=Paper= {{Paper |id=Vol-2259/aics_8 |storemode=property |title=Vehicular Traffic Flow Intensity Detection and Prediction Through Mobile Data Usage |pdfUrl=https://ceur-ws.org/Vol-2259/aics_8.pdf |volume=Vol-2259 |authors=Maurice Saliba,Charlie Abela,Colin Layfield |dblpUrl=https://dblp.org/rec/conf/aics/SalibaAL18 }} ==Vehicular Traffic Flow Intensity Detection and Prediction Through Mobile Data Usage== https://ceur-ws.org/Vol-2259/aics_8.pdf

Vehicular traffic flow intensity detection and
prediction through mobile data usage

Maurice Saliba1 , Charlie Abela1 , and Colin Layfield2
1
Department of Artificial Intelligence,
Faculty of ICT, University of Malta, Malta,
mauricesaliba@gmail.com, charlie.abela@um.edu.mt
2
Department of Computer Information Systems,
Faculty of ICT, University of Malta, Malta,
colin.layfield@um.edu.mt

Abstract. A novel approach, consisting of an ensemble of data-mining
and machine learning techniques, is proposed to prove that it is possible
to extract and predict vehicular traffic patterns from mobile usage data.
An anonymized mobile phone usage dataset from a telecommunications
provider in Malta was used to generate an origin-destination (OD) ma-
trix that defines the top two locations towards which each user travels
to through clustering. The OD matrix was used to infer user trips over
fastest routes between these top two locations across time. We then ap-
plied spatial binning techniques to deduce the aggregate distribution of
traffic load on the traffic network. A predictive model based on an arti-
ficial neural network was trained with grid nodes’ traffic levels in a time
series to predict traffic level for specific nodes.
Our findings are promising and show that the built models are more ef-
fective to measure and predict traffic flow demand for specific locations
rather than the actual traffic flow rate. The proposed solution needs
improvement by adding a dynamic traffic assignment to the whole al-
gorithm. This would give more accurate results, especially for traffic
flow points that tend to be congested, by capturing user route selection
changes and get more precise localization of delay causes.

Keywords: Mobility patterns · Vehicular traffic congestion · Artificial
intelligence · Machine learning.

1 Introduction

The dynamics of traffic flow are determined by the travel needs of the masses.
The daily commutes of every individual impacts those of others. The interac-
tion on a large scale of all the vehicles in a time series is difficult to model
and to predict in a robust and responsive manner [19]. Traffic sensors, cameras
and induction loops are all sources of information that can be used to both
detect high traffic intensity or even forecast it beforehand. However, traffic in-
tensity measurement with these methods is physically limited. Camera feeds and
2 Maurice Saliba et al.

inductive-loop detectors cannot be installed in every road of the transport in-
frastructure. Devices carried by travellers, or embedded in vehicles, are more
practical to build smart solutions for traffic management [19].
Mobile traces can be processed and used to offer location based services that
have a wide application spectrum that go beyond solving mobility issues [10, 4,
8, 9]. This formidable data source, however, poses a challenge. Location data,
which usually comes in large amounts, has to be harvested, ingested efficiently
and ideally processed in real time for the required final purpose which is value
added location based services.
The range of applications and branches of research abound on remote col-
lection of mobile users’ geolocation information. To name a few, applications
include: traffic patterns and prediction modelling, crowd management, hotspot
detection, lost device recovery, emergency rescue, use for investigative author-
ities, location-based recommendation and advertising systems, contextualized
information, social interaction based application, epidemiology etc.
Calabrese et al [4] emphasized that studies on human mobility patterns would
be vital for improved and sustainable urban planning and could boost the en-
vironment’s well being given that transportation in 2004 already accounted for
22% of primary energy use.
Steenbruggen et al [17] discuss how mobile geolocation data can be used to
differentiate weekday traffic patterns from those of weekends. Another specific
type of prediction based on mobile usage discussed in [10] is jam detection.
Macroscopic monitoring and analysis of vehicle mobility through mobile traces
is a wide area of study that has ramifications in many areas of research [17].
In this paper we focus on measuring traffic flow and the prediction of traffic
flow changes over time for a selection of locations by using mobile data usage. A
combination of data mining and machine learning techniques are used to devise
a data processing pipeline. Through this pipeline it is possible to:

1. process raw event data records containing cell tower locations, date and time
based on which we carry out preliminary descriptive statistical analysis;
2. zoom into the main areas of activity of users by using unsupervised machine
learning techniques that cluster the most dense groups of geolocation data
points;
3. determine routes between these main activity areas and collect spatial-grid
aggregated data from daily trips done along these routes from thousands of
users;
4. use the transformed data that is representative of traffic flow in various loca-
tions to train and validate a predictive model using artificial neural networks
[4, 18, 10, 2];
5. feed visualization tools that enable insightful visual inspection of traffic pat-
terns projected on maps.

