Overview of Publicly-Available Data Sources on Road Trafﬁc Accidents

in Russia

Alexey Girin

, Nikolay Teslya

2 a

and Nikolay Shilov

2 b

ITMO University, St. Petersburg, Russia

SPC RAS, St. Petersburg, Russia

Keywords:

Road Trafﬁc Accident, Risk Factors, Data Sources, Data Analysis.

Abstract:

It is very important to develop measures to prevent trafﬁc accidents to increase road safety. The paper provides

identiﬁcation and description of various publicly-available sources of data related to road trafﬁc accidents in

Russia. All the described data could be combined to support an detailed analysis of accidents. A review and

classiﬁcation of risk factors associated with road accidents was conducted, and the data required for accident

analysis as well as publicly accessible sources of such data in Russia were described. Additionally, a review

of the methods used for analyzing and predicting accidents was undertaken.

1 INTRODUCTION

Road trafﬁc accidents (RTAs) represent a signiﬁcant

socio-economic problem that is relevant both glob-

ally and in Russia. According to ofﬁcial data, in 2023

in Russia there were more than 132.4 thousand RTA

(4.5% more than a year earlier), in these accidents

more than 14.5 thousand people died (2.3% more than

a year earlier) and more than 166.5 thousand were in-

jured (4.3% more than a year earlier) (Main Direc-

torate for Trafﬁc Safety of the Ministry of Internal

Affairs of Russia, 2024). RTAs also lead to mate-

rial losses, encompassing both direct costs associated

with property damage and medical care, and indirect

costs related to workforce loss and reduced labour

productivity (M

asilkov

a, 2017).

To ensure road safety, it is essential to develop

measures to prevent RTAs. One of the main aspects

of road safety involves analyzing RTAs. The knowl-

edge gained about the causes of RTAs through this

analysis can increase the efﬁciency of the decision-

making process related to road safety (Goniewicz

et al., 2016). One of the important steps in RTA anal-

ysis is the collection and preparation of data for analy-

sis. This involves describing data related to accident-

prone factors, and identifying sources of such data.

The accuracy of analysis results directly depends on

the ability to integrate data from diverse sources,

https://orcid.org/0000-0003-0619-8620

https://orcid.org/0000-0002-9264-9127

as well as the completeness and reliability of those

sources (Gutierrez-Osorio and Pedraza, 2020).

Despite the fact that some researchers manage to

cooperate with public authorities in Russia in order to

gain access to data on RTA (Kudryavtsev et al., 2013),

the general tendency is that most of the RTA data are

used only by public authorities for internal analysis

and are not publicly available. This lack of an ac-

cessible tool for researchers to access comprehensive,

open data on RTAs in the country negatively impacts

the quality of RTA analysis results. The aim of this

paper is to identify and describe various data sources

on RTAs in Russia that are open for public access and

can be combined for further analysis by researchers.

The paper is organized as follows. Section 2 out-

lines the methods used in this study. Section 3 re-

views the RTA risk factors. Section 4 describes the

RTA data sources speciﬁcally in Russia. In section 5,

methods of RTA analysis and prediction are reviewed.

A discussion of the study follows in section 6, and the

paper concludes in section 7.

2 RESEARCH METHODS

Firstly, we systematizes the potential causes of RTAs

described in the review literature of RTA risk factors,

and identify data associated with these factors. Based

on this review, we identify a classiﬁcation of RTA risk

factors into three categories: human factors, environ-

480

Girin, A., Teslya, N. and Shilov, N.

Overview of Publicly-Available Data Sources on Road Trafﬁc Accidents in Russia.

DOI: 10.5220/0012737100003702

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 10th International Conference on Vehicle Technology and Intelligent Transport Systems (VEHITS 2024), pages 480-487

ISBN: 978-989-758-703-0; ISSN: 2184-495X

mental factors, and vehicle factors. Next, we review

papers on RSA research to identify data sources that

correlate to the data usually used in RTA analysis.

These sources are categorized as follows: public au-

thorities, onboard devices, road infrastructure, and so-

cial media. Using this categorization, we search and

analyze publicly available data sources on RTAs in

Russia. To identify RTA data sources from public au-

thorities in Russia, we examine regulations regarding

RTA data registration and information resources pro-

vided by those authorities. We also review mapping

and weather services that provide additional data for

contextualising RTAs. Finally, we review the methods

that use these data to analyse and predict RTAs.

