Analysing THE Impact of Road Information System on Traffic Safety

Analysing THE Impact of Road Information System on Traffic Safety

Available online at www.sciencedirect.com ScienceDirect Procedia Engineering 187 (2017) 712 – 721 10th International Scientific Conference Transbalt...

1022KB Sizes 0 Downloads 106 Views

Available online at www.sciencedirect.com

ScienceDirect Procedia Engineering 187 (2017) 712 – 721

10th International Scientific Conference Transbaltica 2017: Transportation Science and Technology

Analysing THE Impact of Road Information System on Traffic Safety Árpád Török*, Gábor Pauer, Tamás Berta KTI Nonprofit Ltd., Budapest, Hungary

Abstract

This paper provides a complex analysis framework for evaluating the role of appropriate transmission (appropriate quality, quantity, location and timing) of information in case of infrastructure safety and accident categories. The aim of the “infrastructure module” of the methodology is to evaluate, how much the appropriate transmission of information influences road safety at a given section of the road infrastructure. The aim of the “accident module” of the methodology is to define those types and causes of accidents, which are remarkably influenced by the appropriateness of information transmission. © 2017 2017Published The Authors. Published by Elsevier Ltd. © by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license Peer-review under responsibility of the organizing (http://creativecommons.org/licenses/by-nc-nd/4.0/). committee of the 10th International Scientific Conference Transbaltica 2017: Transportation and Technology. Peer-review underScience responsibility of the organizing committee of the 10th International Scientific Conference Transbaltica 2017 Keywords: road safety, information system, accident analysis, crash risk, infrastructure safety, safety indicator, risk factor analysis

1. Introduction Drivers constantly make decisions during their travel. These decisions are influenced by their rule knowledge, driving patterns and skills, as well as available information and interpretation during driving.

* Corresponding author. E-mail address: [email protected]

1877-7058 © 2017 Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license

(http://creativecommons.org/licenses/by-nc-nd/4.0/). Peer-review under responsibility of the organizing committee of the 10th International Scientific Conference Transbaltica 2017

doi:10.1016/j.proeng.2017.04.445

Árpád Török et al. / Procedia Engineering 187 (2017) 712 – 721

Information may come from the environment or the vehicle’s passenger compartment. There are direct and indirect information, and they may be related to the transport system (useful information) or not (useless information). Drivers make decisions based on their knowledge and information available, which determine the chain of events in transportation. An error occurring in the chain causes disturbance in the transport process that must be solved as soon as it is reaching a critical level. If it fails to resolve, road traffic accidents occur. The level of road safety is therefore significantly influenced by the quality of information [1]. It is important to systematically analyse the quality and accordance of information come from the environment, and to eliminate critical errors. Analysation of the complete transport process (instead of a specific situation), as well as interpretation of interactions of situations within the chain of events are the most important components of the method. 1.1. The theoretical approach of information One of the first designs of the information theory has been elaborated by C.E. Shannon and W. Weaver in 1949. Actually, it is a model of communication, which is not able to examine the meaning of the news [2]. In their research, transmissions of information between the transmitter and receiver sides, and potential barriers of understanding have been analyzed. Information is basically a word for unpredictability – relation of information and uncertainty is defined in information science. Information is the third basic component of the universe beside materials and energy. It occurs in the communication of any two objects, and which is more important, it is able to multiply by itself. Thus, information can not only be derived from structured data, but it is also present in chaotic conditions. It is reducing entropy and incoordination in both cases [3]. Receiver of a news usually know less about the state of a system that the transmitter. This different level of knowledge is called entropy. Entropy is to be reduced by transmission of information [4]. The concept of information takes place on the border of natural sciences and humanities; and perhaps helps to reduce gaps between the two disciplines. According to N. Wiener (the developer of cybernetics): “Information is information, not matter or energy. No materialism which does not admit this can survive at the present day” [5]. Apparently, opinions about the concept of information are going through a similar unification that happened with the concept of energy in the 19th century. 1.2. The place and role of information in the road transport system According to classical subdivision of the road transport system, basic components are the following: • Human, • Vehicle, • Infrastructure. Technological, economic and social development induces broader interpretation of the system. Some aspects are not sufficiently highlighted by the trichotomy. They are to be revealed by analyzing the elements of road transport system as well as the information system and relations between the drivers and the transport environment [6]. These aspects are becoming more and more important in modern information systems as market value of information is rising. Development of telematics systems and innovative solutions provides opportunity to give personalized information to the drivers about transport processes related to road operations, route-planning, accident situations, weather conditions, etc. These “tools” are becoming more integrated in the process of controlling the transport [7]. In the trichotomy of road transport system, some relations are not strong and efficient enough without information contact. This subdivision requires effective operation of information connections between the system components. However, beside conventional road information (e.g. static information like traffic signs and road markings); broader interpretation is necessary, which can be observed for example in the philosophy of road designing (e.g. in the design principles of self-explaining roads [8]).

