Optimizing Transport Network to Reduce Municipality Mobility

Budget

Irina Arhipova

, Nikolajs Bumanis

, Liga Paura

, Gundars Berzins

, Aldis Erglis

Gatis Vitols

Evija Ansonska

, Vladimirs Salajevs

and Juris Binde

Faculty of Information Technologies, Latvia University of Life Sciences and Technologies, Liela street 2, Jelgava,

LV 3001, Latvia

Faculty of Business, Management and Economics, University of Latvia, Aspazijas bulv. 5, Riga, LV 1050, Latvia

Latvian Mobile Telephone, Ropazu street 6, Riga, LV 1039, Latvia

{gundars.berzins, aldis.erglis, evija.ansonska}@lu.lv, juris.binde@lmt.lv

Keywords: Transportation Network Optimisation, Mobility Budget, Public Transportation, Mobile Activity.

Abstract: Mobility budgets dictate the limit of CO

per capita, which is calculated based on the mode of travel and

distance. Mobility budgets are one of the final goals of the optimisation of transportation network, when the

aspects of fairness and equity are considered. The main problem arises when we focus on multiple criteria of

fairness and equity. In addition, it was observed that any drastic change in behaviour leads to inadequate initial

parametrisation, especially under the effects of COVID-19. This can also mean that optimising transportation

network according to class-to-be is most likely to cause behaviour changes in relation to the use of public

transport. The aim of this article is to define the structure of optimisation task, based on mobility budget

provided on a monthly basis. This research was based on public transportation data and mobile activity data.

The former was used to determine the usage of public transport during 2017 and 2022, while the latter

provided enough information to determine exactly how COVID-19 affected the behaviour of city districts and

provide concrete information regarding necessary re-planning measures for public transportation station

locations. In result, the optimisation solution was proposed by defining case-specific objective functions and

constraints.

1 INTRODUCTION

In the contrast of advantages brought by economic

growth, the increase in overall Greenhouse Gas

emissions (GHGE) (International Energy Agency,

2022) raises an issue of balancing the use of

technological operations versus their environmental

cost. It includes almost all fields – from industrial

operations like food production (Boke Olén et al.,

2021) and waste utilisation (Yasmin et al., 2022) to

personal carbon footprint (Khanam et al., 2022).

In order to address GHGE, European Commission

of European Union (EU) adopted a series of proposals

https://orcid.org/0000-0003-1036-2024

https://orcid.org/0000-0002-6625-9475

https://orcid.org/0000-0002-1302-527X

https://orcid.org/0000-0002-4131-8635

https://orcid.org/0000-0002-7029-2745

https://orcid.org/0000-0001-8545-4690

with the main goal of achieving climate-resilient

society (European Comission, 2020). The results of

developing technologies and strategies aligning with

these proposals can already be seen with reduction of

overall GHGE in EU from 2019 to 2020 by 0.37 Gt

(European Environment Agency, 2021). While it

seems positive, during these years the COVID-19

pandemic was most active. As a result, based on

global CO

emissions (International Energy Agency,

2022) the reduction in 2020 may be a pandemic’s

post-effect dip that returned even higher in 2021.

It was found that the COVID-19 pandemic did

indeed have negative impact on CO

emissions

Arhipova, I., Bumanis, N., Paura, L., Berzins, G., Erglis, A., Vitols, G., Ansonska, E., Salajevs, V. and Binde, J.

Optimizing Transport Network to Reduce Municipality Mobility Budget.

DOI: 10.5220/0011941500003494

In Proceedings of the 5th International Conference on Finance, Economics, Management and IT Business (FEMIB 2023), pages 38-47

ISBN: 978-989-758-646-0; ISSN: 2184-5891

 2023 by SCITEPRESS – Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)

(Jawadi et al., 2022; Kareinen et al., 2022; Nicolini et

al., 2022). This relates, for example, to the

enforcement of homestay, including remote work,

and reduction of leisure group sizes at early stages of

the pandemic until now when improved protection

measures are in place. The adoption of these

principles, including people wish for less exposure to

public places, led to population behaviour changes –

more people switched from public transport (PT) to

personal transport for commute (Delbosc et al.,

2022), as the latter proved to be a safer and less

restrictive mode of travel during COVID-19 (Zafri et

al., 2022). This, in turn, resulted in an increase of

personal carbon footprint in mobility, and the term

“rebound effect” was introduced (Rojas et al., 2022).

As Jawadi et al. (Jawadi et al., 2022) point out, short-

term strategies have no significant contributing effect

and sustainable long-term solutions should be

developed instead. This includes changing the way of

perception of daily carbon footprint (Hoffmann et al.,

2022).