3 OVERVIEW OF RTA RISK

FACTORS

Road trafﬁc, as a complex dynamic system, includes

several interrelated elements like trafﬁc participants

(drivers, passengers, pedestrians), vehicles and the

external environment, which consists of the road net-

work, weather and time conditions, as well as the

surrounding terrain and trafﬁc ﬂow (Fig. 1) (Vogel

and Bester, 2005). Disturbances in the functioning of

these elements can lead to RTAs. Each trafﬁc element

deﬁnes a group of RTA risk factors.

Figure 1: RTA risk factors.

3.1 Human Factors

Human-related RTA risk factors are divided according

to two criteria: the intentionality of the road user’s

behaviour and the duration of the factor’s inﬂuence

(Fig. 2) (Bucsuh

azy et al., 2020).

Long-term RTA risk factors caused by intentional

actions are associated with negative habits and be-

Figure 2: RTA. Human factors.

havioural characteristics of road users. Such factors

include:

• Neglect of passive safety (ignoring the use of

child seats and seat-belts) (Fattakhov, 2018).

• Regular reckless or dangerous behaviour (speed-

ing and failure of drivers to maintain lateral spac-

ing) (Alonso Pl

a et al., 2013).

The actions of road users that are consciously

taken immediately before and cause a accident are the

short-term accident risk factors caused by the inten-

tional behaviour of road users:

• Underestimation of risk (e.g. overtaking on an

unsafe road section, overtaking in heavy traf-

ﬁc, crossing the road in an inappropriate place)

(Parker et al., 1995).

• Distraction while driving (e.g. using a mo-

bile phone while driving) (Horsman and Conniss,

2015).

• Alcohol and drug intoxication (Rodionova et al.,

2022).

Persistent conditions that adversely affect road

users identify the following long-term risk factors for

RTA: medical causes (chronic neurological diseases,

narcolepsy) (Lindsay and Baldock, 2008), distracted

behaviour (Parker et al., 1995), insufﬁcient driving

experience (Hu et al., 2020).

Short-term factors caused by unintentional driv-

ing behaviour include: mental and somatic abnormal-

ities (acute stress, bouts of illness) (Taylor and Dorn,

2006), fatigue state (microsleep) (De Mello et al.,

2013), panic reaction (Bucsuh

azy et al., 2020).

The following data on RTA’s participants are cor-

related with the identiﬁed risk factors: common in-

formation (age, driving experience), medical informa-

Overview of Publicly-Available Data Sources on Road Trafﬁc Accidents in Russia

481

tion (chronic diseases, mental and somatic abnormal-

ities), condition and behaviour of the participants.

3.2 Inﬂuence of the External

Environment

Trafﬁc is inﬂuenced by the external environment,

such as the condition of the road network, weather

and time conditions, the characteristics of the sur-

rounding area, and trafﬁc ﬂow.

RTA risk factors caused by the condition of the

road network include (Fattakhov, 2018; Wang et al.,

2013): the arrangement of roads (location of trafﬁc

lights and regulatory signs, etc.), road surface condi-

tions (low level of pavement, presence of potholes),

road geometries, conditions of artiﬁcial lighting, and

the presence of foreign objects in trafﬁc areas.

RTA risk factors caused by weather and time con-

ditions (Zou et al., 2021; Hazaymeh et al., 2022): pre-

cipitation, wind, fog; time of day, and natural lighting

conditions.

RTA risk factors caused by the characteristics of

the surrounding terrain (Hazaymeh et al., 2022): ter-

rain relief (ﬂat, hilly, or mountainous), and the density

and type of buildings and other constructions.

RTA risk factors caused by the characteristics of

trafﬁc ﬂow: speed regime and trafﬁc intensity are sep-

arately identiﬁed (Zhang et al., 2020).

Consequently, the following data on the environ-

mental conditions at the accident site during the RTA

correlate with the identiﬁed risk factors: the state and

layout of the road infrastructure, the time and weather,

terrain, the presence, type, and density of points of in-

terest, and the characteristics of the trafﬁc ﬂow.