713

714

Árpád Török et al. / Procedia Engineering 187 (2017) 712 – 721

As new possibilities to publish, classify and process information have been developed, information has been raised to system level [9]. Therefore, valuable information in the transport system has a key role in the field of road safety. New approach related to safety has been brought by road traffic management, however, system-oriented analysis of the role of information in transport has not been implemented yet [10]. Therefore, the aim of our research was to elaborate methods that are able to analyse the importance and necessity of the appropriate transmission of information in transport. To achieve this, two approaches have been applied: a method to characterize the infrastructure, as well as a method to identify critical types of accidents has been elaborated. 2. Method for the evaluation of road infrastructure (“infrastructure module”) The aim of the elaborated method is to evaluate, how much the appropriate transmission (appropriate quality, quantity, location and timing) of information influences road safety at a given section of the road infrastructure. The different attributes of an infrastructure section (e.g traffic volume, number of lanes, etc.) can significantly affect the information perception and processing ability of the driver, hence can strongly influence accident risk. In the first step basing the foundation of the methodology those typical accident situations have been defined where the state of information system could have been played an important role: 1. Collisions with opposing traffic due to entering motorway in irregular direction, when drivers have defined the direction of the traffic probably wrongly; 2. Accidents due to delayed decisions related to exiting a motorway; 3. Accidents due to relative speeding, when drivers have chosen their speed wrongly; 4. Accidents due to temporary traffic signs; 5. Collisions with opposing traffic after exiting highway, when drivers have define the road category wrongly or missed to recognize the end of motorway; 6. Accidents due to crossing different traffic direction. After this, those infrastructure attributes have been defined which can significantly influence the spatial or temporal conformance of the information transmission process: • • • • •

Average daily traffic (T), Sum of heavy good vehicle traffic (H), Road category (R), Curve radius (C), Number of lanes (L).

The investigated infrastructure attributes were required to be available in the national road database. The investigated geographical information system of the road network have had to be composed of homogenous sections in regard the investigated attributes. To simplify the process of parameterization, the considered infrastructure attributes have been represented by unified scales (scales are increasing quantitatively and corresponding to existing coding in case of the road category) (Table 1). Representing scale values (from 1 to 5) of these 5 infrastructure attributes (T; H; R; C; L) can be assigned to every homogenous sections of the infrastructure. In the next step, it has been weighted, how much is the importance of providing appropriate quality information in case of the typical accident situations influenced by road attributes (e.g. it is probably more important to give the right information, in the right place at the right time on a road section with high-traffic volume, than on a section with low-traffic). 10 points have been distributed between the road attributes in proportion to the rate of the influence, in case of every accident situations (Table 2.). Beside the rate, the direction of influence has also been identified and indicated in Table 2. The bold numbers indicate positive correlation: these are the cases when higher scale values of the road attributes mean increased risk of the occurrence of the defined accident types (e.g. the probability of missing to exit the motorway is greater on a road section with larger number of lanes, as well as on sections with high-traffic volume). Italic numbers indicate the opposite effect: in this cases, higher scale values of the road attributes means a lower risk in the accident situations (e.g. higher traffic volume decreases the probability of entering a motorway in irregular direction as the driver can see the right traffic behaviour around himself).

715

Árpád Török et al. / Procedia Engineering 187 (2017) 712 – 721 Table 1. Scales of the infrastructure attributes. Infrastructure attributes

Real value

Representing scale value

Average daily traffic (T)

0 Vehicle/day – 3000 Vehicle/day

1

3001 Vehicle/day – 6000 Vehicle/day

2

6001 Vehicle/day – 10000 Vehicle/day

3

Sum of heavy good vehicle traffic (H)

Road category (R)

Curve radius (C)

Number of lanes (L)

