This article was downloaded by: [National Pingtung University of Science and Technology]

On: 13 January 2015, At: 00:25

Publisher: Taylor & Francis
Informa Ltd Registered in England and Wales Registered Number: 1072954 Registered office: Mortimer House,
37-41 Mortimer Street, London W1T 3JH, UK

Traffic Injury Prevention

Publication details, including instructions for authors and subscription information:
http://www.tandfonline.com/loi/gcpi20

A Method for Determining the Number of Traffic

Conflict Points Between Vehicles at Major–Minor
Highway Intersections
a a b c a
Fuquan Pan , Lixia Zhang , Jian Lu , Jiguang James Zhao & Fengyuan Wang
a
School of Automobile and Traffic , Qingdao Technological University , Qingdao , Shandong ,
P. R. China
b
School of Naval Architecture, Ocean and Civil Engineering , Shanghai Jiaotong University ,
Shanghai , P. R. China
c
Department of Civil Engineering , Southern Illinois University Edwardsville , Edwardsville ,
Illinois
Accepted author version posted online: 22 Aug 2012.Published online: 26 Mar 2013.

To cite this article: Fuquan Pan , Lixia Zhang , Jian Lu , Jiguang James Zhao & Fengyuan Wang (2013) A Method for
Determining the Number of Traffic Conflict Points Between Vehicles at Major–Minor Highway Intersections, Traffic Injury
Prevention, 14:4, 424-433, DOI: 10.1080/15389588.2012.713148

To link to this article: http://dx.doi.org/10.1080/15389588.2012.713148

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the “Content”) contained
in the publications on our platform. However, Taylor & Francis, our agents, and our licensors make no
representations or warranties whatsoever as to the accuracy, completeness, or suitability for any purpose of the
Content. Any opinions and views expressed in this publication are the opinions and views of the authors, and
are not the views of or endorsed by Taylor & Francis. The accuracy of the Content should not be relied upon and
should be independently verified with primary sources of information. Taylor and Francis shall not be liable for
any losses, actions, claims, proceedings, demands, costs, expenses, damages, and other liabilities whatsoever
or howsoever caused arising directly or indirectly in connection with, in relation to or arising out of the use of
the Content.

This article may be used for research, teaching, and private study purposes. Any substantial or systematic
reproduction, redistribution, reselling, loan, sub-licensing, systematic supply, or distribution in any
form to anyone is expressly forbidden. Terms & Conditions of access and use can be found at http://
www.tandfonline.com/page/terms-and-conditions
 Traffic Injury Prevention (2013) 14, 424–433
Copyright  C Taylor & Francis Group, LLC

ISSN: 1538-9588 print / 1538-957X online

DOI: 10.1080/15389588.2012.713148

A Method for Determining the Number of Traffic Conflict

Points Between Vehicles at Major–Minor Highway
Intersections
FUQUAN PAN1, LIXIA ZHANG1, JIAN LU2, JIGUANG JAMES ZHAO3, and FENGYUAN WANG1
Downloaded by [National Pingtung University of Science and Technology] at 00:25 13 January 2015

1
School of Automobile and Traffic, Qingdao Technological University, Qingdao, Shandong, P. R. China
2
School of Naval Architecture, Ocean and Civil Engineering, Shanghai Jiaotong University, Shanghai, P. R. China
3
Department of Civil Engineering, Southern Illinois University Edwardsville, Edwardsville, Illinois

Received 18 January 2012, Accepted 11 July 2012

Background: Traffic conflict points are the areas formed by conflict movements in the traffic flow. The numbers and distributions
of different types of traffic conflict points are often used to evaluate access management designs and the safety performance of
intersections. Customarily, determining the number of different types of traffic conflict points is based on the manual drawing
method, which is troublesome and makes using computerized procedures for calculation of the number of traffic conflict points
difficult.
Objective: The present work aims to quickly compute the number of traffic conflict points between vehicles at major–minor highway
intersections.
Methods: The concept of “virtual lanes” is defined, and the virtual lane conversion method is proposed. Based on actual and
virtual lanes, vehicular movements, and traffic control modes, a generalized method to determine the number of traffic conflict points
between vehicles at major–minor highway intersections is developed.
Results: A procedure for determining the number of conflict points is presented. A similar HCM LOS-style form and computer
software were designed to help calculate the number of conflict points.
Conclusions: Applications show that the method can efficiently and accurately determine the number of traffic conflict points
between vehicles at major–minor highway intersections.
Keywords: traffic conflict points, virtual lanes, conflict movements, highway intersections