3.3 Inﬂuence of Vehicles

RTA risk factors related to vehicles (Zovak et al.,

2016): wear and tear of the vehicle (correlated with

the following data: vehicle mileage, period of use);

unsatisfactory technical condition of the vehicle (data

on the presence of damage and faults of the vehicle).

4 OVERVIEW OF RTA DATA

SOURCES

RTA data sources that are used in accident analysis

and prediction studies:

• Public Authorities: generally collect data on the

type, location, time and participants, as well as

data on the causes and consequences of the acci-

dent (Rabbani et al., 2022).

• On-board Devices: devices that are mounted on

a vehicle to collect vehicle characteristics and

driver state information (e.g. drowsiness, anxiety,

distraction) (Gutierrez-Osorio and Pedraza, 2020;

Chand et al., 2021).

• Road Infrastructure Devices: technical devices

that monitor the movement of vehicles: radars,

cameras, loop trafﬁc detectors. (Gutierrez-Osorio

and Pedraza, 2020).

• Social Media: information extracted from acci-

dent publications has been used to identify RTA

locations and to identify clusters of high accident

rates (Chand et al., 2021).

4.1 RTA Data Sources in Russia

The main source of data on RTA in Russia are organ-

isations that keep records of RTA: internal affairs au-

thorities; road owners; organisations owning vehicles;

and medical organisations. Table 1 provides informa-

tion about the data on RTA collected by these organ-

isations. The data on RTA records from internal af-

fairs authorities are partially open for public access.

These data can be accessed through a specialised ser-

vice ”Indicators of the Road Safety State” provided

by the Trafﬁc Police of the Russian Ministry of In-

ternal Affairs and described in the subsection 4.1.1.

This service is the primary, and in most cases, the only

source of data on trafﬁc accidents used in the analysis

of RTAs in Russia.

For instance, the research (Donchenko et al.,

2020) considered the potential for creating a model

that could predict trafﬁc accidents based on data about

accidents in Russia between the beginning of 2015

and April 2018. These data was provided by this

service. Another example of using this service is re-

search (Kasatkina and Vavilova, 2023), which devel-

oped an intelligent system that determines the con-

centration of road accidents in the Udmurt Republic

(Russia) through the use of accident data from 2022.

Some accident analysis studies conducted in Rus-

sia use accident data acquired from the ”Indicators of

the State of Road Safety” service in conjunction with

information from various other sources. For example,

the authors of the study (Fatkulin et al., 2017) used

data obtained from the service and data extracted from

posts in VKontakte social network during 2017 when

developing a system for monitoring accidents through

real-time analysis of social media.

For a more in-depth contextualisation of RTA, it

is necessary to consult additional data sources. These

include road and weather conditions at the time of the

RTA, the condition and behaviour of the participants

in the RTA, and the environment of the RTA scene.

VEHITS 2024 - 10th International Conference on Vehicle Technology and Intelligent Transport Systems

482

Table 1: RTA data in Russian organisations.

Organisations Description of RTA

record data

Public

access

Internal affairs

authorities

General information

about the RTA, in-

formation about the

external environment

(weather conditions,

road network state,

surroundings of the

accident site), and par-

ticipants and vehicles

involved in the RTA

Partial

Road owners Data on the condition

of the road network in

the area of the RTA

Organisations

owning vehi-

cles

General information

about the accident

and results of internal

investigation: infor-

mation about driver

and injured (severity),

information about

vehicles, details of

accident (violation

of trafﬁc rules fact),

consequences of acci-

dent (material losses,

number of deaths and

injuries)

Medical or-

ganisations

Data on those killed

and injured in RTA

who were treated by a

medical organisation

These sources are described in the following subsec-

tions.

4.1.1 Indicators of the State of Road Safety

”Indicators of the State of Road Safety” is a service

provided by the Russian Ministry of Internal Affairs

Trafﬁc Police. This service provides statistical data on

the state of accident rates and detailed accident cards

(Main Directorate for Trafﬁc Safety of the Ministry of

Internal Affairs of Russia, 2024). The cards contain

detailed information on the circumstances of speciﬁc

RTA. The cards follow a ﬁxed structure with sections

for general information, scheme of the accident, road

conditions on the scene, data on vehicles, and other

participants (Table 2). The service does not provide

information on permissible values for each ﬁeld of the

card. However, this information can be obtained from

regulations and recommendations which are publicly

available. The procedure for ﬁlling out the ﬁelds is

Table 2: Structure of RTA card.