Carbon footprint can be related to multiple

activities and fields, while most prominent in the

effect on CO

during and after COVID-19 was

transportation (Li et al., 2021; Vélez, 2022). There

can be various approaches to manage carbon

footprint. In general, carbon footprint management is

implemented by applying carbon budgets to regions,

countries and industries. Although the debates on

fairness of allocation of these budgets is still ongoing

(Pan et al., 2022; Williges et al., 2022), the common

consensus is that equity must be supported (Hänsel et

al., 2022). In this instance, equity refers to the relative

distribution of benefits and costs. Therefore, while it

is possible to manage carbon footprint on national and

industry levels by applying government regulations to

stakeholders (Bocken & Allwood, 2012), the

individual carbon footprint management cannot be

forced (Khanam et al., 2022).

Individual carbon footprint includes mostly

choices of food, both its production and initialization,

technologies and modes of mobility. Researchers

suggest that the mode of mobility leaves the largest

impact on the carbon footprint per capita (Bhoyar et

al., 2014; Li et al., 2021). Application of mobility

budgets offers a more specific approach to travel-

related carbon footprint management. Following the

main aspect of localised carbon footprint

management, the mobility budgets are calculated and

specified according to region, objective and target

groups.

There are various ways to approach specification

of mobility budgets. For example, in study of carbon

footprint in college mobility the basis was put on

survey data and existing local policies in order define

multimodal logit model that shaped optimal

combination of mobility modes (Crotti et al., 2022).

Another study (Sánchez-Barroso et al., 2022) was

using surveys to acquire data on mobility modes of

people traveling to healthcare centres. This data was

then used to estimate GHG emissions and it was

found that managing the mobility modes does lead to

reduction of GHG emissions.

In essence, mobility budgets dictate the limit of

per capita on daily/weekly/monthly basis. The

level of CO

is thus calculated based on the mode of

travel and distance. Currently, there are no

standardized characteristics of an individual mobility

budget, even though multiple attempts have been

made to define them. Focusing on multiple aspects of

fairness and equity raises the main challenges. These

aspects include Millonig et al. (Millonig et al., 2022)

hypothesised variants to characterise mobility

budgets based on average ceiling values such as daily

distance travelled by a particular mobility mode and

specified mobility budget for Austria.

The conclusion suggests that even though it is

possible to put limits on mobility budgets, it is strictly

region-based and requires in-depth analysis of the

given region’s socio-geographic needs which makes

one-size-fits all approach impossible even in EU

Mobility budgets are not always the final goal of

transportation network optimization, however, many

such tasks are based on criteria of fairness and equity

(Caggiani et al., 2017; Lucas et al., 2019). In addition,

the fairness in terms of transportation policies is not

yet clearly defined (Randal et al., 2020).

The problem of designing an optimised

transportation system is not novel, but it has attracted

more attention in recent years (Caggiani et al., 2017).

Farahani et al. (Farahani et al., 2013) provided a

sophisticated review on transportation network

design problems almost a decade ago and concluded

that a lot of work will be required in the future to

address questions such as objective functions and

constraints, correct decision-making strategies and

multi-modality of networks. Later, an overview of

multilayer network designs was provided (Crainic et

al., 2022) by applying own classification

methodology with the focus on taxonomy. There are

various approaches to optimising a transportation

system.

Gu et al. (Gu et al., 2018) used simulation-based

optimization to propose toll pricings that would result

in maximised network density. The last decade also

showed a switch of research and application focus

towards multi-objective optimization (MOOP) in

Optimizing Transport Network to Reduce Municipality Mobility Budget

various fields (Al-Ashhab, 2022; Ogumerem et al.,

2018; Sun et al., 2014).

Zhang et al. (Zhang et al., 2022) proposed a

network design based on hidden Markov model and

an Equilibrium Optimizer to solve a MOOP. Authors

used information about passenger paths, amount and

duration of trips to optimize public transportation

aspects – minimising costs and maximising the

number of trips. The work shows good results

compared to algorithm such as Floyd-Warshall.

Wang et al. (Wang et al., 2020) proposed a model

based on Non-dominated Sorting Genetic Algorithm

II to solve MOOP of customizing bus routes in real-

time. The model aimed at achieving two goals – to

minimise travel time and minimise total operating

costs. Authors used information about the travel time,

including boarding times and total travel time, and the

conditions of bus stations such as station’s

accessibility, total demand and the availability of

parking lots. The authors used demand data obtained

from local citizens. It means that particular objective

functions and constraints may vary depending on

transportation network and, therefore, they must be

defined as case-specific.

MOOP involves multiple objective functions that

are to be minimised or maximised. Compared to

single-objective optimisation problem, the best

solution cannot be determined by comparing values

of objective functions; instead, dominance must be

used as a goodness value. For example, an angle-

based constrained dominance principle (Fan et al.,

2019). The answer of MOOP is a set of solutions that

define the best trade-off between competing

objectives subjected to constraints. These constraints

are often defined by case-specific boundaries.

