Effects of the Coronavirus Pandemic on Youth Mobility: A Case

Study Analysis through Floating Car Data

Simone Porru

, Francesco Edoardo Misso

, Silvia Manca and Cino Repetto

T Bridge S.p.A., Via Garibaldi 7/10, Genova, Italy

Keywords: Floating Car Data, GPS Data Analysis, Youth Mobility.

Abstract: Among mobility data sources, Floating Car Data plays a very, and increasingly, significant role, and has been

extensively used to obtain traffic information. In this study, FCD has been used to shed light upon the youth

mobility changes occurred during the first two years of the coronavirus pandemic by focusing on five selected

high schools in Modena Municipality (Italy). Mobility indicators computed within the areas under

investigation show that two out of the three schools’ areas are not associated to a significant variation in the

number of detected distinct private probe vehicles from November 2019 to November 2021 (-3% and +1%),

whereas within the Modena Municipality results show a 8% decrease. However, one out of the three schools’

areas shows a significant decrease in 2021 when compared to 2019 (-11%), suggesting a noticeable decrease

in private vehicles traffic density that could be due to an increased use of personal mobility vehicles, such as

bikes. Moreover, results within the Modena Municipality suggest that in 2021, even if the number of detected

vehicles was lower than in 2019, each vehicle not only covered a longer distance on average, but also the total

distance covered by all the vehicles together was longer (14% increase).

1 INTRODUCTION

Floating Car Data (FCD) refers to data collected

directly by a vehicle as it is in motion, typically

covering its location and speed (ECOSOC, 2021;

Zannat, 2019). FCD has been used for a variety of

purposes, including estimating the level of service on

traffic networks (Dailey, 2002), investigating traffic

safety (Axer, 2013; Biral, 2021; Guido, 2012; Kerner,

2005; Vaiana, 2014), and regulating traffic signals

(Astarita, 2017). Connected vehicles providing FCD

are to be considered as moving sensors, and as such

they are extremely different from other data sources

traditionally used for traffic prediction such as fixed

traffic detectors (e.g., induction loops and traffic

cameras) (Altintasi, 2017). The latter, contrary to

FCD data sources, provide information about the total

traffic, even if only relevant to the road section where

the detectors are deployed. In general, however, it is

yet unclear how much FCD is necessary for traffic

estimations and predictions, considering that not all

the vehicles can be equipped with FCD devices such

as GPS, smartphones, V2X onboard units, and also

https://orcid.org/ 0000-0002-8448-9282

https://orcid.org/ 0000-0002-6660-3360

considering that these devices may not always

actively transmit traffic data (Mena-Oreja, 2021).

Apart from traffic density, by leveraging the

information provided by connected vehicles via FCD

it is possible to calculate a range of mobility

indicators relevant to selected areas and time frames.

This study, based on work undertaken for the project

YOUMOBIL (T Bridge, 2022)—a project aimed at

enhancing the passenger transport system for young

people that live in rural areas and at improving their

access to the European and national transport

networks—attempts to analyse FCD to shed light

upon the youth mobility changes occurred during the

first two years of the coronavirus pandemic. In

particular, since areas and time frames that are mostly

related to youth activities—such as schools’

neighbouring areas around start times—may be

further investigated to find potential clues on how

youth mobility has changed between 2019 and 2021,

the FCD analysis reported in this article attempts to

evaluate the youth mobility changes during the first

two years of the coronavirus pandemic by focusing

on five selected high schools in the Modena

144

Porru, S., Misso, F., Manca, S. and Repetto, C.

Effects of the Coronavirus Pandemic on Youth Mobility: A Case Study Analysis through Floating Car Data.

DOI: 10.5220/0011318900003280

In Proceedings of the 19th International Conference on Smart Business Technologies (ICSBT 2022), pages 144-151

ISBN: 978-989-758-587-6; ISSN: 2184-772X

Municipality, located in the territory of Emilia-

Romagna, a region of northern Italy.