Introduction
Compared to highway segments, intersections are more likely
to be crash-prone areas due to the existence of many traffic
conflict points. A traffic conflict is a traffic event involving the
interaction of 2 or more road users, usually motor vehicles,
where one or both drivers take evasive actions, such as brak-
ing or swerving, to avoid a collision (Parker and Zegeer 1989).
A traffic conflict point is the point where traffic flow (move-
ment) diverges or 2 traffic flows (movements) cross or merge.
Thus, the concepts behind these 2 terms differ. In accordance
with traffic rules, traffic flow should be in the designated or
approved lanes. Thus, from the view of the traffic flow lane, a
traffic conflict point is the location of traffic lane divergence
or where 2 or more traffic lanes cross or merge. In theory, traf-
fic conflict points are the spots where traffic conflicts occur.
However, due to the complexity of actual traffic operations,
Address correspondence to Fuquan Pan, School of Automobile
and Traffic, Qingdao Technological University, No. 777 Jial-
ingjiang East Road, Huangdao District, Qingdao, Shandong
266520, China. E-mail: panfuquan@gmail.com
the place where actual traffic conflicts occur may not only
be traffic conflict points. However, as long as traffic conflict
points are created, traffic conflicts may occur. Studies in the
United States showed that for a cross (4-way) intersection
with 16 crossing conflict points, 8 merge conflict points, and
8 diverge conflict points between vehicles, the average crash
rate is 0.4 crashes per million vehicles. After introducing some
improvements with access management techniques by pro-
hibiting left-turn maneuvers from the minor road, however,
the number of conflict points was reduced to only 2 crossings,
4 merge conflict points, and 4 diverge conflict points, and the
average crash rate dropped to 0.2 crashes per million vehicles
(Preston and Albrecht 2001). Obviously, safety performance
at cross-intersections is highly correlated with the number of
conflict points and can be directly improved by eliminating the
number of conflict points.
Conflict points and their corresponding severity character-
istics are widely used in evaluating the safety performance of
intersections. Glauz and Migletz (1980) evaluated the safety
performance of intersections based on conflict points between
vehicles. Stone et al. (2002) also estimated pedestrian safety in
roundabouts by considering the conflict points between vehi-
cles and between vehicles and pedestrians. Determining traffic
 Traffic Conflict Points
conflict points is usually a prerequisite in evaluating the safety
performance of an intersection. For example, the concept of
“intersection level of safety service” proposed by Pan et al.
(2008) and Lu et al. (2008) was based on the exact number of
conflict points and their corresponding severities at intersec-
tions. Most improvements in intersection safety merely reduce
the amount and/or severity of the conflict points or simply
separate them from each other with access management tech-
niques (Access Management Committee 2003; Maze et al.
2000; Pan et al. 2011).
The current article is mainly focused on major–minor high-
Downloaded by [National Pingtung University of Science and Technology] at 00:25 13 January 2015
way intersections, which are the most common type of inter-
sections in rural/suburban areas in China and some other
countries where major roads (arterials, collectors) intersect
with minor roads (collectors, local roads). To promote safety
performance at major–minor intersections, determining dif-
ferent types of conflict points, evaluating the level of safety
service, and proposing proper improvements to the intersec-
tions are critical.
This article aims to develop a new method to determine
the number of traffic conflict points between vehicles at
major–minor intersections. The method can be applied to
both T-shaped (3-way) and cross-shaped (4-way) intersec-
tions, as well as 2-way stop/yield sign intersections, 2-phase
signalized intersections, and 3-phase signalized intersections.
In addition, this method aims to be adapted by 2-way, 2-
lane T, and cross-intersections, as well as by complex T and
cross-intersections with more than 2 lanes. Applications of
roundabouts are rare in intersections of major–minor high-
ways; thus, this method does not consider roundabouts. In the
present study, the major road was in the north–south direction
and the minor road was in the west–east direction.
The traffic conflict point can be determined with the tra-
ditional manual drawing method. In this method, the tra-
jectories of vehicular movements are drawn on the intersec-
tion configuration picture, and the traffic conflict points—that
is, where 2 different vehicular movements share the right of
way—can be identified directly. The drawing method can be
used in any intersection type under any traffic control. Its lim-
itation is that the traffic conflict points can only be determined
manually, and the process is troublesome.
To overcome the shortcomings of the drawing method, Zhai
(1992) and Zhu (1996) explored a similar method to calcu-
late the number of traffic conflict points between vehicles.
However, their respective methods are suitable only for sim-
ple 2-way, 2-lane stop/yield control intersections. Presently, no
method is able to compute the overall number of traffic conflict
points in all categories. Existing methods determine the num-
ber of traffic conflict points only for simple intersections. For
2-lane T and cross-intersections under stop/yield traffic con-
trol signs, the number of crossing conflict points, merge con-
flict points, and diverge conflict points between vehicles can be
determined using Eqs. (1) and (2) (Zhai et al. 1992; Zhu 1996).
NCC = n 2 (n − 1)(n − 2)/6 (1)
NMC = NDC = n(n − 2), (2)
where NCC , NMC , and NDC represent the number of crossing
conflict points, merge conflict points, and diverge conflict
425
points, respectively, and n is the number of legs at the
intersection (n = 3 or n = 4). However, these equations
cannot calculate the number of traffic conflict points between
vehicles at highway intersections with more than 2 lanes
under stop/yield traffic control or signalized traffic control.
At the Transportation Research Board (TRB) annual meet-
ing, Lu et al. (2011) explored a calculation procedure for the
number of conflict points at unsignalized intersections. Al-
though the subject is similar, there are many differences be-
tween this article and Lu et al. (2011): (1) Lu et al. (2011)
used the concept of intersection quadrants, whereas this ar-
ticle does not and is able to compute the number of conflict
points. (2) In this article, traffic conflict movements and lane
configurations are used to determine the number of conflict
points; Lu et al. (2011) did not use traffic conflict movement.
(3) Lu et al. (2011) were only concerned with unsignalized in-
tersections, but this article provides computational models for
2-way stop/yield intersections and signalized intersections. (4)
In this article, a calculation table similar to style of HCM LOS
(Highway Capacity Manual Level of Service) is designed to
help engineers use the proposed method. (5) In this article,
a computer program was developed; Lu et al. (2011) did not
develop a computer program.
Method
Basic Assumptions
The proposed method is based on normal intersection lane
settings and prevailing traffic rules. For clarity, the following
assumptions are explained.
The assumptions of entrance approach and exit approach
at an intersection:
1. Lane setting is not restricted; that is, exclusive lanes and
shared lanes can be configured.
2. Except for the exclusive left lanes or exclusive right lanes at
the entrance approach, the number of lanes of an entrance
approach equals the number of lanes of the corresponding
exit approach. Otherwise, the lanes become inconsistent
with the traffic rules. The number of lanes at an entrance
approach is not limited.
When a vehicle enters an intersection and passes the in-
tersection, compliance with the following traffic rules is as-
sumed:
1. Although the lane line is not marked in the inside of an in-
tersection, the passing vehicle is not allowed to change its
direction to the former lane; that is, the vehicle should pro-
ceed toward the lane in front of the former lane. Otherwise,
the driver is in violation of traffic rules.
2. Vehicles in the exclusive left (right) lane or those with left
(right)-turn right of way on the shared lane can only go
to the inside (outside) lane of the corresponding exit ap-
proach. If there are 2 lanes with left (right) turns, vehi-
cles can only turn left (right) to the corresponding inside
(outside) 2 lanes of the corresponding exit approach. The
rest may be deduced by analogy. Noncompliance equates
to violation of traffic rules.
 426
Definition of Virtual Lanes
In this study, traffic conflict points is used in the following con-
texts. To help determine the number of traffic conflict points
at intersections, virtual lanes are defined. Vacek et al. (2007),
Stiller et al. (2008), Fujii and Yoshimura (2010), and Minh
et al. (2012) applied virtual lanes in a simulation framework for
cognitive automobiles, an autonomous system, vehicle emis-
sions, and maneuvers of motorcycles in queues, respectively.
However, the concept of virtual lanes in those applications
is different from the definition of virtual lanes in this arti-
Downloaded by [National Pingtung University of Science and Technology] at 00:25 13 January 2015
cle. In the present research, virtual lanes refers to traffic flow
paths from an entering lane at an intersection, such as through
movement, left-turn, and right-turn paths (traffic movements
run in the corresponding lanes). For instance, if an entering
lane allows traffic to make a through movement and right-
turn movement, this lane is considered to have 2 virtual lanes
leaving the entering lane after the stop line. It is defined as a
virtual lane because it is not often marked on the pavement
and does not actually exist.
A maximum of 3 virtual lanes can be derived from one
actual entering lane. For example, 2 virtual lanes, one for the
through and one for left-turn movement, could be derived
from a lane shared by through and left-turn movements. An
exclusive lane can only be converted into an exclusive virtual
lane. In general, the number of virtual lanes resulting from an
actual lane is equal to the number of allowed traffic movement
paths in this lane. Thus, the number of virtual lanes at an
intersection should be greater than, or at least equal to, the
number of actual lanes.
Virtual Lane Conversion
For the given approach, the number of virtual lanes can be
determined using the following methods:
1. The number of virtual lanes for through movements is
equal to the number of entering lanes with through move-
ments; that is, the number of lanes with exclusive through,
left/through/right, left/through, and through/right move-
ments.
2. The number of virtual lanes for left-turn movements
is equal to the number of entering lanes with left-turn
movements; that is, the number of lanes with exclusive
left-turn, left/through/right, left/through, and left/right
movements.
3. The number of virtual lanes for right-turn movements is
equal to the number of entering lanes with right-turn move-
ments; that is, the number of lanes with exclusive right-turn,
through/right, left/through/right, and left/right move-
ments.
Traffic Control Mode
The following 3 types of intersections under different traffic
control modes are discussed in this article:
1. Two-way stop/yield sign intersection: With stop/yield
signs installed at minor road approaches, all movements
from the minor road must stop/yield to those from the
major road.
Pan et al.
2. Two-phase signalized intersection: Phase 1 is on the major
road, where the through and right-turn movements are pro-
tected and left-turn movements are permitted; phase 2 is
on the minor road, where the through and right-turn move-
ments are protected, and left-turn movements are permit-
ted. Right-turn-on-red (RTOR) movements are permitted
all the time.
3. Three-phase signalized intersection: Phase 1 is on the ma-
jor road, where through movements are protected; Phase
2 is on the major road, where left-turn movements are
protected; Phase 3 is on the minor road, where through
movements are protected and left-turn movements are per-
mitted. RTOR movements are permitted all the time.
In theory, 4-phase signalized intersections have no conflict
points. Thus, this article will not discuss the 4-phase control
mode.
Conflict Movements and Right of Way
Traffic conflict occurs between 2 or more traffic movements
that share the same space during maneuvers when one move-
ment crosses, merges into, or diverges from another movement.
The traffic conflict point is located where the trajectories of
the 2 conflict movements intersect. Typically, conflict move-
ments can be classified into crossing, merge, and diverge con-
flict movements, as shown in Figure 1. Not all of the conflict
movements in Figure 1 occur under the given traffic control
mode because not all the traffic movements would be given
right of way under the special traffic control mode. As an ex-
ample, no through movements from major roads and minor
roads would be allowed at the same time for 2-phase signalized
intersections. Therefore, for 2-phase signalized intersections,
no crossing conflict movements exist between through move-
ments from major and minor roads. For 2-way stop/yield sign
intersections, the through movements from major and minor
roads are allowed at the same time, thereby creating crossing
conflict movements between them.
The methodology proposed is mainly based on the traffic
control mode at intersections. Altogether, 3 types of inter-
sections are defined, namely, the 2-way stop/yield sign inter-
section, the 2-phase signalized intersection, and the 3-phase
signalized intersection. In most cases, conflict movements can
be determined using the method developed for the particular
traffic control mode, considering the relationship to the right
of way. However, other irregular intersections are under spe-
cial traffic control modes. In these cases, conflict movements
should be determined directly based on the right of way con-
ditions in these intersections.
Number of Traffic Conflict Points
The conflict point category is based on the category of the
conflict movements; that is, crossing conflict points, merge
conflict points, and diverge conflict points.
Crossing Conflict Points
Crossing conflicts occur between conflict movements that
come from different entrance approaches and go toward dif-
ferent exit approaches. Examples of crossing conflicts include
 Traffic Conflict Points
Downloaded by [National Pingtung University of Science and Technology] at 00:25 13 January 2015
Fig. 1. Conflict movements at a cross-intersection: (a) crossing
conflict movements; (b) merge conflict movements; and (c) diverge
conflict movements.
the conflict between through movements and opposite left-
turn movements at signalized intersections and the conflict
between through movements from major and minor roads at
2-way stop/yield sign intersections. The characteristic of the
crossing conflict is that one movement needs to cross the tra-
jectory of the other movement completely to complete the
maneuver and vice versa.
427
(a)
(b)
(c)
Fig. 2. Crossing conflict movements at intersections: (a) 3-phase
signalized intersection; (b) 2-phase signalized intersection; and
(c) 2-way stop/yield sign intersection.
Figure 1a shows all of the possible crossing conflicts at
an intersection. For given intersections, only those conflict
movements in Figure 1a that have the right of way at the same
time are real conflict movements. The total number of crossing
conflict points at the intersection is the sum of the conflict
points for each individual pair of crossing conflict movements.
The number of crossing conflict points for each individual pair
of crossing conflict movements is the product of the number
of virtual lanes for the 2 crossing conflict movements. The
number of crossing conflict points can be calculated using
Eqs. (3) and (4).
NCCi j = NVLi × NVLj , (3)