Considering the mobility budget as a goal for

MOOP, the objective functions should address

fairness and equity principles of travel – accessibility,

mobility, variety of types of transport, social equity,

while constraints should address both mobility

aspects and GHGE in general. The result of this

MOOP is a mobility budget provided on a monthly

basis.

Therefore, the aim of this article is to define the

structure of optimization task address fairness and

equity principles of travel, based on mobility budget

provided on a monthly basis.

The remainder of this paper is organised as

follows. The next section describes procedures for

data gathering and analysis. Afterwards, the results of

public bus service data and mobile network data

analysis are presented. Finally, the aspect of mobility

fairness is chosen and objective functions criteria for

the optimisation task are selected.

2 METHODOLOGY

As it was already mentioned, COVID-19 had a

negative impact on overall CO

emissions. In order to

determine particular MOOP parameters, the data

gathered was analysed in consideration of the said

impact. The choice of MOOP objective functions and

constraints was a prerequisite for determining

parameters of PT and selecting those affecting

mobility budgets.

Four types of data were considered: general data,

PT data, monitoring data and mobile activity data.

Aspects of transportation data can be defined as

follows:

 A trip is any movement (>100m) originating

from a location where the resident has an

activity to a destination where the resident has

another activity;

 Destination may be any public place;

 The trip is over when the destination is reached;

 The trip may be followed by another trip

(movement) to another destination.

2.1 General and Public Transportation

Data

This data includes the following parameters for all

types of transport:

 Average trip distance (km) per resident per day;

 Average trip time (min) per day;

 Average and maximum moving speed (km/h);

 Total number and types of roads available for

PT, private vehicles and trucks per day;

 Total number and types of vehicles in the city

per day.

Average trip distance and trip duration dictate the

general need for transportation (both public and

private) including the number of PT vehicles, number

and location of stations. Statistical data around these

parameters can be used to determine commute routes

in order to develop transportation hubs. Moving

speed determines the distance of travel, especially

during commute hours.

Total number and types of available roads allow

calculation of optimal routes for trucks, including city

bypass routes (for example, city of Jelgava in Latvia

has a regulation forbidding large trucks to pass

through city centre). Additionally, it serves as a data

source for PT route planning as it defines actual road

capacity, i.e., the maximum number of vehicles per

day.

FEMIB 2023 - 5th International Conference on Finance, Economics, Management and IT Business

Total number and types of vehicles in the city

allow calculating the distribution between private,

freight and PT vehicles, and it can be used as one of

the parameters in multi-criteria optimisation. PT data

includes the following parameters that relate to public

service vehicles such as busses and trains:

 Total number of PT vehicles on duty per day;

 Total number of transportation routes;

 Average fullness (%) of each PT vehicle per

route;

 Total number of trips per region;

 Most active period of the day (h);

 Ticket price;

 Fuel price;

 Use of multi-modal PT routes.

Total number of PT vehicles on duty and

corresponding total number of transportation routes

shows the usage of PT versus private vehicles.

Average fullness is one of parameters for calculating

effectiveness of transportation system. Total number

of trips determine the need to either increase or reduce

the number or PT vehicles depending on fullness

statistics.

The most active period of the day, or several such

periods, define the required amount of stations and

frequency of PT lines. The ticket price versus fuel

price (including use of private vehicles) shows how

attractive PT is to residents compared to other travel

options. Multi-modal PT routes defined by the

number of different types of PT vehicles used for one

trip provide necessary information to determine if

improvements to an existing route are required and/or

a new PT route is needed and/or change towards hub-

type mobility model must be considered.

2.2 Monitoring Data

This data includes the following parameters relating

to observatory data that can serve as supportive

information:

 Number of vehicles entering and exiting the

city;

 Moving speed of vehicles within city limits;

 Parking violations;

 Environmental data.

This type of data is typically provided by

monitoring devices embedded in urban infrastructure:

video cameras, speed sensors, meteorological

stations, etc. Cameras may be used at city entrances

in order to register vehicles entering and exiting the

city in order to determine internal city traffic and

improve or add/remove existing entry/exit routes.

Speed sensors embedded in road surface determine

the moving speed of vehicles in order to optimise the

traffic flow (including road capacity and traffic flow

intensity). Parking violations can decrease road

capacity, thus breaking the optimised traffic flow.

Finally, the environmental parameters such as air

temperature, wind direction, wind speed, humidity

and rain/snow precipitations can affect traffic

conditions, resulting in the need to adjust traffic flow

(for example, changing the traffic light algorithm in

real-time).