To this purpose, we pursue the following research

questions:

RQ1: Does traffic near selected high schools in

2021 significantly differ from 2019? We want to

shed light on how youth mobility changed during the

coronavirus pandemic. Consequently, we decided to

focus on mobility indicators related to areas relevant

to youth education, namely, schools’ neighbouring

areas, in 2019 and 2021.

RQ2: How do the traffic changes near high

schools compare to traffic changes in the whole

Municipality of Modena in 2019 and in 2021? To

better evaluate the traffic changes within areas linked

to youth activities, we should compare them to an

appropriate baseline. As a consequence, we also

evaluated the mobility changes in the whole

Municipality of Modena, where the schools’ areas are

located.

The rest of the paper is organized as follows:

Section 2 presents the methodology, starting from the

territorial context under investigation and the

reference time frames, and then introducing the FCD

dataset and the mobility indicators; Section 3 reports

the results of the analysis, focusing on the Modena

Municipality, the schools’ areas, and the FCD

penetration rate; Section 4 addresses the research

questions and, finally, Section 5 draws the

conclusions.

2 METHODOLOGY

2.1 Territorial Context

The FCD analysis focuses primarily on the traffic

density around five selected high schools in the

Modena Municipality, on the south side of the Po

Valley, in the Province of Modena in the territory of

Emilia-Romagna, a region of northern Italy (Figure

1). The schools, all located to the west of the old city

centre, are the following:

• F. Selmi Institute of Higher Education, located

at Via Leonardo da Vinci 300, in Villaggio

Artigiano district

;

• F. Corni Institute of Higher Education, located

at Via Leonardo da Vinci 300, next to the F.

Selmi institute, in Villaggio Artigiano

district

;

• Wiligelmo High School, located at Viale

Corassori 101, in San Faustino district

;

• G. Guarini Institute of Higher Education. The

school is located at Via Corassori 95, very

close to Wiligelmo High School, in San

Faustino district

;

• Cattaneo-Deledda Professional Institute. The

school is located at Strada degli Schiocchi

110, in San Faustino district, close to G.

Guarini Institute of Higher Education

Figure 1: Location of Modena within Italy (left) and location of the five selected schools (red dots, right) within the

Municipality of Modena (green shaded area, right).

www.istitutoselmi.edu.it

https://www.istitutocorni.edu.it/

https://www.liceowiligelmo.edu.it/

https://www.istitutoguarini.edu.it/

https://www.cattaneodeledda.edu.it/

Effects of the Coronavirus Pandemic on Youth Mobility: A Case Study Analysis through Floating Car Data

145

Since two or more schools located next to each

other share their surroundings, to the purpose of the

mobility analysis the five schools can be grouped in

three main areas (Selmi and Corni, Wiligelmo and

Guarini, and Cattaneo-Deledda) according to the

schools’ proximity to each other. To allow for a better

detection of the impact of the traffic flow generated by

the schools, such areas can be made large enough to

include the main parking areas and roads where

students can get out of a private vehicle after taking a

ride to school at start times, leading to the identification

of the areas shown in Figure 2 and Figure 3.

Table 1 reports the total number of enrolled

students for each area in school year 2021/2022.

Figure 2: The 3 schools’ areas under study (Selmi and

Corni, Wiligelmo and Guarini, Cattaneo-Deledda).

Table 1: Enrolled students by area in school year

2021/2022.

Area District Schools Students

(2021-2022)

Selmi and

Corni

Villaggio

Artigiano

2 3,033

Wiligelmo

and Guarini

San

Faustino

2 1,587

Cattaneo-

Deledda

San

Faustino

1 1,281

2.2 Reference Periods for Comparison

To allow for an effective comparison on mobility

changes between the period before the coronavirus

pandemic and the most recent times, we first

identified a recent time period in which tough

coronavirus restrictions affecting mobility (and

especially youth mobility) were not in place (e.g., no

lockdowns, limited distance learning), so that the

impact of the coronavirus pandemic on mobility

patterns could possibly be observed devoid of the

temporary effects of harsh COVID-19 measures.