NCC = NCCi j , (4)
where NCC is the number of conflict points at the intersection,
NCCij is the number of crossing conflict points between con-
flict movements i and j, NVLi is the number of virtual lanes
for movement i, and NVLj is the number of virtual lanes for
movement j.
Merge Conflict Points
Merge conflict points occur in the exit approach where 2 or
more movements merge. Four situations can form merge con-
flict points. (1) Merge conflicts occur in the following sequence:
the left-turn and through movements merge first, followed by
their merging with the right-turn movements, as shown in
Figure 1b. (2) Through movement and right-turn movement
merge to form the merge conflict point. (3) Through move-
ment and left-turn movement form the merge conflict point.
(4) Left-turn movement and right-turn movement merge to
form the merge conflict point, and this situation is applica-
ble for T intersections and the protected left-turn movements
phase in signalized cross-intersections. For signalized intersec-
 428
(a)
Downloaded by [National Pingtung University of Science and Technology] at 00:25 13 January 2015
(b)
(c)
Fig. 3. Merge conflict movements at intersections: (a) 3-phase
signalized intersection; (b) 2-phase signalized intersection; and
(c) 2-way stop/yield sign intersection.
tions, some merge conflicts do not exist because of protected
left-turn movements. In fact, the merge conflict point is lo-
cated where the trajectories of 2 merge conflict movements
intersect.
The number of merge conflict points at an intersection is
the sum of merge conflict points at all exit approaches. The
number of merge conflict points at an exit approach is equal to
the number of lanes that accommodate 2 or more movements.
NMCi = Max (NVLLi + NVTLi + NV RLi − NDLi , 0) (5)