2.3 Mobile Activity and Data

Acquisition

This data includes the following parameters that

depict mobile activity, including the location and

movement, of residents:

 Mobile call density in a territory;

 Mobile user density in a territory;

 Mobile internet usage in a territory;

 Mapping of mobile base stations within a

municipal territory;

 GPS coordinates for mobile base stations;

 Spatial grid of 1km x 1km size;

 Mapping of mobile base stations in the spatial

grid of 1km x 1km size.

In general, this data should be used together with

other types of data in order to determine such aspects

as the proximity of residents’ commute route to PT

stations, the type of district based on commute hours,

i.e., business or dwelling area. Based on the

availability of data, including communication and

legal agreements with local municipality, traffic

authorities, PT companies and telecommunication

companies, two types of data were selected – PT data

and mobile activity data. The data was gathered in

Jelgava City, Latvia.

Based on the legal agreement between Latvia

University of Life Sciences and Technologies and

Jelgava City Municipality, PT data was provided by

Jelgavas Autobusu Parks Ltd, the main PT service

operator in Jelgava. The provided data cover two

comparable periods – February 2017 and February

2022, to demonstrate the effect of COVID-19

restrictions on the use of PT vehicles. The following

data was provided:

 Number of entries and exits from a PT vehicle

at a particular bus stop, % of total number;

 Difference between entries and exits at a

particular bus stop, n count;

 Distance travelled per trip and per day, km;

Optimizing Transport Network to Reduce Municipality Mobility Budget

 Distance travelled depending on passenger

category, % and n count;

 Duration of travel per trip and per day, min;

 Average duration of travel per week, min.

Mobile activity data was provided by the

telecommunication company, Latvijas Mobilais

Telefons Ltd (LMT). Mobile activity data was

selected for the same periods as for the PT data. The

data acquired was call data records that is a by-

product resulting from mobile operator network

events in Jelgava, Latvia. The location of particular

subscribers at the start of the phone call was

determined by the nearest mobile base station. Each

station is an infrastructure object that ensures mobile

connectivity. In addition, this data was consolidated

with 15-minute steps; therefore, identification of

movement of specific persons was not possible. In

addition, it was impossible to identify specific

persons; therefore, the privacy aspect was not an issue

that would require additional regulations. The

obtained data includes:

 The number of unique subscribers in Jelgava

(for years 2017 and 2022), n count;

 The number of unique subscribers per station

per day of the week and per hour of the day, n

count.

Analysis of data was performed using RStudio

(Posit, PBC, version 2022.02.0) and Python (Python

Software Foundation, version 3.10.4).

3 RESULTS

As it was previously mentioned, in order to get a

statistical impact of COVID-19 on PT vehicle use, the

data from February 2017 and February 2022 were

compared. On average, there were 2.24 times more

trips on weekdays in February 2017 than in February

2022. This shows a significant difference between the

number of trips on weekdays and holidays. In

comparison, there were 2.03 times more trips on

weekdays than on weekends in 2022.

At the same time, due to the COVID-19 restriction

measures, the total number of trips in 2022 decreased

by 62.9% compared to 2017, but the general tendency

between weekdays and weekends remained the same.

The distance travelled per single passenger trip was

calculated (see Table 1).

The results show that the distance travelled does

not differ significantly between weekdays and

weekends. The average distance travelled was 4.61 km

per passenger trip in February 2017 (see Figure 1).

Table 1: The distance travelled (km) per single passenger

trip in February 2017.

Day Mean Median Std.

Deviation

Number

of trips

Weekend day 4.63 3.90 2.67 32 724

Weekday 4.61 3.83 2.67 183 139

Total 4.61 3.85 2.67 215 863

Similarly, the distance travelled per single

passenger trip was calculated for February 2022. The

results show that the travelled distance does not differ

significantly between weekdays and weekend days –

the average distance was 4.73 km per passenger trip

in February 2022 (see Table 2).

Figure 1: Average distance travelled (km) per single

passenger trip in February 2017.

In 2017 and 2022, the average observed trip

duration was up to 13 minutes and no significant

differences between years regarding trip duration was

observed. In addition, trip duration did not differ

significantly between weekdays and weekends. The

same tendencies with average distance travelled per

trip per weekday were observed for 2022.

Table 2: The distance travelled (km) per single passenger

trip in February 2022.

Day Mean Median Std.

Deviation

Number

of trips

Weekend day 4.68 4.06 2.62 13 218

Weekday 4.74 4.12 2.68 67 052

Total 4.73 4.10 2.68 80 270

However, the behaviour patterns affected

by COVID-19 show significant changes – in 2022,

the travel distance decreased on weekends, while

in 2017, the travel distance increased on weekends

(see Figure 2).