To this purpose, we selected November 2021 as a

suitable candidate time frame for the study

(Lombardi, 2021). This choice also led to select

November 2019 as the corresponding counterpart in

the period before the coronavirus pandemic, since the

two time frames correspond to the same period of the

year and consequently allow for a meaningful

comparison between 2019 and 2021.

More specifically, we focused on the periods 9-29

November 2019 and 8-28 November 2021, because

the two time frames:

1. Are the same length (21 days);

2. Cover the same period of the year almost

exactly (1-day difference);

3. Each features exactly 3 occurrences of each

day of the week;

4. Do not include public holidays (1 November),

only regular weekends.

In addition, to specifically address the impact of

schools’ activity on mobility, we also focused on the

hours 7:00-9:00 a.m., since the start time of the

schools under study is 8:00 a.m.

Figure 3: Schools' areas under study and main entrances.

ICSBT 2022 - 19th International Conference on Smart Business Technologies

146

2.3 The FCD Dataset

To the purpose of the study, an appropriate FCD

dataset was queried in order to extract meaningful

information on mobility changes.

Such dataset, acquired from a third-party provider

and constituted by FCD collected by private and

commercial vehicles equipped with an onboard GPS

device, comprises of two tables. The most significant

table is a collection of GPS records, each

corresponding to a specific probe vehicle’s position.

Each record contains a trip identifier, and all the

records with the same trip identifier are part of the

same trip. A trip is defined as a collection of GPS

records between two points where the vehicle

remained stationary for at least 5 minutes. The other

table of the dataset is constituted by records each

containing the start and end GPS position of each

identified trip, with additional information on the

distance covered during the trip as provided by the

vehicle’s odometer.

Each record of the main table of the dataset also

includes additional data other than the GPS position

(latitude and longitude) and the trip identifier; among

these additional data is the vehicle identifier, the

corresponding timestamp, the vehicles’ direction (0-

360), the vehicle type (private/commercial), and other

data related to the vehicle’s position, including the

address and the municipality.

To focus on the Modena case study, the dataset

only comprises the trips which have at least one GPS

record within the Modena Municipality in November

2019 and November 2021.

2.4 Mobility Indicators

To perform the FCD analysis, different queries were

devised to allow for the extraction of meaningful

mobility indicators relevant to the context (e.g.,

number of distinct vehicles, average distance per

trip), both on the whole of the Modena Municipality,

as it constitutes the baseline against which schools’

areas were compared, and the schools’ areas

themselves, as such areas constitute the main subject

of the analysis. In particular, we focused on the

number of distinct probe vehicles identified in each

area, as this is an indicator which might reflect traffic

density.

For each of the areas considered, all the queries

were executed separately for each time frame of

interest—namely, 9-29 November 2019 and 8-28

November 2021—averaging the results per each day

of the week and per the five-day school week

(Monday through Saturday). Since the FCD dataset

allows for a separate analysis of private and

commercial vehicles, we specifically focused on

private vehicles, due to the fact that hardly any

commercial vehicle may be associated to students’

home-to-school trips.

After extracting the data for the selected time

frames in 2019 and 2021, relative deltas were

calculated to compare the difference between the

most meaningful indicators in each area. As a result,

the deltas relevant to the schools’ areas could also be

compared to the deltas relevant to the Modena

Municipality.

Finally, we estimated the FCD penetration rate

both in 2019 and 2021 to evaluate its impact on the

resulting key mobility indicators.

3 RESULTS

3.1 Modena Municipality

To the purpose of focalising on the changes directly

related to schools’ activities, mobility changes around

schools were compared to the changes in the whole of

the Modena Municipality, whose FCD were analysed

specifically to build a baseline for comparison. As

previously stated, the municipality’s FCD were

analysed focusing between 7 and 9 am, as such time

slot is the most relevant to schools-related mobility,

being 8:00 am the start time for lessons for all the 3

selected schools.