NMC = NMCi , (6)
where NMC is the number of merging conflict points at the
intersection, NMCi is the number of merge conflict points
at the exit approach i, NVLLi is the number of virtual left-
turn lanes that enter the exit approach i, NVTLi is the num-
ber of virtual through lanes that enter the exit approach i,
NVRLi is the number of virtual right-turn lanes that enter ap-
proach i, and NDLi is the number of departure lanes in the exit
approach i.
Diverge Conflict Points
Diverge conflict points occur at the entrance approach when a
lane accommodates 2 or more movements. The diverge conflict
occurs in the following sequence: first, right-turn movements
from the entire traffic, followed by left-turn movements from
the remaining traffic, as shown in Figure 1c. For signalized
intersections with protected left-turn movements, only diverge
conflicts occur between through and right-turn movements.
Exclusive lanes have no diverge conflict points. Diverge conflict
points are found where 2 or more movements diverge.
For an intersection, the number of diverge conflict points is
the sum of diverge conflict points at all entrance approaches.
Pan et al.
(a)
(b)
(c)
Fig. 4. Diverge conflict movements at intersections: (a) 3-phase
signalized intersection; (b) 2-phase signalized intersection; and
(c) 2-way stop/yield sign intersection.
The number of diverge conflict points at the entrance approach
i is the sum of lanes that accommodate 2 or more movements,
as shown in Eqs. (7) and (8):
NDCi = NLLTi + NLLRi + NLT Ri + 2NLLT Ri (7)