Section Title Section Description

General infor-

mation

Place and time of the RTA,

number of participants and

number of vehicles

Road condi-

tions

Description of objects at the

accident site and near it, road

conditions and deﬁciencies,

weather conditions, and other

information about the sur-

roundings of the accident

Vehicles Description of vehicles in-

volved in the accident, pres-

ence or absence of technical

faults

RTA partici-

pants

Categories of participants

(driver, passenger, cyclist,

etc.), level of injury, breach of

road trafﬁc rules

RTA diagram A graphical representation of

the circumstances surround-

ing the accident (the location

and direction of motion of

participants and vehicles dur-

ing the accident)

described in the ”Recommendations for recording and

analysing RTA on Russian highways” from the Fed-

eral Road Agency Rosavtodor. Table 2 gives a de-

scription of each section of the cards. Maps can be

downloaded separately for each region of the Russian

Federation. The period of time for which all cards

can be downloaded at once is limited to one month.

The service only allows downloading cards for those

RTAs that occurred before 2015. The following doc-

ument formats are supported: PDF, XLS, CSV, and

XML. The service is a major source of data on RTAs

in Russia, providing the most comprehensive basic in-

formation about the circumstances of these RTAs.

4.1.2 Road Fund Control System

Road Fund Control System from ROSDORNII pro-

vides information about roads and road papers plans,

as well as data on RTA (The Federal Autonomous

Institution ”Russian Road Scientiﬁc-Research Insti-

tute”, 2024). The data can be accessed through the

user interface; the system does not provide a tool for

uploading them. The system provides detailed up-to-

date data on both federal and regional roads and local

roads. Table 3 provides a description of the informa-

tion page sections for road data, some of the ﬁelds

(road characteristics) are tempered, i.e. their start and

Overview of Publicly-Available Data Sources on Road Trafﬁc Accidents in Russia

483

Table 3: Structure of the road information page in Road

Fund Control System.

Section Title Section Description

Legal informa-

tion

Road owner and operating or-

ganisation

Project activity Information on the types of

papers on a particular road

Technical data Road class and category,

pavement type, number

of lanes, maximum speed,

climatic zone, axle load,

capacity, carriageway width,

subgrade width

Normative

road condition

Compliance of the road with

the normative condition

Safety Availability of the accident-

hazardous status, accidents

with dead and injured people

on the road

end times are speciﬁed for the values of these ﬁelds.

The data on RTA provided by the system correspond

in structure and content to the data of the ”Indicators

of the State of Road Safety” service. Data on road

plan documents for a particular motorway include:

type of documents, timing of documents, technical in-

dicators of road sections before and after documents,

etc. The information system Road Fund Control Sys-

tem can be used as an additional source of data on

RTA, speciﬁcally to obtain information about the con-

dition of the road network in the area where the acci-

dent occurred.

4.1.3 Mapping Services

The OpenStreetMap (OSM) service (OpenStreetMap

Contributors, 2024a), which is the leading open-

source of geographic information (Mooney et al.,

2017), can be used to collect additional information

about the accident site environment. The process of

deﬁning the accident site environment using the OSM

can be described as follows:

1. Deﬁne a set of objects types that may affect trafﬁc

safety.

2. Match each type of objects from the obtained set

with a template for search in OSM.

3. Implement a function that searches for objects of

speciﬁc types within the given radius around the

coordinates of the accident.

To obtain OSM data it is possible to download

a snapshot for a certain region using Geofabrik tool

(Geofabrik GmbH Karlsruhe, 2024). An alternative

option is to use the Overpass service (OpenStreetMap

Contributors, 2024b), which provides an API for ob-

taining OSM objects by user requests. The obtained

data can be used in addition to the data from the ”road

conditions” section of RTA cards.

4.1.4 Sources of Weather Data

Data on weather conditions at the location and time

of the accident can be obtained directly from the ac-

cident card. For a detailed report on weather con-

ditions it may be necessary to use external services.