Table 2 reports the number of distinct private

probe vehicles detected within the municipality of

Modena between 7 and 9 am, averaged from Monday

to Saturday over the time frames 9-29 November

2019 and 8-28 November 2021, and the resulting

delta between 2019 and 2021.

Table 2: Average daily distinct private probe vehicles

(Monday to Saturday, 7-9 am) within Modena

Municipality.

Area

Daily

average

2019

Daily

average

2021

Delta

2021/2019

Modena

Municipality

685.8 632.9 -8%

Results show that, within the Modena

Municipality, the number of distinct private probe

vehicles detected in 2021 is 8% lower than in 2019.

In addition, we also calculated the variation of the

total distance covered by all the private probe

vehicles in the same time frame, and also the average

Effects of the Coronavirus Pandemic on Youth Mobility: A Case Study Analysis through Floating Car Data

147

distance covered per vehicle, to the purpose of

detecting if such variations reflected that of the

number of distinct vehicles (Table 3).

Table 3: Deltas 2021/2019 of the average of total distance

covered and distance per vehicle relative to private probe

vehicles (Monday to Saturday, 7-9 am) within Modena

Municipality.

Area Total distance

(delta

2021/2019)

Distance per

vehicle (delta

2021/2019)

Modena

Municipalit

+14% +22%

The results show that the total distance travelled

by private probe vehicles in 2021 increased by 14%

on an average day between Monday to Saturday

between 7 and 9 am, and that the average travelled

distance per vehicle increased by 22%. Thus, in 2021

were detected 8% fewer private probe vehicles than

in 2019, but the total distance travelled by the

detected vehicles increased by 14%.

3.2 Schools’ Areas

The FCD detected within the 3 identified schools’

areas were analysed focusing between 7 and 9 am, as

such time slot is the most relevant to schools-related

mobility, being 8:00 am the start time for lessons for

all the 3 selected schools.

Table 4 reports the number of distinct private

probe vehicles detected within each of the 3 schools’

areas between 7 and 9 am, averaged from Monday to

Saturday over the time frames 9-29 November 2019

and 8-28 November 2021, and the resulting delta

between 2019 and 2021.

Table 4: Average daily distinct private probe vehicles

(Monday to Saturday, 7-9 am) within the schools’ areas.

Area Daily

average

2019

Daily

average

2021

Delta

2021/2019

Selmi and

Corni

10.3 9.2 -11%

Wiligelmo

and Guarini

19.7 19.0 -3%

Cattaneo-

Deledda

8.6 8.7 +1%

Results show that in 1 of the 3 schools’ areas the

number of distinct private probe vehicles detected in

2021 is around 1% higher than in 2019, whereas in

the other 2 schools’ areas the number decreased.

Within the area where the number decreased, only

that of Selmi and Corni (-11%) decreased more (and

only slightly so) than that of the whole Modena

Municipality (-8%), whereas in the Wiligelmo and

Guarini’s area the extent of the decrease is

considerably less noticeable (-3%) when compared to

the whole Modena Municipality (-8%).

3.3 FCD Penetration Rate

As previously mentioned, the FCD analysed in this

study might be used as a basis to estimate traffic

within the case study areas, which means we have to

first determine how many vehicles the employed FCD

actually represent. To this purpose, assuming that the

probe vehicles are uniformly distributed across the

network—even if this is not realistic in many cases—

we calculated the ratio of the probe vehicles (FCD

penetration rate), as suggested by Nagle and Gayah

(Nagle, 2014). More specifically, Nagle and Gayah

suggested to use such ratio to estimate network-wide

variables from FCD, also proposing to acquire the

ratio by dividing the number of vehicles tracked by

GPS in the analysis area for a specific time period to

the number of vehicles that crossed the fixed traffic

detectors in the same area and period. Traffic density

can be then estimated by multiplying the number of

vehicles detected with FCD in a specific area by the

reciprocal of the penetration rate of the FCD devices.