4,59

4,60

4,61

4,62

4,63

4,64

Monday

Tuesday

Wednesday

Thursday

Friday

Saturday

Sunday

distance, km

FEMIB 2023 - 5th International Conference on Finance, Economics, Management and IT Business

Figure 2: Average distance travelled (km) per single

passenger trip in February 2022.

Due to the introduced COVID-19 restriction

measures, an overall decrease in the number of

residents travelling by PT was expected. Analysis of

passenger turnover was performed, and it was found

that the turnover in February 2022 was 2.6 times

lower than in February 2017.

3.1 Mobile Activity Data Analysis

Mobile activity data was analysed using Principal

Component Analysis (PCA) with an aim to categorise

districts based on their economic activity. PCA was

selected to generate complex factors that have linear

correlation with original features, i.e., unique

subscribers per base station during a day and time.

As a result, two complex factors that describe

94.9% of overall data variability PC1 (65.2%) and

PC2 (29.7%) were obtained. The average values of

PCs were calculated based on weekdays in February

2017. It was concluded that PC1 has higher values on

Fridays and Sundays and lower values on other days.

Therefore, it hypothesised that the city districts

included in PC1 are residential districts, but those in

PC2 - business districts.

Upon analysing individual base stations, it was

concluded that three types of districts could be

distinguished – business, residential and mixed

districts. However, only one base station (Liela iela

2) belonged to a business district – others were

categorised as either mixed or residential types.

Similarly, the average values of PC1 and PC2

were calculated based on the time of the day (in

hours). On weekdays, PC1 had lower values between

00:00 and 15:00 and higher values after working

hours (15:00-22:00), while PC2 had higher values

between 00:00 and 15:00. Therefore, the previously

stated hypothesis that PC1 referred to residential

districts, but PC2 – to business districts is valid.

Based on these results and data on the total

number of subscribers per base station, there were

three groups of districts in February 2017 (see Figure

3). Only one base station (Liela iela 2) referred to a

business district category.

Analysis of data for February 2022 lead to the

conclusion that the overall division of districts have

changed due to COVID-19 restrictions, causing

increase of remote work from home. Using PCA, two

complex factors that describe 95.7% of overall data

variability PC1 (51.0%) and PC2 (44.7%) were

obtained.

Instead of falling into strict categories of

residential or business type districts, every district is

now considered as partly “mixed”. Therefore, two

groups of districts may be distinguished – mixed

business districts and mixed residential districts.

Figure 3: Base station classification based on district type:

blue=residential districts, black=mixed districts, orange=

business districts, February 2017.

For example, the only business district of 2017

became one of the multiple mixed business districts

in 2022 (see Figure 4). Overall, these results show

that initial classification of districts as a parameter for

determining point of interest (POI) for further

planning of mobility budget may vary depending on

situation.

Optimisation of PT network in order to assess and

assign a correct mobility budget to each resident is

based on the notion of appropriately selected

optimisation criteria. Analysis of bus network

4,65

4,70

4,75

4,80

Monday

Tuesday

Wednesday

Thursday

Friday

Saturday

Sunday

distance, km

Optimizing Transport Network to Reduce Municipality Mobility Budget

Figure 4: Base station classification based on district type:

blue=mixed residential districts, orange=mixed business

districts, February 2022.

operations and mobility activity data in Jelgava

provided the basis for understanding what exactly is

required to define such criteria.

3.2 Optimization Criteria for Public

Transportation Network

The primary goal of optimisation is to find a solution

that meets the needs of passengers in terms of

fairness, including the choice of PT vehicle as type of

transport, availability of such PT vehicles on the city

roads and accessibility to PT vehicle stops. Addition

of mobility budget on top of passenger needs results

in a requirement to find a solution for optimal PT

vehicle trip described by appropriate route in relation

to district classification.

The optimisation problem solutions are defined as

a mobility budget provided on a monthly basis as:

 Decision space in which the individual person

can choose from different options to cut down

emissions and stay within the limit;

 Indicator for authorities and transport providers

where improvements in accessibility and

transport options are needed to “unburden”

narrow budgets.

Therefore, the objective for optimising the

transportation network in Jelgava, Latvia, can be

defined as follows:

 Minimise the travel distance required to access

the point of interest when using single type of

transportation, i.e., a bus.

 Minimise travel duration to access the point of

interest when using multiple types of

transportation or multiple PT vehicles for a

single trip.

The objective functions are subjected to multiple

defined constraints:

 Decarbonisation of a specific region (e.g. a

city), for example, reduction by 10% compared

to the previous year, divided among all

residents;

 Local circumstances: accessibility and

availability of alternatives;

 Social circumstances: supply and care

obligations, financial situation;

 Basic functions of living: work, education,

daily needs;

 Constants of human mobility, e.g. the travel

time budget (60-90 minutes per day, regardless

of the means of transport or location), and 3-4

trips per day;

 Trading option for a limited part of the

emission allowances per capita (e.g. 10%).