NDC = NDCi , (8)
where NDC is the number of diverge conflict points at the in-
tersection, NDCi is the number of diverge conflict points at the
entrance approach i, NLLTi is the number of lanes shared by
left-turn and through movements at the entrance approach i,
NLTRi is the number of lanes shared by through and right-turn
movements at the entrance approach i, NLLRi is the number of
lanes shared by left-turn and right-turn movements at the en-
trance approach i, and NLLTRi is the number of lanes shared by
left-turn, through, and right-turn movements at the entrance
approach i.
Steps in Determining Conflict Points
Based on the methodology proposed above, the conflict points
at major–minor intersections can be determined with the fol-
lowing steps.
Step 1: Collect information on the intersection, such as inter-
section shape, traffic control mode, and lane configurations,
among others.
Step 2: Determine the number of virtual lanes at each entrance
approach and the number of lanes at each exit approach.
 Traffic Conflict Points 429

Table 1. Form for determining the number of conflict points at an intersection

Intersection Major Minor Date Analyst

road road

Step1 Intersection geometric information

Traffic × Signalized Unsignalized

Downloaded by [National Pingtung University of Science and Technology] at 00:25 13 January 2015

control × Two-phase Three-phase Stop control Yield control

mode

Intersection shape T intersection × Cross-intersection

Lane Northbound Southbound Westbound Eastbound

configuration NLLN NLTN NLRN NLLTN NLTRN NLLRN NLLTRN NLLS NLTS NLRS NLLTS NLTRS NLLRS NLLTRS NLLW NLTW NLRW NLLTW NLTRW NLLRW NLLTRW NLLE NLTE NLRE NLLTE NLTRE NLLRW NLLTRE

at

entrance 0 0 0 1 1 0 0 0 0 0 1 1 0 0 0 0 0 0 0 0 1 0 0 0 0 0 0 1

approach

Step 2 Determine the number of virtual lanes

Number Northbound Southbound Westbound Eastbound

of NVLLN = NVLTN = NVLRN = NVLLS = NVLTS = NVLRN = NVLLW = NVLTW = NLTW NVLRW = NVLLE = NVLTE = NVLRE = NLRE

virtual NLLN + NLTN + NLRN + NLLS + NLTS + NLRS + NLLW + + NLLTW + NLRW + NLLE + NLTE + + NLTRE +

lanes at NLLTN + NLLTN + NLTRN NLLTS + NLLTS + NLTRS + NLLTW + NLTRW + NLTRW + NLLTE + NLLTE + NLLRE +

entrance NLLRN + NLTRN + + NLLRN NLLRS + NLTRS + NLLRS + NLLRW + NLLTRW NLLRW + NLLRE + NLTRE + NLLTRE

approach NLLTRN NLLTRN + NLLTRN NLLTRS NLLTRS NLLTRS NLLTRW NLLTRW NLLTRE NLLTRE

1 2 1 1 2 1 1 1 1 1 1 1

Number NDLN (Northbound) NDLS (Southbound) NDLW (Westbound) NDLE (Eastbound)

of lanes at 2 2 1 1

exit

approach

Step 3 Conflict movements

(Continued on next page)

 430 Pan et al.

Table 1. Form for determining the number of conflict points at an intersection (Continued)

