE html>
<html lang="en">
<head>
<meta charset="UTF-8" />
<meta
content="width=device-width, user-scalable=no, initial-scale=1.0, maximum-scale=1.0, minimum-scale=1.0"
name="viewport"
/>
<meta content="ie=edge" http-equiv="X-UA-Compatible" />
<title>Interagency Work Zone Traffic Data Modeling and Analysis</title>
<script src="https://ajax.googleapis.com/ajax/libs/jquery/3.5.1/jquery.min.js"></script>
<script src="src/js/bootstrap.min.js"></script>
<script src="src/js/bootstrap.bundle.min.js"></script>
<link href="src/css/bootstrap-grid.min.css" rel="stylesheet" />
<link href="src/css/bootstrap.min.css" rel="stylesheet" />
<link href="src/css/main.css" rel="stylesheet" />
</head>
<body data-offset="150" data-spy="scroll" data-target="#sidebar">
<div class="container-fluid">
<div class="row" id="body__row">
<!-- <div id="nav_fix-top">
<img alt="navigation-icon" id="nav__open-btn" onclick="nav_show()" src="src/image/nav.png">
<button id="nav__close-btn" onclick="nav_close()">Close</button>
</div> -->
<nav class="col-md-3 col-lg-3" id="sidebar">
<ul class="nav__content-wrapper nav__main">
<li class="nav-item">
<a
class="nav-link .active"
href="#main__cover-page"
id="nav__home-btn"
>Home</a
>
</li>
<li class="nav-item">
<a class="nav-link" href="#main__overview">Overview</a>
</li>
<li class="nav-item">
<a class="nav-link" href="#main__data">Project Data</a>
</li>
<li class="nav-item">
<a class="nav-link" href="#main__methodology">Methodology</a>
<ul>
<li>
<a
class="nav-link nav-subsection"
href="#methodology__wrangling"
>Data Wrangling</a
>
</li>
<li>
<a
class="nav-link nav-subsection"
href="#methodology__clustering"
>Clustering</a
>
</li>
</ul>
</li>
<li class="nav-item">
<a class="nav-link" href="#main__results">Results</a>
<ul>
<li>
<a class="nav-link nav-subsection" href="#results__dashboard"
>Dashboard</a
>
</li>
<li>
<a
class="nav-link nav-subsection"
href="https://workzone-collision-predict.herokuapp.com/"
target="_blank"
>Web application</a
>
</li>
</ul>
</li>
<li class="nav-item">
<a class="nav-link" href="#main__about">Team</a>
</li>
<li class="nav-item">
<a
class="nav-link"
href="https://github.com/workzone-collision-analysis/capstone"
target="_blank"
>Github</a
>
</li>
<li class="nav-item">
<a
class="nav-link"
id="report_download"
download="report.pdf"
href="src/report.pdf"
>Report</a
>
</li>
</ul>
</nav>
<main class="col-md-12 col-lg-9 ml-sm-auto" id="main">
<section class="container-fluid main__section" id="main__cover-page">
<div id="logos">
<img alt="NYU CUSP LOGO" src="src/image/cusp.png" />
<img alt="HDR" src="src/image/hdr.png" />
</div>
<h1>
Interagency Work Zone<br />
Traffic Data Modeling and Analysis
</h1>
<h4>NYU | Center for Urban Science and Progress</h4>
<h4>Capstone Project | 2020</h4>
</section>
<section class="container-fluid main__section" id="main__overview">
<h2>Overview</h2>
<h3>Background</h3>
<p>
Road construction events are a necessary part of keeping road
infrastructure in good condition but can pose significant safety
problems when implemented. Temporary work zones for roadway
constructions have the potential to significantly impact mobility
and safety for all roadway users. An increase in the number of
people using local streets because of work zone diversion plans
may increase the likelihood of crashes, including crashes
involving vulnerable populations (e.g., cyclists, seniors, and
individuals with disabilities). To aid transportation authorities
in the city of New York better understand the extent of mobility
impacts associated with work zones, we propose a clustering
approach to predict the probability of a vehicle collision
occurring in the proximity of a road construction event (i.e. work
zone).
</p>
<h3 id="problemState">Problem Statement</h3>
<div>
<h1>
Can we use characteristics of streets and roadway construction
events to predict collision probability in future work zones?