4 CONCLUSIONS

Published research suggest that the mode of travel

leaves the largest impact on the carbon footprint per

capita. Analysis of people behaviour also show that

personal vehicles were preferred mode of mobility

during COVID-19 pandemic.

Data analysis suggests also that the behavioural

changes caused by the initial enforcement and later

encouragement of COVID-19 restriction measures

led to rapid decrease in the use of PT vehicles. In

February 2022, the total number of passenger trips

had decreased by 62.9% compared to February 2017,

leading to a dramatic decrease of overall passenger

turnover.

It is interesting, however, that the intensity of PT

vehicle uses on various days of the week did not

change much – in February 2017, there were 2.24

times more passenger trips on weekdays than on

weekends, but in February 2022, the same parameter

was 2.03 times more trips on weekdays compared to

weekends.

At the same time, the observed travel time and

total duration per passenger did not decrease, and

remained at about 13 minutes for both periods.

Multiple researchers found similar situations in their

countries (Chang et al., 2021; Delbosc et al., 2022;

Shaer & Haghshenas, 2021; Sogbe, 2021).

FEMIB 2023 - 5th International Conference on Finance, Economics, Management and IT Business

It is important to mention that except of analysing

general transportation statistics, no data relating to

equity principles (Lucas et al., 2019) was observed.

This would include segregation into genders, age (for

example, more detailed than just “children”),

ethnicity and overall ability of movement. Even

though these parameters play a significant role in

optimisation and mobility budget assignments, no PT

service company would gather such data.

Regarding the classification of mobile phone base

stations, it was observed that any drastic change in

behaviour leads to inadequate initial parametrisation.

This can also mean that optimisation of transportation

network according to class-to-be is most likely to

cause behaviour changes regarding the use of PT

vehicles.

However, mobile activity data reveal enough

information to determine exactly how COVID-19

affected the behaviour of city districts and provide

concrete information about the necessary re-planning

measures for PT station locations.

As the modes of mobility even after the COVID-

19 pandemic shifted more to use private vehicles,

particularly cars, government institutions may need to

initiate switch from extensive use of private vehicles

to public vehicles or private vehicles which produce

less CO2. This could be done through campaigns,

providing optimized PT routes and times, offering

free tickets and other ways. Thus balancing, for

example, between reaching decarbonisation goals and

providing sustainable PT network which is used by

citizens.

ACKNOWLEDGEMENTS

The research leading to these results has received

funding from the Horizon 2020 ERA-NET Cofund

Urban Accessibility and Connectivity (EN-UAC)

project “Individual Mobility Budgets as a Foundation

for Social and Ethical Carbon Reduction”

(MyFairShare).

REFERENCES

Al-Ashhab, M. S. (2022). A multi-objective optimization

modelling for design and planning a robust closed-loop

supply chain network under supplying disruption due to

crises. Ain Shams Engineering Journal, 101909.

https://doi.org/10.1016/j.asej.2022.101909

Bhoyar, S. P., Dusad, S., Shrivastava, R., Mishra, S., Gupta,

N., & Rao, A. B. (2014). Understanding the Impact of

Lifestyle on Individual Carbon-footprint. Procedia -

Social and Behavioral Sciences, 133, 47–60.

https://doi.org/10.1016/j.sbspro.2014.04.168

Bocken, N. M. P., & Allwood, J. M. (2012). Strategies to

reduce the carbon footprint of consumer goods by

influencing stakeholders. Journal of Cleaner

Production, 35, 118–129. https://doi.org/10.1016/j.jcle

pro.2012.05.031

Boke Olén, N., Roger, F., Brady, M. V., Larsson, C.,

Andersson, G. K. S., Ekroos, J., Caplat, P., Smith, H.

G., Dänhardt, J., & Clough, Y. (2021). Effects of farm

type on food production, landscape openness, grassland

biodiversity, and greenhouse gas emissions in mixed

agricultural-forestry regions. Agricultural Systems,

189, 103071. https://doi.org/10.1016/j.agsy.2021.103

071

Caggiani, L., Camporeale, R., & Ottomanelli, M. (2017).

Facing equity in transportation Network Design

Problem: A flexible constraints based model. Transport

Policy, 55(January), 9–17. https://doi.org/10.1016/

j.tranpol.2017.01.003

Chang, H. H., Meyerhoefer, C. D., & Yang, F. A. (2021).

COVID-19 prevention, air pollution and transportation

patterns in the absence of a lockdown. Journal of

Environmental Management, 298, 113522.

https://doi.org/10.1016/j.jenvman.2021.113522

Crainic, T. G., Gendron, B., & Akhavan Kazemzadeh, M.