A selection of methods that are encountered in literature are applied and evalu-
ated. The real challenges arise in the quest for a high spatio-temporal resolution
when modelling traffic, given that mobile usage records’ geolocation dataset is
Vehicular traffic flow intensity detection and prediction 3

sparse and tracks the position of users with a considerable margin of displace-
ment error [10, 8]. In section 2, background and related work, we will expound
on techniques used for general human mobility modelling from mobile call detail
records (CDRs) found in literature. In section 3 the methodology to form the
data processing pipeline is described. Sections 4 and 5 follow with results and
evaluation and a conclusion with possible future work respectively.

2 Background and Related work
In this section we will go over mainstream techniques and approaches that make
use of mobile data for traffic flow detection and prediction.

2.1 Mobile location data sources
Our research revolves around geolocation attributes of mobile data usage. Mo-
bile device location traces have their limitations when used for vehicular traffic
analyses. In contrast to surveys, they lack demographics [4, 6] and the market
share of a mobile service provider that made the dataset available for scientific
research might not be really representative of the commuting patterns [3]. Many
studies highlighted the importance of removing bias when pre-processing such
datasets before any further processing is done [12, 18]. Passive data, gathered in
the form of CDRs, are not suited to extract different modes of travelling, route
assignment and the classification of detailed activity types [6].
Mobile device location data is not only limited to data that originates from
cellular networks. Global Positioning System (GPS) is the most current reliable
source of geolocation because of their higher resolution with a lower margin of
error [9]. Using GPS data for a mobility study is more challenging because it
needs the continuous consent of users to get such data and drains the battery
quickly especially because of long signal acquisition time [20, 1].
Research generally focuses on voice CDRs to trace mobility. Mobile data
usage was rarely found in literature to be used to detect vehicular traffic or
predict it because of unavailability of such datasets [10, 3].

2.2 Origin and destination matrices computation
A recurrent topic in traffic flow analyses is the study of how to deduce origin
and destination (OD) locations for travelling vehicles [12]. Numerous research
work focused on tackling the problem posed by traffic congestion detection by
first deducing the OD matrix [18, 12, 2–4, 6].
ODs are used to extract main activity hubs. Gonzalez et al [8] state that
40% of the time users are at their two preferred locations. Therefore most trips
can be mostly explained as being between several locations since users tend to
be highly inclined to be regular in spatial and temporal terms. All this leads to
the safe assumption that the majority of trips are between home and work. In
literature it is commonly found that locations that were likely to be recorded in
4 Maurice Saliba et al.

OD matrices were home and work [3, 6, 16]. In [3] home location is detected by
checking which 500 metres square cell has the most activity during the night for
every specific user. Colak et al [6] and Calabrese et al [3] also label zones such
as home and work and try to find purpose behind other types of trips.

2.3 Trip generation from OD matrices

Route selection is necessary to link origins and destinations from OD matrices
to generate OD trips. In Iqbal et al[12] the route is determined by a function
of least travel time path. In Toole et al[18], Open Street Maps3 (OSM), which
is an open source map editing framework, is used to infer routing. Some studies
assign trips to a user when there are consecutive calls in the same day and the
calls are done from different locations. Typically, two consecutive ’stays’ that
are not more than 1 day apart would constitute a trip [6, 18].

3 Methodology

In this section we describe step by step how we extracted traffic flow information
and built predictive models from mobile data usage.

3.1 Dataset used

Experimentation was done on an anonymized cell generated CDRs’ dataset that
was provided by GO Plc Malta4 , which is one of the main Malta telecommu-
nication service providers. The dataset was recorded in October 2016 and had
approximately 100 million mobile data usage records. Main data fields of interest
in the data structure were an anonymous identifier, timestamp, volume, duration
and geo-coordinates.