Part 1 Part 2 Part 3

Crossing

× × × ×
Downloaded by [National Pingtung University of Science and Technology] at 00:25 13 January 2015

Part 1 Part 2 Part 3

Merging

× × × ×

Part 1 Part 2 Part 3

Diverging

× × × × × ×
× ×

Conflict Three-phase signal control Two-phase signal control Stop/yield control sign

movements Part 1 only Part 1 + part 2 Part 1 + part 2 + part 3

Step 4 Calculation of number of traffic conflict points

Part 1 Part 2 Part 3

Number NVLTENVLLW + NVLTWNVLLE NVLLNNVLTS + NVLLSNVLTN NVLTSNVLTE + NVLTSNVLTW + NVLTNNVLTE + NVLTNNVLTW +

of NVLTSNVLLE + NVLTNNVLLW + NVLTENVLLN + NVLTWNVLLS +

crossing NVLLENVLLN + NVLLWNVLLN + NVLLENVLLS + NVLLWNVLLS

conflict 2 4

points

Number Max(NVLTN + NVLRW NDLN, 0) + Max(NVLTS 0 Max(NVLLS + NVLTE + NVLRN NDLE, 0) + Max(NVLLW + NVLTS

of merge + NVLRE NDLS, 0) + Max(NVLLN + NVLRS + NVLRE NDLS, 0) + Max(NVLLN + NVLTW + NVLRS NDLW, 0) +

conflict NDLW, 0) + Max(NVLLS + NVLRN NDLE, 0) + Max(NVLLE + NVLTN + NVLRW NDLN, 0)

points Max(NVLTW + NVLRS NDLW, 0) + Max(NVLTE

+ NVLRN NDLE, 0) + Max(NVLLE + NVLRW

NDLN, 0) + Max(NVLLW + NVLRE NDLS, 0)

6 0

(Continued on next page)