R. (2022). A taxonomy of multilayer network design

and a survey of transportation and telecommunication

applications. European Journal of Operational

Research, 303(1), 1–13. https://doi.org/10.1016/j.ejor.

2021.12.028

Crotti, D., Grechi, D., & Maggi, E. (2022). Reducing the

carbon footprint in college mobility: The car

commuters’ perspective in an Italian case study.

Environmental Impact Assessment Review, 92, 106702.

https://doi.org/10.1016/j.eiar.2021.106702

Delbosc, A., Currie, G., Jain, T., & Aston, L. (2022). The

‘re-norming’ of working from home during COVID-19:

A transtheoretical behaviour change model of a major

unplanned disruption. Transport Policy, 127, 15–21.

https://doi.org/10.1016/j.tranpol.2022.08.007

European Comission. (2020). 2050 long-term strategy.

https://climate.ec.europa.eu/eu-action/climate-

strategies-targets/2050-long-term-strategy_en

European Environment Agency. (2021). Data viewer on

greenhouse gas emissions and removals, sent by

countries to UNFCCC and the EU Greenhouse Gas

Monitoring Mechanism (EU Member States).

https://www.eea.europa.eu/data-and-maps/data/data-

viewers/greenhouse-gases-viewer

Fan, Z., Fang, Y., Li, W., Cai, X., Wei, C., & Goodman, E.

(2019). MOEA/D with angle-based constrained

dominance principle for constrained multi-objective

optimization problems. Applied Soft Computing

Journal, 74, 621–633. https://doi.org/10.1016/j.asoc.20

18.10.027

Farahani, R. Z., Miandoabchi, E., Szeto, W. Y., & Rashidi,

H. (2013). A review of urban transportation network

design problems. European Journal of Operational

Optimizing Transport Network to Reduce Municipality Mobility Budget

Research, 229(2), 281–302. https://doi.org/10.1016/

j.ejor.2013.01.001

Gu, Z., Shafiei, S., Liu, Z., & Saberi, M. (2018). Optimal

distance- and time-dependent area-based pricing with

the Network Fundamental Diagram. Transportation

Research Part C: Emerging Technologies, 95, 1–28.

https://doi.org/10.1016/j.trc.2018.07.004

Hänsel, M. C., Franks, M., Kalkuhl, M., & Edenhofer, O.

(2022). Optimal carbon taxation and horizontal equity:

A welfare-theoretic approach with application to

German household data. Journal of Environmental

Economics and Management, 116, 102730.

https://doi.org/10.1016/j.jeem.2022.102730

Hoffmann, S., Lasarov, W., & Reimers, H. (2022). Carbon

footprint tracking apps. What drives consumers’

adoption intention? Technology in Society, 69, 101956.

https://doi.org/10.1016/j.techsoc.2022.101956

International Energy Agency. (2022). Global energy

review: CO2 emissions in 2021. https://www.iea.org/

reports/global-energy-review-co2-emissions-in-2021-2

Jawadi, F., Rozin, P., & Bourghelle, D. (2022). Insights into

CO2 emissions in Europe in the context of COVID-19:

A panel data analysis. International Economics.

Kareinen, E., Uusitalo, V., Kuokkanen, A., Levänen, J., &

Linnanen, L. (2022). Effects of COVID-19 on mobility

GHG emissions: Case of the city of Lahti, Finland.

Case Studies on Transport Policy, 10(1), 598–605.

https://doi.org/10.1016/j.cstp.2022.01.020

Khanam, T., Rahman, A., Xiaoqian, X., Mola-Yudego, B.,

Moula, M. M. E., & Pelkonen, P. (2022). Assessing the

awareness and willingness of European experts to

reduce their carbon footprint in everyday consumption.

Environmental Impact Assessment Review, 97, 106889.

Li, X., Zhang, R., Chen, J., Jiang, Y., Zhang, Q., & Long,

Y. (2021). Urban-scale carbon footprint evaluation

based on citizen travel demand in Japan. Applied

Energy, 286, 116462. https://doi.org/10.1016/

j.apenergy.2021.116462

Lucas, K., Martens, K., Di Ciommo, F., & Dupont-Kieffer,

A. (2019). Measuring Transport Equity. In Measuring

Transport Equity. Elsevier. https://doi.org/10.1080/

01441647.2021.1983068

Millonig, A., Rudloff, C., Richter, G., Lorenz, F., & Peer,

S. (2022). Fair mobility budgets: A concept for

achieving climate neutrality and transport equity.