3.2 Traffic flow detection by trip generation assigned traffic

The method we adopted to detect traffic on the road network involved first the
generation of an OD matrix that contained main stay locations for users in a time
series. A trip was generated between each main location for each user as it will
be explained in section 3.4. The trip information includes turn by turn directions
with longitude and latitude coordinates. Traffic load was assigned to junctions
and turns depending on the time retrieved from OSM data. The major challenge
here proved to be the traffic assignment, given that there is an interaction of a
lot of vehicles at a given point in time with a complex structure of roads and
possible unexpected events such as weather, accidents and road closures.
3
https://www.openstreetmap.org
4
https://go.com.mt
Vehicular traffic flow intensity detection and prediction 5

3.3 Main activity hubs extraction through DBSCAN clustering
One of the main steps of the proposed solution was to derive main areas of activ-
ity from the mobile data usage of subscribers. This was achieved using clustering.
Clustering was also used to remove noise in the form of sudden displacements
through frequent oscillations between cell towers by finding a centroid of activity.
DBSCAN (density based spatial clustering of applications) clustering was chosen
over k-means for reasons similar to those explained in [5, 11]. DBSCAN does not
need to set the number of clusters at the outset. Moreover it finds clusters of
non-spherical nature and leaves noisy elements out of the computed clusters [5].
The algorithm is more sensitive to density rather than to aggregate distances of
surrounding points.

Fig. 1. DBSCAN clustering to find main user activity hubs. (Sample illustration)

The chosen values for DBSCAN hyper-parameters were 500m for radius
(based on average distance of 350m between cell towers in urban areas), min-
imum required points was set to 3 and euclidean distance was chosen as the
distance metric. Figure 1 shows plotting of a sample of mobile activity clusters
delineated by rectangular boundaries.

3.4 OD Matrix trip Generation
We decided to focus on two main areas of activity per user as the basis of our OD
matrix generation, namely home and work location. This was based on results
reported in the literature review (refer to section 2).
The top two mobile data usage activity clusters per user where retrieved from
the resulting users’ clusters computed through the DBSCAN. The user’s CDRs
6 Maurice Saliba et al.

that had geographical coordinates located in the two main activity cluster areas
were then filtered into a new dataset through the spatial joining technique [7].
This process resulted in a dataset containing all data usage records that had a
location in either of the top two clusters for any user in the time series.

3.5 Trip generation, route choice and traffic assignment

We inferred the routes between origin and destination from the OSM using a
method discussed in Toole et al [18]. The fastest route was assigned for each
entry in the OD matrix together with duration information from the trip. The
routing engine Open Source Routing Machine (OSRM)5 was used for this pur-
pose. Choosing the fastest route by default is a limitation of this research and
must be considered as a source of bias.

Fig. 2. Traffic flow count snapshot at 7:00 a.m. October 2016. Illustrated through
CartoDB temporal mapping. Tool allows to visually investigate traffic by sliding the
date-time setting. Circle size is proportional to traffic flow count. Large circles are
depicted in notorious traffic hotspots in Malta.

5
http://project-osrm.org/docs/v5.15.2/api/#route-service
Vehicular traffic flow intensity detection and prediction 7

The difference between the actual trip duration retrieved from observed de-
partures and arrivals per user, and the OSRM derived trip duration, was consid-
ered to be the global trip delay. After computing delays for each trip per user,
aggregated statistics were collated to describe typical delays at different hours
for both weekdays and weekends. Trip delay differences are evident even between
Saturdays and Sundays but they were highly similar for weekdays.
Traffic was assigned to the road network depending on manoeuvres’ steps
with geolocation given by the OSRM. These steps have the time information
when user travelled through the geolocation. This information was used to dis-
tribute traffic count on the road network. Figure 2 illustrates traffic flow count
data at a given point in time.

3.6 Prediction using a Multilayer Perceptron Classifier (MLPC)
The next step in the data processing pipeline consisted in predicting traffic from
a stipulated time ahead for a given location point. This prediction had to be
based on data that was harvested some time before. All of the original datasets
had records with timestamps set in the past, so we simulated prediction of traffic
flow by trying to forecast traffic at a certain point in time which is ahead of a
given timestamp. Evaluation was then carried out with a variable number of first
principle component analysis (PCA) components, prediction multi-steps ahead
and possible classes that describe level of traffic. The training and testing inputs
for the MLPC model were traffic counts in the grid and the output was the level
of traffic at a certain location in the relative future.
The multi-step time series prediction model was trained and tested with a
variable amount of steps ahead. Each step was already defined to be 5 minutes
long. The experimentation was performed with 3, 6, 12 and 288 steps ahead that
transalte to 15 minutes, 30 minutes, 1 hour and 1 day.