The main criteria for choosing such a service are:

the ability to access historical weather data based on

geographic coordinates and time, the availability of

data for Russia, the maximum data update interval

should not exceed one hour (to ensure the most ac-

curate readings), the ability to retrieve values for a

speciﬁc range of weather variables. One of the ser-

vices that meets these criteria is Open-Meteo, which

provides an API for accessing historical weather data

up to 10,000 times per day for personal use, making

it the best option compared to similar services (Open-

Meteo contributors, 2024).

4.1.5 Court Websites

Data on the circumstances of a particular RTA can be

obtained from the text of a court judgment on the acci-

dent. The sources of documents with verdicts in RTA

cases include court websites, where the texts of court

decisions are published. When working with these

sources, the following difﬁculties arise: the need to

compare court decisions with speciﬁc RTAs; the need

to handle unstructured texts from court decisions; and

the dispersion of places where court decisions are

posted. With a court decision text analysis it is pos-

sible to determine information about the participants

in an accident (a description of their behaviour during

the accident, their medical condition, the fact of al-

cohol or drug intoxication, and any injuries received

as a result of the RTA), information about vehicles

(model, period of use, and presence and type of me-

chanical malfunctions), and the state of the surround-

ing environment (the weather and time conditions and

condition of the road network).

4.1.6 Other Sources

Alternative sources of open data on RTAs in Russia

include publications on social media and in the public

media, as well as publicly available recordings from

onboard video recorders. Further research is required

to explore the possibilities for searching, interpreting

and comparing these data with speciﬁc RTA.

VEHITS 2024 - 10th International Conference on Vehicle Technology and Intelligent Transport Systems

484

5 METHODS OF RTA ANALYSIS

AND PREDICTION

In order to further use of RTA data from various pub-

lic sources for RTA analysis and prevention purposes,

it is necessary to have an understanding of the cur-

rent methods for analyzing and predicting RTAs. It is

worth noting that a signiﬁcant number of techniques

used in analyzing and predicting RTAs are related to

artiﬁcial intelligence and machine learning.

5.1 Identiﬁcation of High Accident

Clusters

One of the steps in road safety is to identify high-

accident clusters in a given area. This task can be for-

mulated as a search for road segments with the high-

est accident density. To solve this problem, cluster-

ing methods based on the analysis of historical data

on RTA are used (Chang et al., 2022). For example,

in the study (Moosavi et al., 2019) DBSCAN clus-

tering algorithm is used to identify clusters of higher

accident density. Another common method is Ker-

nel Density Estimation (KDE), which allows to cal-

culate the density of RTA points in the study area and

identify areas with the highest density values (Santos

et al., 2021).

5.2 Identiﬁcation of RTA Risk Factors

The high fatality rate in RTA is attributed to the lack

of awareness of RTA risk factors, and many stud-

ies have aimed at identifying these factors, which is

also an important step towards road safety (Alkheder

et al., 2020). For example, the analysis of RTA data

in (Wang et al., 2021) is helped to identify RTA risk

factors and to propose a model to quantify them in

order to compare the probability of RTA in different

scenarios. Part of the researches in this area focuses

on evaluating the impact of RTA risk factors on the

severity of the accident consequences, which can be

expressed through the number of fatalities and casu-

alties in the accident, as well as the extent of injury

to victims. For example, in (Eboli et al., 2020) this

problem is solved using Binary Logistic Regression,

and in (Alkheder et al., 2020) the decision tree, Sup-

port Vector Machine (SVM) and Bayesian network

are used to estimate the degree of inﬂuence of risk

factors on accident severity.

5.3 RTA Prediction

Challenges in the ﬁeld of RTA prediction include pre-

dicting the occurrence and estimating the severity of

RTA. One of the advanced trends in this area is the

application of deep learning algorithms (Gutierrez-

Osorio and Pedraza, 2020). For example, (Rolison,

2020) presented a method for predicting the severity

of RTA based on encoding a matrix of RTA data into

special grey images, which are then used as inputs

for a convolutional neural network. The results of the

study showed that the proposed method of predicting

RTA severity using a convolutional neural network

provides higher accuracy compared to other methods

such as K-nearest neighbours algorithm, SVM and re-

current neural networks.