10000 Vehicle/day – 20000 Vehicle/day

4

20001 Vehicle/day –

5

0 Vehicle/day – 700 Vehicle/day

1

701 Vehicle/day – 1400 Vehicle/day

2

1401 Vehicle/day – 2300 Vehicle/day

3

2301 Vehicle/day – 4500 Vehicle/day

4

4501 Vehicle/day –

5

Motorway

1

Highway

2

Primary main road

3

Secondary main road

4

Other roads

5

0 m – 120 m

1

121 m – 450 m

2

451m – 720 m

3

721 m – 1100 m

4

1101 m –

5

1

1

2

2

3

3

4

4

5 or more

5

Table 2. Rate and direction of influencing effect of information. Typical accident situations/Infrastructure attributes

a

b

c

d

e

f

T

7

4

-

3

2

2

H



3

-

1

2



R





5

2

2

4

C

3



5

1

2

3

L



3



3

2

1

Sections of the infrastructure are to be evaluated according to every attribute in case of every accident types defined; therefore 6 evaluating functions (with 5 factors) have been formulated as follows:

f a = (6 − T )aT ⋅ H a H ⋅ R a R ⋅ (6 − C )aC ⋅ La L ,

(1)

f b = T bT ⋅ H b H ⋅ R b R ⋅ C bC ⋅ LbL ,

(2)

f c = T cT ⋅ H c H ⋅ R c R ⋅ (6 − C )cC ⋅ Lc L ,

(3)

716

Árpád Török et al. / Procedia Engineering 187 (2017) 712 – 721

f d = T d T ⋅ H d H ⋅ (6 − R )d R ⋅ (6 − C )d C ⋅ (6 − L )d L ,

(4)

f e = (6 − T )eT ⋅ (6 − H )e H ⋅ (6 − R )e R ⋅ (6 − C )eC ⋅ Le L ,

(5)

ff =T

fT

⋅H

fH

⋅ (6 − R ) f R ⋅ (6 − C ) f C ⋅ L f L .

(6)

Evaluating functions have been separated according to the defined accident situations. Functions take influencing effects of infrastructure attributes into account by their exponents. Values of exponents (aT, aH, …, bT, …, fL) are indicated in Table 2. Bases of the multiplier factors are the representing scale values of the road infrastructure attributes. Absolute value of the scale values have been used in the formulas in case of positive correlations, while negative correlations have been taken into account by subtracting scale values of a constant 6 value (which value is the sum of the maximum and the minimum of the representing scale). Namely, if the scale value representing the average daily traffic (T) is 1 (which is the lowest possible value of T, and used in case of less than 3001 Vehicle/day) at a given section of the infrastructure, then the value of the basis of the first multiplier factor in function fa is 5 (5 = 6 − 1, which is the highest possible value of T); that is because the risk of entering a motorway in irregular direction is higher in case of lower traffic volumes (negative correlation). Evaluating indicator (F) of the investigated infrastructure section is finally formulated by the summation of the results of the 6 evaluating functions as follows:

F i = f ai + f bi + f ci + f di + f ei + f ei ,

(7)

where i = the ID of the investigated infrastructure section. Consequently, our method is applicable for comparing different road sections by a simple evaluating indicator (Fi), which takes various aspects (related to the importance of providing appropriate quality information) into account. The method has been implemented on the Hungarian road network. Results have been illustrated on the following map (Fig. 1). Sections of the road infrastructure have been coloured by the value of their evaluating indicator applying a ten degree scale. Red colour indicates infrastructure elements where the importance of providing appropriate information (quality information in the right place at the right time) is the highest, while the less critical road sections are marked by green colour.

Fig. 1. Evaluation of the Hungarian road network (related to the importance of providing appropriate quality information).

Árpád Török et al. / Procedia Engineering 187 (2017) 712 – 721

The histogram in Fig. 2. facilitates the interpretation of the map. Heights of the column indicate the number of infrastructure sections within a given range of the values of the evaluating indicators. Critical infrastructure elements are located in the right side of the histogram. 30% of the investigated road sections are located in the two most critical categories.

Fig. 2. Distribution of the investigated sections of the Hungarian road network (related to the importance of providing appropriate quality information).