</h1>
</div>
<h3>Scope</h3>
<div class="row">
<div class="col-8 col-sm-5 col-md-4 col-lg-4 scope__column">
<img alt="Dataset Diagram" src="src/image/database.png" />
<h4>Data Wrangling</h4>
</div>
<div class="col-8 col-sm-5 col-md-4 col-lg-4 scope__column">
<img alt="Clustering Diagram" src="src/image/clustering.png" />
<h4>Clustering</h4>
</div>
<div class="col-8 col-sm-5 col-md-4 col-lg-4 scope__column">
<img alt="Dashboard Diagram" src="src/image/dashboard.png" />
<h4>Dashboard</h4>
</div>
<div class="col-8 col-sm-5 col-md-4 col-lg-4 scope__column">
<img alt="Dashboard Diagram" src="src/image/webapp.png" />
<h4>Predictive Application</h4>
</div>
</div>
<p>
This project proposes using a k-means clustering approach to
predict the probability of a vehicle collision occurring in the
proximity of a work zone. The proposed clustering method is
applied to over 20,000 construction and emergency construction
events of relatively short duration in New York City to identify
types of work zones that may present greater safety risks. The
results of this project enables practitioners to employ
appropriate mitigation strategies during project programming and
develop effective transportation management plans.
</p>
</section>
<section class="container-fluid main__section" id="main__data">
<h2>Project Data</h2>
<p>
This project is unique in that all data sources used in this
research are publicly available datasets. The team did not have
access to a single, unified data source that provided information
about the location and timing of construction work zones within
NYC. Instead, the team brought together disparate datasets to
produce an approximation of work zone events and their attributes
(road type, duration, length, etc.). The information below shows
the sources and features taken from each dataset to produce a
single dataset of construction events.
</p>
<div class="row">
<div class="col-8 col-sm-5 col-md-4 col-lg-4 data__column">
<div class="data__column-icon">
<img
alt="road construction"
id="icon_511"
src="src/image/511.png"
/>
</div>
<h4>511 Events</h4>
<span>Geometry: Point</span>
<ul class="data__column-attribute">
<li>Location,</li>
<li>Duration,</li>
<li>Season</li>
</ul>
</div>
<div class="col-8 col-sm-5 col-md-4 col-lg-4 data__column">
<div class="data__column-icon">
<img
alt="road construction"
src="src/image/construction.png"
/>
</div>
<h4>Street Closure</h4>
<span>Geometry: Line</span>
<ul class="data__column-attribute" id="sc__attr">
<li>Length</li>
</ul>
</div>
<div class="col-8 col-sm-5 col-md-4 col-lg-4 data__column">
<div class="data__column-icon">
<img
alt="crash symbol"
id="crash_icon"
src="src/image/crash_icon.png"
/>
</div>
<h4>Crash</h4>
<span>Geometry: Point</span>
<ul class="data__column-attribute" id="crash__attr">
<li>Number of crashes</li>
</ul>
</div>
<div class="col-8 col-sm-5 col-md-4 col-lg-4 data__column">
<div class="data__column-icon">
<img alt="crash symbol" src="src/image/lion.png" />
</div>
<h4>LION</h4>
<span>Geometry: Line</span>
<ul class="data__column-attribute">
<li>Roadway type,</li>
<li>Street width,</li>
<li>Posted speed</li>
</ul>
</div>
</div>
</section>
<section class="container-fluid main__section" id="main__methodology">
<h2 id="section3">Methodology</h2>
<div id="methodology__wrangling">
<h3>Data Wrangling</h3>
<p>
The data sources were produced by various entities and used
different geometric representations. Therefore, there was a need
for a standardized street network onto which data sets, namely
LION, crashes, and work zones, could be joined. SharedStreets –
a tool that creates a shared reference system for disparate
street networks – provided a solution to connect the data. Each
record in the LION, collisions and WZ data was assigned a
‘SharedStreets geometry id’. Based on this id, characteristics
of streets, collisions, and work zones were linked. The data was
filtered to include only durations less than 24 hours, and
number of crashes happening within 900 feet of a work zone was
calculated. Then the Street Closures data was joined to get the
length attribute in feet.