Although the FCD penetration rate can vary in

space and time, it must be noted that the mobility

indicators calculated in the previous sections were

averaged over a period of three weeks both in 2019

and 2021, thus limiting the impact of temporary

variations in the FCD penetration rate and, especially

regarding the indicators calculated for the Modena

Municipality, rate variation in space were smoothed

out by the considerable extension of the area of the

analysis.

Given the previous considerations, we estimated

the 2019 and 2021 FCD penetration rate related to the

area under study on the basis of the number of

vehicles that crossed the fixed traffic detector with

identification number 328, located around 3.5 km

south-west of the 5 schools under investigation on

SP486 road (Figure 4), on the peak day of November

2019 and November 2021 (13/11/2019 and

11/11/2021) .

To determine the FCD penetration rate we

identified all the trips that crossed the fixed traffic

detector on the monitored road section by analysing

on map the private probe vehicles’ GPS positions in

chronological order. We then divided the number of

identified trips by the traffic detected by the fixed

traffic detector, thus estimating the FCD penetration

rate both in 2019 and 2021 (Table 5).

ICSBT 2022 - 19th International Conference on Smart Business Technologies

148

Figure 4: Location of the fixed traffic detectors on road

SP486 with respect to the schools' area.

Table 5: Estimation of the 2019 and 2021 FCD penetration

rate for private vehicles on the monitored SP486 road

section.

Year Detected

traffic

FCD

traffic

FCD penetration rate

2019 26,294 195 0.85%

2021 25,389 184 0.82%

The difference between the estimated FCD

penetration rates in 2019 and 2021 does not appear to

be significant. This suggests that the deltas calculated

in the previous sections between the number of

detected private probe vehicles in the 3 schools’ areas

and the Modena Municipality in 2019 and 2021 could

effectively reflect the deltas between the real traffic

densities in the same areas, even if it is not possible

to draw such a conclusion from the sole comparison

of the estimates in one road section.

4 DISCUSSION

In this section, we discuss the results of the FCD

analysis and investigate the possible reasons behind

the difference between the mobility indicators

calculated within the Modena Municipality and the 3

selected schools’ areas, thus trying to answer the

research questions introduced at the beginning of this

chapter. We first address RQ1, and then RQ2.

RQ1. Does traffic near selected high schools in

2021 significantly differ from 2019?

As reported in Table 6, two out of the three

schools’ areas under investigation, namely Wiligelmo

and Guarini’s and Cattaneo-Deledda’s areas, do not

show a significant variation in the number of detected

distinct private probe vehicles from November 2019

to November 2021 (-3% and +1% respectively). This

would suggest that the traffic density around schools

in November 2021 grew back to the traffic density

characterizing the period before the COVID-19

pandemic. However, one out of the three schools’

areas, namely Selmi and Corni’s area, shows a

significant decrease in the number of detected private

probe vehicles in 2021 when compared to 2019 (-

11%), suggesting a noticeable decrease in private

vehicles traffic density within such area. As a

consequence, these results are further investigated

and discussed at the end of this section, so as to first

provide additional insights on the local context

through addressing the second research question.

Table 6: Average daily distinct private probe vehicles

(Monday to Saturday, 7-9 am) within the 3 selected

schools’ areas and Modena Municipality.

Area Daily

average

2019

Daily

average

2021

Delta

2021/2019

Wiligelmo

and Guarini

19.7 19.0 -3%

Cattaneo-

Deledda

8.6 8.7 +1%

Selmi and

Corni

10.3 9.2 -11%

Modena

Munici

alit

685.8 632.9 -8%

RQ2. How do the traffic changes near high

schools compare to traffic changes in the whole

Municipality of Modena in 2019 and in 2021?