4 Results and evaluation
In our evaluation process we evaluated four main experimental procedures:
1. Average trip count per hour for weekdays and weekdays;
2. Average global trip delay per hour for weekdays and weekends;
3. Traffic flow count in a selection of locations;
4. Traffic count prediction for a selection of locations.

4.1 Average trip count per hour
Through a linear regression we showed that the trip distribution derived from
mobile usage CDRs’ generated OD Matrix has a significant correlation with the
trip distribution as reported in a National Household Travel Survey (NHTS)
conducted in 2010 [14] (see figure 3). The same linear regression model was used
to scale up the trip distribution in 2010 to the one registered in 2016 for this
study. A correlation statistical analysis gave the result of a Pearson correlation
coefficient of 0.94 with a p-value of 1.13628e − 11∗ .
8 Maurice Saliba et al.

Fig. 3. Comparison between OD average trip distribution over a month and Transport
Malta 2010 survey results ([14])

4.2 Trip average delay per hour
Seven whole days of Google Distance matrix API (DMAPI) data from June 2018
was scraped by retrieving duration information for every quarter of an hour.
Estimated trip delay was calculated by subtracting the estimated trip duration
from the trip duration in traffic. Average trip delay was then computed per hour
for four different Malta routes that link cities, namely Mosta to Marsa, Mellieha
to Swieqi (MS), Birkirkara to Sliema (BS) and Valletta (the capital city) to
Mgarr (VM). The overall trip delay average was calculated for these routes.
Correlation results showed that there is a strong linear relationship between
the routes’ trip delay pattern which were investigated with the DMAPI. Corre-
lation between DMAPI and OD-OSRM trip delay estimation was less but still
considerable. Between DMAPI average overall trip delay and OD-OSRM non
shifted expected trip delay data, there was a correlation of 0.69 (p < 0.001).
Correlation was computed between data retrieved in June 2018 for DMAPI and
data retrieved in October 2016 for OD-OSRM. In October traffic in Malta is
much heavier than in June because schools start in this period. It is important
to note that during summer, government employees work half days and schools
are closed.

4.3 Traffic flow count
The ground truth to evaluate traffic flow count experimentation came from work
done by Nigel Pace in his dissertation submitted in 2017 [15]. Directional traffic
flow counts were manually gathered from web camera streams recorded from four
locations. These were gathered from Kappara and Marsa roadways for traffic
which was both northbound and southbound. The Marsa roadway is referred to
Vehicular traffic flow intensity detection and prediction 9

as the Marsa-Hamrun bypass, which is the road leading to and from the Santa
Venera tunnels. The Kappara roadways get and feed traffic to the old Kappara
roundabout which today has been replaced by a flyover. The dates for the data
collection were from Monday 17th October to Friday 21st October 2016. Data for
the day of Tuesday 18th October was missing from the dataset. The traffic flow
count consisted of an average traffic flow count per minute taken over intervals
of 15 minutes. This resulted into 11 samples per location for data gathered from
6.00 a.m to 8.45 a.m. for every day. A daily average for every quarter of an hour
was then taken for both the actual data and the one generated with OD-OSRM.

Table 1. Correlation statistics for linear regression models. OD-OSRM traffic flow
count is the predictor variable and video stream traffic count is the dependent variable.
∗
Results are all significant with p < 0.05.

A regression was established between OD-OSRM traffic count and video
stream traffic counts and results are shown in table 1. There is strong corre-
lation with Kappara traffic flows6 but a weak negative one with Marsa7 located
traffic flows. This can be attributed to the fact that traffic tends to be slower in
Marsa traffic flow points when compared with the Kappara traffic flow points.
OD-OSRM measurements were based on trips that had been detected but if
actual vehicular traffic slows down due to congestion the OD-OSRM traffic flow
count does not reflect actual traffic counts. Therefore, two conclusions are de-
rived from this. The first conclusion is that reliable regression models can be
trained on actual traffic data for traffic flow road sections which do not experi-
ence heavy traffic slow down. Secondly the regression model mapped traffic flow
counts gave a reliable account of what flow capacity is ’expected’ to be serviced
at any given point in time from a given road section in order that traffic flows
smoothly.