Rough set theory can be used in RTA data anal-

ysis to assess the inﬂuence of risk factors on RTA

outcomes. It can also be used to reduce the number

of attributes associated with risk factors and to detect

hidden relationships between the data and for classi-

ﬁcation tasks (Jianfeng et al., 2019).

In papers on RTA analysis and prediction, there is

a tendency to use multiple data sources to improve

the accuracy of the results. Researchers explicitly

point out that the success of prediction models de-

pends mainly on how data from different sources can

be integrated (Chand et al., 2021; Suat-Rojas et al.,

2022). In addition to the above studies, another exam-

ple of such integration is the paper of (Marcillo et al.,

2022), which combined heterogeneous data on trafﬁc

accidents, weather conditions and hotspots. Based on

a deep neural network, a real-time trafﬁc accident pre-

diction model was built, which performs better than

other models.

6 DISCUSSION

6.1 RTA Risk Factors

The following RTA factors were identiﬁed: human

factors; external environment; vehicle factors. Data

that relate to these factors can be used to provide a

comprehensive understanding of the accident context.

6.2 RTA Data Sources

The sources of RTA data include: public authorities,

onboard devices, road infrastructure devices, and so-

cial media. The main source of RTA data in Russia

is the trafﬁc police service (public authorities), which

is open to the public and provides basic information

about the circumstances of the RTAs, accident partic-

ipants, and vehicles. Other data collected by public

authorities on RTAs are used only for internal pur-

poses and are closed to the public. Additional sources

Overview of Publicly-Available Data Sources on Road Trafﬁc Accidents in Russia

485

of public RTA data that could be integrated into a uni-

ﬁed system are:

• Road Fund Control System: road network data.

Can be used to determine the road condition at

the location of the accident, as well as to assess

the intensity of trafﬁc ﬂow.

• OSM mapping service: can be used to obtain data

on the environment of the accident site.

• Weather APIs: can be used to determine weather

conditions at the location and time of the accident.

• Court websites and aggregators of court decisions

can be used to obtain information on the circum-

stances of a road accident, a detailed description

of the state and behaviour of RTA participants,

and characteristics of vehicles and the external en-

vironment. The data is unstructured and requires

pre-processing.

• Publications on social media, as well as publicly

available recordings from onboard devices (video

recorders), need further research to explore the

possibilities of searching and interpreting data.

These sources are not included in this paper.

6.3 Methods of RTA Analysis and

Prediction

The following road safety tasks were identiﬁed: iden-

tiﬁcation of clusters of increased accident rate (DB-

SCAN clustering algorithm and KDE); identiﬁcation

of RTA risk factors (decision tree, SVM and Bayesian

network); accident prediction (convolutional neural

network and recurrent neural network).

7 CONCLUSION

This paper classiﬁes risk factors for RTAs and de-

scribes data that can be used in RTA analysis. The

sources of RTA data were reviewed, and the pub-

licly available sources in Russia were searched and

described. Additionally, a review was conducted on

methods for analysing and predicting RTAs. In the fu-

ture work, it is planned to implement a software sys-

tem to automatically collect RTA data from the public

sources, develop methods of RTA analysis and pre-

diction based on the data obtained.

ACKNOWLEDGEMENTS

The study was carried out within State Research,

project number FFZF-2022-0005.

REFERENCES

Alkheder, S., AlRukaibi, F., and Aiash, A. (2020). Risk

analysis of trafﬁc accidents’ severities: An applica-

tion of three data mining models. ISA transactions,

106:213–220.

Alonso Pl

a, F. M., Esteban Mart

ınez, C., Calatayud Mi

nana,

C., and Sanmart

ın, J. (2013). Speed and road acci-

dents: behaviors, motives, and assessment of the ef-

fectiveness of penalties for speeding. American Jour-

nal of Applied Psychology, 2013, vol. 1, num. 3, p.

58-64.

Bucsuh

azy, K., Matuchov

a, E., Z

uvala, R., Moravcov

a, P.,

Kost

ıkov

a, M., and Mikulec, R. (2020). Human fac-

tors contributing to the road trafﬁc accident occur-

rence. Transportation research procedia, 45:555–561.