As reported on Table 6, in the Municipality of

Modena the decrease in the number of detected

distinct private probe vehicles from November 2019

to November 2021 is around 8% and, therefore,

noticeably different from the results obtained within

Wiligelmo and Guarini’s and Cattaneo-Deledda’s

areas. Nevertheless, as previously mentioned, the

results from Selmi and Corni’s areas differ

considerably from those obtained around the other

schools’ areas, suggesting an even higher decrease in

traffic density than that detected in the whole

Municipality of Modena. This prompted a more in-

depth investigation into the local context to find

possible reasons that could explain such difference,

as Selmi and Corni’s 11% decrease in distinct private

(probe) vehicles could actually be due to many

factors, or a combination of them. Among the various

possible reasons, one could be that of an increased use

of personal mobility vehicles, such as e-scooters and

bikes/e-bikes, also considering that a nearby brand-

new bicycle route was inaugurated at the beginning

of October 2021—just one month before the time

frame investigated in 2021— which effectively

Effects of the Coronavirus Pandemic on Youth Mobility: A Case Study Analysis through Floating Car Data

149

connect Selmi and Corni schools to the Central

Station in a very convenient way. Such infrastructure

could have led many citizens that regularly move to

or from the areas crossed by the cycling route—thus

including the Selmi and Corni’s students—to rely on

personal mobility vehicles instead of private ICE

vehicles.

With regard to the RQ2, we must also highlight

the results concerning the total distance travelled by

private probe vehicles in 2019 and 2021 within the

Municipality of Modena (Table 7).

Table 7: Deltas 2021/2019 of the average of total distance

travelled and distance per vehicle relative to private probe

vehicles (Monday to Saturday, 7-9 am) within Modena

Municipality.

Area Total distance

(delta

2021/2019)

Distance per

vehicle (delta

2021/2019)

Modena

Municipalit

+14% +22%

As previously reported, the total distance

travelled by the detected vehicles increased by 14%,

with a 22% increase in the average travelled distance

per vehicle, but in 2021 8% fewer private probe

vehicles were detected than in 2019. These results

suggest that in 2021, even if the number of private

vehicles was lower than in 2019, not only each of

these vehicles covered a longer distance, but they all

together also travelled a longer distance (14%

variation). This might have led to a higher traffic

density within the Municipality of Modena on the

whole.

We must also remark that hypothesis such as the

one reported previously are based on the results from

the FCD Analysis, and thus should be further verified

with real traffic density data since, as previously

stated, the FCD penetration rate relative to the

analysed dataset is slightly lower than 1% and could

also vary considerably in space and time.

Nevertheless, the FCD analysis provides interesting

information which can be leveraged to further

investigate the local context and the related changes

in traffic density.

5 CONCLUSIONS

Floating Car Data refers to data collected directly by

a vehicle, typically covering its location and speed

(ECOSOC, 2021; Zannat, 2019). FCD has been used,

among others, to estimate the level of service on

traffic networks (Dailey, 2002), to investigate traffic

safety (Axer, 2013; Biral, 2021; Guido, 2012; Kerner,

2005; Vaiana, 2014), and to regulate traffic signals

(Astarita, 2017).

In this study, FCD has been used to attempt to

investigate the effects of the coronavirus pandemic on

mobility within the areas around five selected high

schools in the Italian Municipality of Modena and the

wider Municipality area. The mobility indicators

computed through the FCD analysis show that, from

November 2019 to November 2021, two out of the

three schools’ areas under investigation are not

associated to a significant variation in the number of

detected distinct private probe vehicles (-3% and

+1%), while in the Modena Municipality results show

a 8% decrease. Nevertheless, one out of the three

schools’ areas shows a significant decrease in 2021

(-11%), suggesting a noticeable decrease in private

vehicles traffic density which can be attributed to an

increased use of personal mobility vehicles, such as

e-scooters and bikes/e-bikes, also considering the

recent inauguration of a nearby brand-new bicycle at

the beginning of October 2021.

In addition, the number of detected vehicles in the

Municipality of Modena was lower in 2021 than in

2019, but the average distance covered by each

vehicle was longer (22% increase), as well as the total

distance covered by all the vehicles together (14%

increase). Even if further validation with real traffic

density data is needed to draw more detailed

conclusions, FCD proved to be able to provide

interesting information which can be leveraged to

investigate the mobility changes within specific

regions.