 Traffic Conflict Points
Table 1. Form for determining the number of conflict points at an intersection (Continued)
Number NLLRN + NLTRN + NLLRS + NLTRS + NLLTW +
of diverge NLLRW + NLTRW + 2NLLTRW + NLLTE + NLLRE +
conflict NLTRE + 2NLLTRE
points 6 2
Conflict Three-phase Crossing conflict points—Part 1 only, Merge conflict points—Part 1 Diverge conflict points—Part 1 Sum =
Downloaded by [National Pingtung University of Science and Technology] at 00:25 13 January 2015
points signal control NCC =
Two-phase signal Crossing conflict points—Part 1
control + part 2, NCC = 6
Stop/yield control Crossing conflict points—Part 1
sign + part 2 + part 3, NCC =
Step 3: List all of the potential conflict movements at the
intersection under the given traffic control mode.
1. Crossing conflict movement: The crossing conflict
movements at intersections under different traffic con-
trol modes are shown in Figure 2 (curves and arrows
represent the traffic flow).
2. Merge conflict movement: The merge conflict move-
ments at intersections under different traffic control
modes are plotted in Figure 3.
3. Diverge conflict movement: The diverge conflict move-
ments at intersections under different traffic control
modes are plotted in Figure 4.
The conflict movements listed above are only for cases
under the general traffic control mode and regular intersec-
tion shape. In special traffic control modes where some of the
movements are prohibited—for example, no RTOR—or some
movements do not occur for irregular intersection shapes; for
example, for through movements from the minor road at a T
intersection, the pairs of conflict movements should be deleted
from the list directly. On the other hand, movements not listed
here are allowed for some irregular intersections as well. For
instance, the southbound left-turn and through movements
are permitted at the same time for some 3-phase T intersec-
tions, because no left-turn movements occur. The reader can
directly determine the conflict movements with the right of
way under the circumstance.
Step 4: Calculate the number of conflict points under each
category.
For convenience, the number of virtual lanes at each en-
trance approach is denoted by N and a series of 4 subscript
characters, where N means number. The first 2 subscripts V
and L represent virtual and lane, respectively. The third is the
movement direction; that is, L for left-turn movement, T for
through movement, and R for right-turn movement. The fourth
431
2NLLTRN + 2NLLTRS + NLLTS +
NLLTN
only, NMC = only, NDC =
Merge conflict points—Part 1, Diverge conflict points—Part 1 Sum = 20
NMC = 6 + part 2, NDC = 8
Merge conflict points—Part 3 Diverge conflict points—Part 1 Sum =
only, NMC = + part 2, NDC =
is the direction of the entrance approach; that is, E, S, W , and
N for eastbound, southbound, westbound, and northbound, re-
spectively. For instance, NVLTE refers to the number of virtual
lanes of through movement for the eastbound direction. Sim-
ilarly, the number of actual entering lanes at each entrance
approach is expressed by N and a series of 3 to 5 subscripts.
The first subscript L represents lane. The second, third, or
fourth subscript denotes the abbreviation of single or several
traffic movements: L for left-turn movement, T for through
movement, R for right-turn movement, LR for left-turn and
right-turn movement, TR for through and right-turn move-
ment, LT for left-turn and through movement, and LTR for
left-turn, through, and right-turn movement. The fifth sub-
script is the direction of the entrance approach, as previously
described. For example, NLLN refers to the number of lanes
with exclusive left-turn movement for northbound. NLLTRW
refers to the number of lanes shared by left-turn, through,
and right-turn movements for westbound. In the same way,
the number of actual departure lanes at each exit approach
is marked by N and 2 subscripts. N means number, the first
subscript D expresses departure, and the second indicates the
direction of the exit approach; that is, E, S, W , and N. For
example, NDLN refers to the number of departure lanes at the
northbound exit approach.
1. Cross conflict points: For the 3-phase signalized inter-
section,
NCC = NVLT E NVLLW + NVLTW NV LLE (9)
where NVLTE NVLLW refers to the variable NVLTE multi-
plied by the variable NVLLW . Others have the same mean-
ing.
For the 2-phase signalized intersection,
NCC = NVLTE NVLLW + NVLTW NVLLE
+ NVLLN NVLTS + NVLLS NVLTN (10)
 432
For the 2-way stop/yield sign intersection,
NCC = NVLTE NVLLW + NVLTW NVLLE
+ NVLLN NVLTS + NVLLS NVLTN
+ NV LTS NVLTE + NV LTS NVLTW
+ NVLTN NVLTE + NVLTN NVLTW
+ NV LTS NVLLE + NVLTN NVLLW
+ NVLTE NVLLN + NVLTW NVLLS
+ NVLLE NVLLN + NVLLW NVLLN
+ NVLLE NVLLS + NVLLW NVLLS (11)
Downloaded by [National Pingtung University of Science and Technology] at 00:25 13 January 2015
2. Merge conflict points: For the 3-phase/2-phase signal-
ized intersection,
NMC = Max(NVLTN +NVLRW − NDLN , 0)
+ Max(NVLTS+NVLRE − NDLS , 0
+ Max(NVLLN +NVLRS − NDLW , 0
+ Max(NVLLS +NVLRN − NDLE , 0)
+ Max(NVLTW +NVLRS − NDLW , 0)
+ Max(NVLTE +NVLRN − NDLE , 0)
+ Max(NVLLE +NVLRW − NDLN , 0)
+ Max(NVLLW + NVLRE − NDLS , 0) (12)
For the 2-way stop/yield sign controlled intersection,
NMC = Max(NVLLS + NVLTE + NVLRN − NDLE , 0)
+Max(NVLLW + NV LTS + NVLRE − NDLS , 0)
+Max(NVLLN + NVLTW + NVLRS − NDLW , 0)
+ Max(NVLLE + NVLTN + NVLRW − NDLN , 0)
(13)
3. Diverge conflict points: For the 3-phase signalized inter-
section,
NDC = NLLRN + NLTRN + NLLRS + NLTRS
+ NLLTW + NLLRW + NLTRW + 2NLLTRW
+ NLLTE + NLLRE + NLTRE + 2NLLTRE (14)
For the 2-phase signalized and 2-way stop/yield sign
intersection,
NDC = NLLRN + NLTRN + 2NLLTRN + NLLRS
+ NLTRS + 2NLLTRS + NLLTW + NLLRW
+ NLTRW + 2NLLTRW + NLLTE + NLLRE
+ NLTRE + 2NLLTRE + NLLTS + NLLTN (15)
Step 5: Determine the total number of conflict points at an
intersection by adding up all of the conflict points calculated
above.
The entire procedure is depicted with a flowchart in
Figure 5.
Illustration
The rationality and correctness of the proposed methodology
were verified in practice, and the methodology was applied
Pan et al.
Intersection geometric
parameter
Traffic
control
mode
Signalized Unsignalized
intersection intersection Lane configuration
Signal phase
Stop/yield sign
location
Intersection shape
Number of virtual lanes at
each entrance approach
Right of way for movements at
different approaches
Conflicting movements with
shared right of way
Number of conflict points under
each category
Total number of conflict points at
the intersection
Fig. 5. Flowchart for determining the number of conflict points
at an intersection.
in some projects on safety performance evaluation and im-
provement in China. In the current study, a 2-phase signalized
cross-intersection was selected to illustrate the methodology.
The major road is a 4-lane highway and the minor road is a
2-lane highway. Figure 6a shows the lane configurations at the
intersection. The right of way at the intersection was allocated
as follows: phase 1, left-turn and through movements on major
road, and phase 2, left-turn and through movements on minor
road. RTOR movements are always permitted.
The conflict points at the intersection were first determined
via the traditional manual drawing method. For the intersec-
tion, a total of 6 crossings, 6 merge, and 8 diverge conflict
points existed, as shown in Figures 6b and 6c. The conflict
points were then determined with the proposed methodology,
as shown in Table 1. The results indicated that the number
of conflict points calculated using this method coincided with
the results derived from the traditional drawing method, which
justified the proposed new method.
An HCM LOS-style form was designed to help calculate the
conflict points with the proposed methodology, as indicated
in Table 1. With this form, the traffic engineers simply need
to encode the intersection parameters to determine the traffic
conflict points at the intersection.
Software Development
Another benefit of the proposed method is that it can be im-
plemented using computer software. Unlike the traditional
manual drawing method, which only works on a case-by-case
basis, the proposed method summarizes the common charac-
teristics of different conflict movements and proposes a general
method for determining conflict points under different traffic
control modes and intersection shapes. This method can be
achieved easily using computer software. The authors have
 Traffic Conflict Points
Downloaded by [National Pingtung University of Science and Technology] at 00:25 13 January 2015
Fig. 6. Lane configurations and conflict points at a 2-phase cross-
intersection: (a) lane configurations at intersection; (b) conflict
points at phase 1; and (c) conflict points at phase 2.
developed software for determining the number of traffic con-
flict points. The computer software can help simultaneously
promote efficiency and avoid errors.
Conclusion
A generalized method to determine the number of traffic
conflict points between vehicles at major–minor highway
intersections under different control conditions was proposed.
The method is not only suitable for simple 2-way, 2-lane T,
and cross-intersections but also for complex T and cross-
intersections with more than 2 lanes. To develop the method,
the concept of virtual lanes was defined and applied. The
characteristics of different conflict movements and the conflict
433
movements under different right of way conditions were dis-
cussed. The procedure involved in computing the number of
conflict points was presented, and a similar HCM LOS-style
form was developed to facilitate the entire process as a new tool
for computing basic data for access management, safety eval-
uation, and safety improvements. The proposed method was
justified in some projects on intersection safety evaluations in
China. Applications show that the method can determine the
number of conflict points with higher efficiency and accuracy.
Programming the method into computer software will follow.
Acknowledgment
The authors are grateful for support from the Humanities and
Social Science Fund Project under the Chinese Ministry of
Education (Grant No. 12YJCZH162) and from the Natural
Science Foundation of Shandong Province, China (Grant No.
ZR2012EEM005).
References
Access Management Committee. Access Management Manual. Washing-
ton, DC: Transportation Research Board; 2003.
Fujii H, Yoshimura S. Precise evaluation of vehicle emission in urban
traffic using multi-agent-based traffic simulator MATES. Paper pre-
sented at: 12th World Conference on Transport Research; July 11–15,
2010; Lisbon, Portugal.
Glauz WD, Migletz DJ. Application of Traffic Conflict Analysis at Inter-
sections. Washington, DC: Transportation Research Board; 1980.
Lu JJ, Ge X, Pan F, Lu C, Li C. A programmable calculation procedure
for number of traffic conflict points at highway intersections. Paper
presented at: 90th Transportation Research Board Annual Meeting;
January 23–27, 2011; Washington, DC.
Lu JJ, Pan F, Xiang Q. Level-of-safety service for safety performance
evaluation of highway intersections. Transp Res Rec. 2008;2075:24–33.
Maze T, Plazak D, Witmer J, Schrock S. Iowa Access Management Hand-
book. Ames, IA: Center for Transportation Research and Education;
2000.
Minh C, Sano K, Matsumoto S. Maneuvers of motorcycles in queues at
signalized intersections. J Adv Transp. 2012;46:39–53.
Pan F, Lu J, Xiang Q, Zhang G. Level of safety service for highway sig-
nalized intersections. Journal of Southeast University (Natural Science
Edition). 2008;38:298–303.
Pan F, Zhang L, Lu J, Xiang Q, Lu L. Application of access management
techniques in safety improvement at highway intersections. Journal of
Beijing University of Technology. 2011;37:237–242.
Parker M, Zegeer C. Traffic Conflict Techniques for Safety and
Operations—Observers Manual. Washington, DC: Federal Highway
Administration; 1989.
Preston H, Albrecht C. Traffic Safety Fundamentals Handbook. St. Paul,
MN: Minnesota Department of Transportation, Office of Traffic En-
gineering; 2001.
Stiller C, Kammel S, Pitzer B, et al. Team annieWAY’s autonomous sys-
tem. Paper presented at: Robot Vision, Second International Work-
shop; February 18–20, 2008; Auckland, New Zealand.
Stone JR, Chae K, Pillalamarri S. The Effects of Roundabouts on Pedes-
trian Safety. Raleigh, NC; Department of Civil Engineering, North
Carolina State University; 2002.
Vacek S, Nagel R, Batz T, Moosmann F. An integrated simulation frame-
work for cognitive automobiles. Paper presented at: 2007 IEEE Intel-
ligent Vehicles Symposium; June 13–15, 2007; Istanbul, Turkey.
Zhai Z, Jiang X, Liu X. Analysis on conflict at urban road intersections.
Road Traffic Manag. 1992;1:32–34.
Zhu Z. Discussion on design of urban road intersection. Nonferrous
Metals Eng Res. 1996;4:62–70.