3. Method for the characterization of road accidents (“accident module”) The aim of the elaborated method is to define those types of accidents, which are remarkably influenced by the appropriateness of information transmission. For this reason the mostly affected accident types and causes have been evaluated based on the impact of the adequateness of information transmission on the given type and cause. The numerical evaluation has been based on the following steps. 1. Accident situations (j) related to problems routed back to information transmission have been defined. These defined situations are the same as applied in infrastructure evaluation methodology, only encoding has been changed in the following way: a = 1; b = 2; c = 3; d = 4; e = 5; f = 6. 2. Those types and causes of accidents have been defined, which are remarkably influenced by the appropriateness of information transmission: • Investigated accident types (t): – Collision among vehicles moving in the same direction (1) – Collision among vehicles moving in the opposite direction (2) – Collision among vehicles moving in crossing direction (3) – Single-vehicle accident (4) – Collision among road and railway vehicles (5) – Pedestrian hit (6) – Collisions in roundabouts (7) • Investigated accident causes (o): – Speeding (8) – Irregular over-taking(9) – Omission of giving the right-of-way (10) – Failing to change direction, turn (11) – Deficient indication of dangerous sites (12) – Error of traffic signs (13)

717

718

Árpád Török et al. / Procedia Engineering 187 (2017) 712 – 721

1. Relative occurrence rate of the investigated accident situations has been estimated related to all accident types (Vjt) and accident causes (Vjo). 2. Rate of accidents happened in the investigated accident situations compared to the total number of accidents happened in the given accident type or cause group has been estimated related to all accident types (Wt) and accident causes (Wo). 3. The role of information transmission as a safety influencing factor have been evaluated applying an evaluation scale from 1 to 5 related to all accident types (Pjt) and accident causes (Pjo). After the determination of rates and evaluation values the evaluation parameters of the accident types (Gt) and accident causes (Go) have been defined according to the following formulas:

Gt = Wt ⋅ ∑ j =1 Pt j ⋅ Vt j ,

(8)

Go = Wo ⋅ ∑ j =1 Poj ⋅ Voj ,

(9)

6

6

where t = 1... 7 is the ID of accident types; o = 8... 13 is the ID of accident causes; j = 1… 6 is the ID of the investigated accident situations. Considering that each accident in the accident database has an unambiguous accident type and cause code, it is possible to order the below presented evaluation parameter (Gi) to each accident:

G i = Gti ⋅ Goi ,

(10)

where i = the ID of the investigated accident, and the value of t and o are unambiguously defined for every accidents. The provided Gi value indicates the possible impact of information system for each accident based on their type and cause, taking into account the investigated accident situations. The above introduced estimations have been provided by a professional working group composed of experts from the field of road safety, infrastructure management and from the vehicle technology R&D sector, the results are presented in Tables 3–4. Table 3. Evaluation of the investigated accident types. Accident types (t)

Value of the evaluation parameter of the accident types (Gt)

Collision among vehicles moving in the same direction (1)

1.075

Collision among vehicles moving in the opposite direction (2)

1.280

Collision among vehicles moving in crossing direction (3)

1.890

Single-vehicle accident (4)

1.400

Collision among road and railway vehicles (5)

0.730

Pedestrian hit (6)

1.160

Collisions in roundabouts (7)

1.650

According to the provided results, the categories of “collision among vehicles moving in crossing direction (3)” and “omission of giving the right-of-way (10)” are influenced in the greatest extent by the appropriateness of the information system except those categories which can unambiguously be routed back to the problem of the information system. The following map (Fig. 3.) presents the value of evaluation parameter (Gi) in case of each accident applying a ten degree scale. Red colour indicates accidents influenced by the information system in an extremely great extent and green colour represents accidents influenced only marginally by the information system.

Árpád Török et al. / Procedia Engineering 187 (2017) 712 – 721 Table 4. Evaluation of the investigated accident causes Accident causes (o)

Value of the evaluation parameter of the accident causes (Go)

Speeding (8)

2.880

Irregular over-taking(9)

0.823

Omission of giving the right-of-way (10)

3.050

Failing to change direction, turn (11)

1.200

Deficient indication of dangerous sites (12)

5*

Error of traffic signs (13)

5*

*Indicated groups refer to categories strongly or totally related to the information system, so maximum evaluation value can automatically applied.

Fig. 3. Evaluation of Hungarian road accidents (related to the role of information system).

Fig. 4. Distribution of Hungarian road accidents related to the role of information system.