Chand, A., Jayesh, S., and Bhasi, A. (2021). Road traf-

ﬁc accidents: An overview of data sources, analysis

techniques and contributing factors. Materials Today:

Proceedings, 47:5135–5141.

Chang, H., Li, L., Huang, J., Zhang, Q., and Chin, K.-S.

(2022). Tracking trafﬁc congestion and accidents us-

ing social media data: A case study of shanghai. Ac-

cident Analysis & Prevention, 169:106618.

De Mello, M. T., Narciso, F. V., Tuﬁk, S., Paiva, T., Spence,

D. W., BaHammam, A. S., Verster, J. C., and Pandi-

Perumal, S. R. (2013). Sleep disorders as a cause of

motor vehicle collisions. International journal of pre-

ventive medicine, 4(3):246.

Donchenko, D., Sadovnikova, N., and Parygin, D. (2020).

Prediction of road accidents’ severity on russian roads

using machine learning techniques. In Proceedings of

the 5th International Conference on Industrial Engi-

neering (ICIE 2019) Volume II 5, pages 1493–1501.

Springer.

Eboli, L., Forciniti, C., and Mazzulla, G. (2020). Factors in-

ﬂuencing accident severity: an analysis by road acci-

dent type. Transportation research procedia, 47:449–

456.

Fatkulin, T., Butakov, N., Krikunov, A., and Voloshin, D.

(2017). Accident monitoring framework based on on-

line social network sensing. Procedia computer sci-

ence, 119:278–287.

Fattakhov, T. (2018). Trafﬁc injuries in russia and its fac-

tors.

Geofabrik GmbH Karlsruhe (2024). Geofabrik, ofﬁcial

website. retrieved february 22, 2024. https://www.

geofabrik.de/.

Goniewicz, K., Goniewicz, M., Pawłowski, W., and Fiedor,

P. (2016). Road accident rates: strategies and pro-

grammes for improving road trafﬁc safety. Euro-

pean Journal of Trauma and Emergency Surgery,

42(4):433–438.

Gutierrez-Osorio, C. and Pedraza, C. (2020). Modern

data sources and techniques for analysis and fore-

cast of road accidents: A review. Journal of traf-

ﬁc and transportation engineering (English edition),

7(4):432–446.

Hazaymeh, K., Almagbile, A., and Alomari, A. H. (2022).

Spatiotemporal analysis of trafﬁc accidents hotspots

VEHITS 2024 - 10th International Conference on Vehicle Technology and Intelligent Transport Systems

486

based on geospatial techniques. ISPRS International

Journal of Geo-Information, 11(4):260.

Horsman, G. and Conniss, L. R. (2015). Investigating evi-

dence of mobile phone usage by drivers in road trafﬁc

accidents. Digital Investigation, 12:S30–S37.

Hu, L., Bao, X., Wu, H., and Wu, W. (2020). A study on cor-

relation of trafﬁc accident tendency with driver char-

acters using in-depth trafﬁc accident data. Journal of

advanced transportation, 2020:1–7.

Jianfeng, X., Hongyu, G., Jian, T., Liu, L., and Haizhu, L.

(2019). A classiﬁcation and recognition model for the

severity of road trafﬁc accident. Advances in Mechan-

ical Engineering, 11(5):1687814019851893.

Kasatkina, E. and Vavilova, D. (2023). Identiﬁcation and

mitigation of road trafﬁc accident concentration areas

by using cluster analysis. In 2023 5th International

Conference on Control Systems, Mathematical Mod-

eling, Automation and Energy Efﬁciency (SUMMA),

pages 247–251. IEEE.

Kudryavtsev, A. V., Kleshchinov, N., Ermolina, M., Lund,

J., Grjibovski, A. M., Nilssen, O., and Ytterstad, B.

(2013). Road trafﬁc fatalities in arkhangelsk, russia in

2005–2010: Reliability of police and healthcare data.

Accident Analysis & Prevention, 53:46–54.

Lindsay, V. and Baldock, M. (2008). Medical conditions as

a contributing factor in crash causation.

Main Directorate for Trafﬁc Safety of the Ministry of In-

ternal Affairs of Russia (2024). Information on road

safety indicators. retrieved february 22, 2024. http:

//stat.gibdd.ru/.