4.4 Traffic flow count prediction for a selection of locations
Table 2 shows the evaluation results obtained for traffic flow count prediction.
It is evident that models trained to predict for smaller time ahead intervals
generally perform better than models that are trained with a lengthier prediction
6
geocordinates: 35.904416, 14.487168
7
geocordinates: 35.898072, 14.486804
10 Maurice Saliba et al.

time interval for the same location. Performance of prediction of four levels of
traffic was done with level one being low or no traffic and level four having the
highest level of traffic. The levels were mapped with a logarithmic function as a
ratio to the highest level of traffic. All models proved to have highest recall and
precision for class one traffic flow counts.

Table 2. Classification evaluation metrics for 4 traffic flow road sections with 4 label
classification and PCA set to extract 324 first components. Testing was done with 4
sizes of prediction time window ahead for each prediction location.

It appears from the table that the best overall classification metric scores were
attained for Hamrun-Valletta roadway. However on examination of a confusion
matrix for classification results per label we noted that the model performed
very badly for high traffic flow count classes. There were no results for class four
and for classes two and three the precision and recall metrics are very low. In
fact, when computing the F1-score for classes two and three, both result were
found to be low at 0.14 and 0.0 respectively. Lv et al stated that ANN models
trained with low traffic counts do not perform well. In this same work evaluation
relative error is greater when traffic flow is small. Results are only being quoted
when traffic flow measurement amounts to 450 vehicles or more for a 15 minute
time window [13].
In contrast predictive results for Marsa road that leads to Aldo Moro street
are less promising than those for Hamrun-Valletta arterial road. Still, the predic-
tive efficacy results are very good, especially when examined in the perspective
Vehicular traffic flow intensity detection and prediction 11

of confusion matrices. Class four cases, which are classified as class one or class
two cases are very few and, even if almost half of class four test values were
predicted as class three, in practice, this would still make the model useful.

5 Conclusions and future work

This research posed questions on whether it is feasible to get vehicular traffic
descriptive and predictive analytics from mobile usage data. We showed how
from top users’ activity locations it is also possible to achieve accurate results in
getting global trip counts and trip delays. Trip data was then used to actually
map traffic flow demand on the road grid. However, it was found that traffic flow
mapping gave more accurate results when the level of traffic congestion was low.
Finally, an MLPC was found to be really efficient in predicting traffic flow for
a set of locations. The confidence level given by the prediction results is high and
if traffic flow input used to train the predictive model is accurate the method we
devised could be used for practical scenarios to forecast traffic in real-time.

References

1. Ahas, R., Tiru, M., Saluveer, E., Demunter, C.: Mobile telephones and mobile
positioning data as source for statistics : Estonian experiences. Presentation for
NTTS (2011)
2. Alexander, L., Jiang, S., Murga, M., González, M.C.: Origin-
destination trips by purpose and time of day inferred from mo-
bile phone data. Transportation Research Part C: Emerging Tech-
nologies 58, 240–250 (2015). https://doi.org/10.1016/j.trc.2015.02.018,
http://dx.doi.org/10.1016/j.trc.2015.02.018
3. Calabrese, C., Giusy, F., Lorenzo, D., Liu, L., Ratti, C., Calabrese, F., Lorenzo,
G.D.: Estimating Origin-Destination flows using opportunistically collected mo-
bile phone location data from one million users in Boston Metropolitan Area
Terms of Use Estimating Origin-Destination flows using opportunistically col-
lected mobile phone location da. IEEE Pervasive Computing 10(4), 36–44 (2011).
https://doi.org/10.1109/mprv.2011.41
4. Calabrese, F., Diao, M., Di Lorenzo, G., Ferreira, J., Ratti, C.: Un-
derstanding individual mobility patterns from urban sensing data: A mo-
bile phone trace example. Transportation Research Part C: Emerging
Technologies 26, 301–313 (2013). https://doi.org/10.1016/j.trc.2012.09.009,
http://dx.doi.org/10.1016/j.trc.2012.09.009
5. Chakraborty NKNagwani Lopamudra Dey, S.: Performance Comparison of Incre-
mental K-means and Incremental DBSCAN Algorithms. International Journal of
Computer Applications 27(11), 975–8887 (2011)
6. Çolak, S., Alexander, L.P., Alvim, B.G., Mehndiratta, S.R., González,
M.C.: Analyzing Cell Phone Location Data for Urban Travel. Trans-
portation Research Record: Journal of the Transportation Re-
search Board 2526, 126–135 (2015). https://doi.org/10.3141/2526-14,
http://trrjournalonline.trb.org/doi/10.3141/2526-14
12 Maurice Saliba et al.