</p>
<div class="methodology__image">
<img
alt="flow chart of the project"
src="src/image/flowchart.png"
/>
</div>
</div>
<div id="methodology__clustering">
<h3>Clustering</h3>
<p>
With the spatial datasets wrangled into a common reference
system, the team employed a k-means clustering algorithm. This
approach was selected to provide a predictive framework to
understand the probability of a crash occurring in the proximity
of a work zone given certain characteristics. The clustering
approach also provides the opportunity to identify different
“cohorts” of work zones and identify work zones that may have a
higher crash probability. Two separate clustering attempts were
made – one with a larger dataset that did not contain a length
attribute and one smaller subset of records that were able to be
matched to permit data that contained a length attribute. The
collision probability was then calculated by dividing the number
of WZs witnessing a crash by the total number of WZs for each
cluster.
</p>
<div class="methodology__image">
<img
id="clustering_img"
alt="clustering"
src="src/image/cluster_method.png"
/>
<img
id="clustering_img_mobile"
alt="clustering"
src="src/image/cluster_method_mobile.png"
/>
</div>
</div>
</section>
<section class="container-fluid main__section" id="main__results">
<h2 id="section4">Results</h2>
<h3>Clustering</h3>
<p>
The silhouette analysis was used to choose an optimal value for
the number of clusters k=4. The predicted probability (using the
train set) of a crash happening in each of the four clusters is
16%, 13%, 14%, and 16% respectively.
<a
href="https://workzone-collision-analysis.github.io/capstone/dashboard/"
target="_blank"
>
An interactive dashboard</a
>
was designed to serve as a tool for transportation authorities to
explore the public safety risks associated with historical work
events as well as the historical crash rates on all New York City
roads and intersections.
<a
href="https://workzone-collision-predict.herokuapp.com/"
target="_blank"
>A predictive web application</a
>
was also built off the results of the clustering methodology to
inform the planning of multiple new construction events
simultaneously. It would aid decision making to avoid situations
with high crash risk.
</p>
<div id="results__dashboard">
<h3>Dashboard</h3>
<iframe
src="https://workzone-collision-analysis.github.io/capstone/dashboard/"
></iframe>
</div>
</section>
<section class="container-fluid main__section" id="main__about">
<h2>Team</h2>
<p>
This project was completed as part of the 2020 Capstone process
for the Center for Urban Science and Progress (CUSP) at NYU. The
team would like to thank capstone sponsor HDR and their partners
for their help in the execution of this research.
</p>
<ul class="team__student">
<li class="student__column">
<div class="student__circle">
<span>Collier, John</span>
<div class="team__link">
<a
href="https://www.linkedin.com/in/john-collier-6557115b/"
>
<img alt="linkedin link" src="src/image/linkedin.png" />
</a>
<a href="https://github.com/johncollier">
<img alt="github link" src="src/image/github.png" />
</a>
</div>
</div>
</li>
<li class="student__column">
<div class="student__circle">
<span>Han, Seunggyun</span>
<div class="team__link">
<a href="https://www.linkedin.com/in/seunggyunhancodes/">
<img alt="linkedin link" src="src/image/linkedin.png" />
</a>
<a href="https://github.com/Aete">
<img alt="github link" src="src/image/github.png" />
</a>
</div>
</div>
</li>
<li class="student__column">
<div class="student__circle">
<span>Jaber, Linda</span>
<div class="team__link">
<a href="https://www.linkedin.com/in/lindajaber/">
<img alt="linkedin link" src="src/image/linkedin.png" />
</a>
<a href="https://github.com/lindaJaber">
<img alt="github link" src="src/image/github.png" />
</a>
</div>
</div>
</li>
<li class="student__column">
<div class="student__circle">
<span>Singh, Akhil</span>
<div class="team__link">
<a href="https://www.linkedin.com/in/akhil-k-singh/">
<img alt="linkedin link" src="src/image/linkedin.png" />
</a>
<a href="https://github.com/akhilksingh">
<img alt="github link" src="src/image/github.png" />
</a>
</div>
</div>
</li>
</ul>
<h3>Mentors</h3>
<ul class="team__mentor">
<li class="mentor__column">
<div class="mentor__circle">
<span>Ozbay, Kaan</span>
<div class="team__link">
<a href="https://www.linkedin.com/in/kaan-ozbay-4805413/">
<img
alt="linkedin link"
class="linkedin"
src="src/image/linkedin.png"
/>
</a>
</div>
</div>
</li>
<li class="mentor__column">
<div class="mentor__circle">
<span>Khan, Junaid</span>
</div>
</li>
</ul>
</section>
</main>
</div>
</div>
<script src="src/js/nav.js"></script>
</body>
</html>