Marcillo, P., Valdivieso Caraguay,

A. L., and Hern

andez-

Alvarez, M. (2022). A systematic literature review

of learning-based trafﬁc accident prediction models

based on heterogeneous sources. Applied Sciences,

12(9):4529.

Mooney, P., Minghini, M., et al. (2017). A review of open-

streetmap data. Mapping and the citizen sensor, pages

37–59.

Moosavi, S., Samavatian, M. H., Parthasarathy, S., Teodor-

escu, R., and Ramnath, R. (2019). Accident risk

prediction based on heterogeneous sparse data: New

dataset and insights. In Proceedings of the 27th ACM

SIGSPATIAL International Conference on Advances

in Geographic Information Systems, pages 33–42.

asilkov

a, M. (2017). Health and social consequences of

road trafﬁc accidents. Kontakt, 19(1):e43–e47.

Open-Meteo contributors (2024). Open-meteo, ofﬁcial

website. retrieved february 22, 2024. https://open-m

eteo.com/en/docs/historical-weather-api.

OpenStreetMap Contributors (2024a). Openstreetmap. re-

trieved february 22, 2024. https://www.openstreetma

p.org.

OpenStreetMap Contributors (2024b). Overpass api. re-

trieved february 22, 2024. https://wiki.openstreetmap

.org/wiki/Overpass\ API.

Parker, D., Reason, J. T., Manstead, A. S., and Stradling,

S. G. (1995). Driving errors, driving violations and

accident involvement. Ergonomics, 38(5):1036–1048.

Rabbani, M. B. A., Musarat, M. A., Alaloul, W. S., Ayub,

S., Bukhari, H., and Altaf, M. (2022). Road accident

data collection systems in developing and developed

countries: a review. International Journal of Inte-

grated Engineering, 14(1):336–352.

Rodionova, M., Skhvediani, A., and Kudryavtseva, T.

(2022). Prediction of crash severity as a way of road

safety improvement: The case of saint petersburg, rus-

sia. Sustainability, 14(16):9840.

Rolison, J. J. (2020). Identifying the causes of road trafﬁc

collisions: Using police ofﬁcers’ expertise to improve

the reporting of contributory factors data. Accident

Analysis & Prevention, 135:105390.

Santos, D., Saias, J., Quaresma, P., and Nogueira, V. B.

(2021). Machine learning approaches to trafﬁc ac-

cident analysis and hotspot prediction. Computers,

10(12):157.

Suat-Rojas, N., Gutierrez-Osorio, C., and Pedraza, C.

(2022). Extraction and analysis of social networks

data to detect trafﬁc accidents. Information, 13(1):26.

Taylor, A. H. and Dorn, L. (2006). Stress, fatigue, health,

and risk of road trafﬁc accidents among professional

drivers: the contribution of physical inactivity. Annu.

Rev. Public Health, 27:371–391.

The Federal Autonomous Institution ”Russian Road

Scientiﬁc-Research Institute” (2024). Road fund con-

trol system. retrieved february 22, 2024. https://xn--d

1aluo.xn--p1ai/.

Vogel, L. and Bester, C. (2005). A relationship between

accident types and causes. SATC 2005.

Wang, C., Quddus, M. A., and Ison, S. G. (2013). The effect

of trafﬁc and road characteristics on road safety: A

review and future research direction. Safety science,

57:264–275.

Wang, X., Qu, Z., Song, X., Bai, Q., Pan, Z., and Li, H.

(2021). Incorporating accident liability into crash risk

analysis: A multidimensional risk source approach.

Accident Analysis & Prevention, 153:106035.

Zhang, Y., Lu, H., and Qu, W. (2020). Geographical detec-

tion of trafﬁc accidents spatial stratiﬁed heterogeneity

and inﬂuence factors. International journal of envi-

ronmental research and public health, 17(2):572.

Zou, Y., Zhang, Y., and Cheng, K. (2021). Exploring the

impact of climate and extreme weather on fatal trafﬁc

accidents. Sustainability, 13(1):390.

Zovak, G., Ku

cini

c, T., and

Sevo, I. (2016). Importance of

technical inspection of vehicles after trafﬁc accidents.

Trans Motauto World, 1(4):3–6.

Overview of Publicly-Available Data Sources on Road Trafﬁc Accidents in Russia

487