7. Eldawy, A., Mokbel, M.F.: Spatialhadoop: A mapreduce framework for spatial
data. In: Data Engineering (ICDE), 2015 IEEE 31st International Conference on.
pp. 1352–1363. IEEE (2015)
8. Gonzalez, M.C., Hidalgo, C.A., Barabasi, A.L.: Understanding indi-
vidual human mobility patterns. Nature 453(7196), 779–782 (2008).
https://doi.org/10.1038/nature06958
9. Hoteit, S., Chen, G., Viana, A., Fiore, M.: Filling the gaps: On the completion
of sparse call detail records for mobility analysis. In: Proceedings of the Eleventh
ACM Workshop on Challenged Networks. pp. 45–50. ACM (2016)
10. Hoteit, S., Secci, S., Sobolevsky, S., Ratti, C., Pujolle, G.: Estimating
human trajectories and hotspots through mobile phone data. Computer
Networks 64, 296–307 (2014). https://doi.org/10.1016/j.comnet.2014.02.011,
http://dx.doi.org/10.1016/j.comnet.2014.02.011
11. Huang, F., Zhu, Q., Zhou, J., Tao, J., Zhou, X., Jin, D., Tan, X., Wang, L.: Research
on the parallelization of the dbscan clustering algorithm for spatial data mining
based on the spark platform. Remote Sensing 9(12), 1301 (2017)
12. Iqbal, M.S., Choudhury, C.F., Wang, P., González, M.C.: De-
velopment of origin-destination matrices using mobile phone
call data. Transportation Research Part C: Emerging Technolo-
gies 40, 63–74 (2014). https://doi.org/10.1016/j.trc.2014.01.002,
http://dx.doi.org/10.1016/j.trc.2014.01.002
13. Lv, Y., Duan, Y., Kang, W., Li, Z., Wang, F.Y.: Traffic flow prediction with big
data: a deep learning approach. IEEE Transactions on Intelligent Transportation
Systems 16(2), 865–873 (2015)
14. Malta, T.: National household travel survey 2010. transport malta (2011)
15. Pace, N.: Investigating the Potential of Big Data in the Management of Traffic in
Malta (2017)
16. Ranjan, G., Zang, H., Zhang, Z.L., Bolot, J.: Are call detail records biased for
sampling human mobility? ACM SIGMOBILE Mobile Computing and Communi-
cations Review 16(3), 33–44 (2012)
17. Steenbruggen, J., Tranos, E., Nijkamp, P.: Data from mobile phone
operators: A tool for smarter cities? Telecommunications Policy
39(3-4), 335–346 (2015). https://doi.org/10.1016/j.telpol.2014.04.001,
http://dx.doi.org/10.1016/j.telpol.2014.04.001
18. Toole, J.L., Colak, S., Sturt, B., Alexander, L.P., Evsukoff, A., González,
M.C.: The path most traveled: Travel demand estimation using big
data resources. Transportation Research Part C: Emerging Tech-
nologies 58, 162–177 (2015). https://doi.org/10.1016/j.trc.2015.04.022,
http://dx.doi.org/10.1016/j.trc.2015.04.022
19. Vlahogianni, E.I., Karlaftis, M.G., Golias, J.C.: Short-term traf-
fic forecasting: Where we are and where were going. Trans-
portation Research Part C: Emerging Technologies 43, 3–19
(2014). https://doi.org/https://doi.org/10.1016/j.trc.2014.01.005,
http://www.sciencedirect.com/science/article/pii/S0968090X14000096
20. Wang, M.h., Schrock, S.D.: Feasibility of Using Cellular Telephone Data to Deter-
mine the Truckshed of Intermodal Facilities. Tech. rep., University of Nebraska -
Lincoln (2012)