719

720

Árpád Török et al. / Procedia Engineering 187 (2017) 712 – 721

The histogram in Fig. 4. facilitates the interpretation of the map. Heights of the column indicate the number of accidents. Accidents influenced by the information system in an extremely great extent are located in the right side of the histogram. The number of them is relatively high. 10 % of the investigated accidents are located in the two most critical categories. 4. Conclusions The paper aimed to analyse the quality and accordance of information come from the environment and the role of information in road accidents. Investigation of the complete transport system (instead of a specific situation) and certain accidents have been provided. New approach related to impact of information transmission related to traffic safety has been introduced. To achieve this, two approaches have been applied: a method to characterize the infrastructure (“infrastructure module”), as well as a method to identify critical types of accidents (“accident module”) has been elaborated. Since the different attributes of the infrastructure (traffic volume, number of lanes, etc.) can significantly affect the information perception and processing ability of the driver, and strongly influence accident risk, the aim of the first pillar of the elaborated method is to evaluate, how much the appropriate transmission (appropriate quality, quantity, location and timing) of information influences road safety at a given section of the road infrastructure. Critical sections of the Hungarian national road network have been presented on a map (Fig. 1). The aim of the second pillar of the elaborated method is to define those types and causes of accidents, which are remarkably influenced by the appropriateness of information transmission. For this reason the mostly affected accident types and causes have been evaluated based on the impact of the adequateness of information transmission on the given type and cause. According to the provided results, the categories of “collision among vehicles moving in crossing direction (3)” and “omission of giving the right-of-way (10)” are influenced in the greatest extent by the appropriateness of the information system. Locations of accidents of the critical accident types have also been presented on a map (Fig. 3). The definition of accidents and infrastructure sections mostly influenced by the information system makes it possible to provide general conclusion regarding the relationship between the accident risk and the information system. Critical road sections, where appropriate transmission of information influences road safety the most, as well as critical accident types that are highly influenced by information have been identified based on the elaborated method. With the comparison of these two data sets, accidents that have been influenced the most likely by the condition (quality, quantity, location and timing) of the information system are to be revealed. As one of the most important results of the project - the research managed to demonstrate that almost 10% of the accidents is strongly related to information transmission processes. Beside of this, the importance of information transmission is estimated to be critical on the 30 % of the investigated road sections. Our future research will focus on in-depth analysis of these accidents and infrastructure sections to reveal connections between road accidents and information flow as well as deficiencies of information systems of the national road network.

References [1] E. Dr Legeza, T. Berta, Zs. Hamza, Közúti információ szerepe és megfelelőségi vizsgálata (In Hungarian), KÖZLEKEDÉSTUDOMÁNYI SZEMLE 63(6) (2013) 65−74. [2] C. E. Shannon, W. Weaver, A Mathematical Model of Communication. Urbana, IL: University of Illinois Press. 1949. [3] A. B. Giczi, Az osztályozás és a káoszelmélet − A rendszer-, a káosz-, az osztályozás-, és az információelmélet találkozása a tudományok dobozai között (In Hungarian), Könyvtári Figyelő 50:(1) (2004). [4] Available from Internet: http://einstein.informatik.uni-oldenburg.de/rechnernetze/informat.htm [5] N. Wiener, Cybernetics- or the Control and Communication in the Animal and the Machine, second edition. (ISBN) 0-262-23007-0 (1985) 132. [6] W. Barfield, T. A. Dingus, Human Factors In Intelligent Transportation Systems, Psychology Press, New York, (ISBN-13) 978-1-315-806624 (2014). [7] B. Hamidi, N. Lajqi, L. Hamidi, Modelling and Sensitive Analysis of the Impact on Telematics System in Vehicles, IFAC-PapersOnLine 49(29) (2016) 232−236. doi: http://dx.doi.org/10.1016/j.ifacol.2016.11.056

Árpád Török et al. / Procedia Engineering 187 (2017) 712 – 721 [8] V. Gitelman, F. Pesahov, R. Carmel, S. Bekhor, The Identification of Infrastructure Characteristics Influencing Travel Speeds on Singlecarriageway Roads to Promote Self-explaining Roads, Transportation Research Procedia 14 (2016) 4160−4169. doi: http://dx.doi.org/10.1016/j.trpro.2016.05.387 [9] T. Berta, Az ember, mint a közlekedési rendszer része (In Hungarian), Közúti és Mélyépítési Szemle (ISSN) 57(12) (2007) 20–25. [10] Á. Dudás, T. Berta, Megoldási lehetőségek egyes közlekedési feltételek biztonságosabb és hatékonyabb kezelésére kritikus közúti pontokon (In Hungarian), Közúti és Mélyépítési Szemle (ISSN) 54(1) (2004) 15–17.

721