Transportation Research Part D: Transport and

Environment, 103(December 2021), 103165.

https://doi.org/10.1016/j.trd.2021.103165

Nicolini, G., Antoniella, G., Carotenuto, F., Christen, A.,

Ciais, P., Feigenwinter, C., Gioli, B., Stagakis, S.,

Velasco, E., Vogt, R., Ward, H. C., Barlow, J.,

Chrysoulakis, N., Duce, P., Graus, M., Helfter, C.,

Heusinkveld, B., Järvi, L., Karl, T., Papale, D. (2022).

Direct observations of CO2 emission reductions due to

COVID-19 lockdown across European urban districts.

Science of the Total Environment, 830, 154662.

https://doi.org/10.1016/j.scitotenv.2022.154662

Ogumerem, G. S., Kim, C., Kesisoglou, I., Diangelakis, N.

A., & Pistikopoulos, E. N. (2018). A multi-objective

optimization for the design and operation of a hydrogen

network for transportation fuel. Chemical Engineering

Research and Design, 131, 279–292. https://doi.org/

10.1016/j.cherd.2017.12.032

Pan, X. Z., Teng, F., Robiou du Pont, Y., & Wang, H. L.

(2022). Understanding equity–efficiency interaction in

the distribution of global carbon budgets. Advances in

Climate Change Research. https://doi.org/10.1016/

j.accre.2022.08.002

Randal, E., Shaw, C., Woodward, A., Howden-Chapman,

P., Macmillan, A., Hosking, J., Chapman, R., Waa, A.

M., & Keall, M. (2020). Fairness in transport policy: A

new approach to applying distributive justice theories.

Sustainability, 12(23), 10102. https://doi.org/10.3390/

su122310102

Rojas, C., Muñiz, I., Quintana, M., Simon, F., Castillo, B.,

de la Fuente, H., Rivera, J., & Widener, M. (2022).

Short run “rebound effect” of COVID on the transport

carbon footprint. Cities, 131, 104039. https://doi.org/

10.1016/j.cities.2022.104039

Sánchez-Barroso, G., González-Domínguez, J., & García-

Sanz-Calcedo, J. (2022). Impact of urban mobility on

carbon footprint in healthcare centers in Extremadura

(Spain). International Journal of Sustainable

Transportation, 16(7), 627–636. https://doi.org/10.10

80/15568318.2021.1914794

Shaer, A., & Haghshenas, H. (2021). The impacts of

COVID-19 on older adults’ active transportation mode

usage in Isfahan, Iran. Journal of Transport and Health,

23, 101244. https://doi.org/10.1016/j.jth.2021.101244

Sogbe, E. (2021). The evolving impact of coronavirus

(COVID-19) pandemic on public transportation in

Ghana. Case Studies on Transport Policy, 9(4), 1607–

1614. https://doi.org/10.1016/j.cstp.2021.08.010

Sun, Y., Cao, C., & Wu, C. (2014). Multi-objective

optimization of train routing problem combined with

train scheduling on a high-speed railway network.

Transportation Research Part C: Emerging

Technologies, 44, 1–20. https://doi.org/10.1016/

j.trc.2014.02.023

Vélez, A. M. A. (2022). Economic impacts, carbon

footprint and rebound effects of car sharing: Scenario

analysis assessing business-to-consumer and peer-to-

peer car sharing. Sustainable Production and

Consumption, 35, 238–249. https://doi.org/10.1016/

j.spc.2022.11.004

Wang, C., Ma, C., & Xu, X. D. (2020). Multi-objective

optimization of real-time customized bus routes based

on two-stage method. Physica A: Statistical Mechanics

and Its Applications, 537(September), 122774.

https://doi.org/10.1016/j.physa.2019.122774

Williges, K., Meyer, L. H., Steininger, K. W., &

Kirchengast, G. (2022). Fairness critically conditions

the carbon budget allocation across countries. Global

Environmental Change, 74, 102481. https://doi.org/

10.1016/j.gloenvcha.2022.102481

Yasmin, N., Jamuda, M., Panda, A. K., Samal, K., &

Nayak, J. K. (2022). Emission of greenhouse gases

(GHGs) during composting and vermicomposting:

Measurement, mitigation, and perspectives. Energy

Nexus, 100092.

FEMIB 2023 - 5th International Conference on Finance, Economics, Management and IT Business

Zafri, N. M., Khan, A., Jamal, S., & Alam, B. M. (2022).

Risk perceptions of COVID-19 transmission in

different travel modes. Transportation Research

Interdisciplinary Perspectives, 13, 100548.

Zhang, Y., Xue, W., Wei, W., & Nazif, H. (2022). A public

transport network design using a hidden Markov model

and an optimization algorithm. Research in

Transportation Economics, 92, 101095. https://doi.org/

10.1016/j.retrec.2021.101095

Optimizing Transport Network to Reduce Municipality Mobility Budget