ACKNOWLEDGEMENTS

This article is based on work undertaken for the

project YOUMOBIL (CE1307), co-funded by the

European Regional Development Fund under

Interreg CENTRAL EUROPE.

REFERENCES

Altintasi, O., Tuydes-Yaman, H., & Tuncay, K. (2017).

Detection of urban traffic patterns from Floating Car

Data (FCD). Transportation research procedia, 22, 382-

391.

Astarita, V., Giofrè, V. P., Guido, G., & Vitale, A. (2017).

The use of adaptive traffic signal systems based on

floating car data. Wireless Communications and Mobile

Computing, 2017.

ICSBT 2022 - 19th International Conference on Smart Business Technologies

150

Axer, S., & Friedrich, B. (2013). Optimization of traffic

safety on rural roads by traffic data based strategies.

Proceedings of 13thWCTR.

Biral, F., Da Lio, M., & Bertolazzi, E. (2005). Combining

safety margins and user preferences into a driving

criterion for optimal control-based computation of

reference manoeuvres for an ADAS of the next

generation. In IEEE Proceedings. Intelligent Vehicles

Symposium, 2005. (pp. 36-41). IEEE.

Dailey, D. J., & Cathey, F. W. (2002). Virtual speed sensors

using transit vehicles as traffic probes. In Proceedings.

The IEEE 5th International Conference on Intelligent

Transportation Systems (pp. 560-565). IEEE.

Guido, G., Vitale, A., Astarita, V., Saccomanno, F., Giofré,

V. P., & Gallelli, V. (2012). Estimation of safety

performance measures from smartphone sensors.

Procedia-Social and Behavioral Sciences, 54, 1095-

1103.

Kerner, B. S., Demir, C., Herrtwich, R. G., Klenov, S. L.,

Rehborn, H., Aleksic, M., & Haug, A. (2005). Traffic

state detection with floating car data in road networks.

In Proceedings. 2005 IEEE Intelligent Transportation

Systems, 2005. (pp. 44-49). IEEE.

Lombardi, F. (2021). Covid, scuola, dad, quarantena: nuovo

protocollo. Classi a casa solo se ci sono 3 positivi. Il

Giorno. https://www.ilgiorno.it/cronaca/scuola-nuovo-

protocollo-covid-dad-quarantene-1.6985457" \l

"Elementari,%20medie%20e%20superiori:%20regole

%20diverse

Mena-Oreja, J., & Gozalvez, J. (2021). On the Impact of

Floating Car Data and Data Fusion on the Prediction of

the Traffic Density, Flow and Speed Using an Error

Recurrent Convolutional Neural Network. IEEE

Access, 9, 133710-133724.

Nagle, A. S., & Gayah, V. V. (2014). Accuracy of

networkwide traffic states estimated from mobile probe

data. Transportation Research Record, 2421(1), 1-11.

T Bridge S.p.A. (2022). Guidelines for the PT Decision

Makers on Smart Rural Transport Tools. Output o

project YOUMOBIL (CE1307), Activity T1.6

“Development of a toolbox for youth-oriented smart

rural transport solutions”.

United Nations Economic and Social Council (ECOSOC)

(2010). Monitoring the transport-related Sustainable

Development Goal indicators in the Economic

Commission for Europe region. (WP.6) Working Party

on Transport Statistics (72nd session).

Vaiana, R., Iuele, T., Astarita, V., Caruso, M. V., Tassitani,

A., Zaffino, C., & Giofrè, V. P. (2014). Driving

behavior and traffic safety: an acceleration-based safety

evaluation procedure for smartphones. Modern Applied

Science, 8(1), 88.

Zannat, K. E., & Choudhury, C. F. (2019). Emerging big

data sources for public transport planning: A systematic

review on current state of art and future research

directions. Journal of the Indian Institute of Science,

99(4), 601-619.

Effects of the Coronavirus Pandemic on Youth Mobility: A Case Study Analysis through Floating Car Data

151