Assessing Level of Service of Two Lane Highways Using User Perception and Its Relationship With Field Measurements

Minu S et al. Int. Journal of Engineering Research and Applications www.ijera.com
ISSN: 2248-9622, Vol. 5, Issue 11, (Part - 2) November 2015, pp.91-95
www.ijera.com 91 | P a g e
Assessing Level of Service of Two Lane Highways Using User
Perception and Its Relationship With Field Measurements
Minu S*, Rohini I**, Lija R L***
*(Department of Civil Engineering, SRR Engineering College, Chennai, India)
** ((Department of Civil Engineering, SRR Engineering College, Chennai, India)
*** (Department of Civil Engineering, SRR Engineering College, Chennai, India)
ABSTRACT
Level of Service (LOS) is an important measure of performance of highways. The primary measure of service
quality for two lane highways based on highway capacity manual are percent time spent following and average
travel speed which are not directly based on user perception. In addition to speed and delay the road users
perceive other attributes in their subjective determination of LOS. A methodology based on fuzzy set approach
and fuzzy clustering is presented to determine the LOS that explicitly accounts for user perception. Several
attributes contributing to road user perceptions of LOS were identified using a questionnaire survey and
composite level of service was obtained by combining all attributes together using fuzzy weighted average and
fuzzy C means clustering techniques. The levels of service obtained were compared with that obtained using
field procedure recommended in HCM 2000.
Keywords – Class II highways, Fuzzy clustering, Fuzzy weighted average, Level of service, HCM
I. INTRODUCTION
With increasing delay and congestion on our
nation‟s streets and highways, finding effective ways
to maintain acceptable levels of service is crucial for
satisfying users as well as protecting the environment.
Level of Service (LOS) is an important measure of
performance in analysis of highways. It is used to
evaluate the performance of a highway. LOS in
Highway Capacity Manual (HCM 2000)was defined
as “a qualitative measure describing operational
conditions in a traffic stream, based on service
measures such as speed and travel time, freedom to
maneuver, traffic interruptions, comfort and
convenience”[1]. This definition is applied in a six
level scale levels of service from A to F. LOS A
representing the best operating conditions and LOS F
gives the worst condition.
Methodology adopted for this study is based on
the user perception of road users. The need of fuzzy
logic, based on fuzzy set approach and fuzzy
clustering provides the foundation for approximate
reasoning with imprecise properties
Fuzzy logic starts with the concept of a fuzzy set
[2]. A fuzzy set is a set without a crisp, clearly defined
boundary. The Fuzzy Logic Toolbox is highly
impressive in all respects. It makes fuzzy logic an
effective tool for the conception and design of
intelligent systems. The Fuzzy Logic Toolbox is easy
to master and convenient to use. It can contain
elements with only a partial degree of membership.
Cluster analysis is the name of a group of
multivariate technique whose primary purpose is to
identify similar entities from the characteristic they
posses [2]. The objectives of the cluster analysis are to
group data points into clusters so that the degree of
association is strong between members of the same
cluster (low variance within) and weak between
members of different clusters (high variance
between).
Fang Clara Fang et al. developed and
implemented a methodology based on fuzzy
clustering to define LOS boundaries based on user
perception [3]. Speed, volume and acceleration noise
were taken as the three criteria for the identification of
LOS in a study conducted by K. Ramesh Kumar. This
study was based on development of level of services
for urban roads through user perception and fuzzy set
approach [4]. Partha Chakroborty and Shinya Kikuchi
studied applicability of fuzzy set theory to the analysis
of capacity and level of service of highway facilities
[5].Lin Zhang conducted a study at signalized
intersection to determine the level of service that
accounts for user perceptions. Fuzzy set approach was
used for the determination of LOS [6].
In the present paper, an attempt has been made to
conduct a questionnaire survey and identify the
important attributes that affect road user perception on
LOS of two lane highways of Trivandrum city in
Kerala State, India and to develop a methodology
based on fuzzy set theory and fuzzy clustering to
define LOS on two lane highways. Level of Service
was also determined by using field procedure
recommended in HCM 2000.
RESEARCH ARTICLE OPEN ACCESS

II. DATA COLLECTION AND
METHODOLOGY
The approach of this study was to select a diverse
sample of road users and have these individuals to
evaluate the three stretches of two lane highways of
Trivandrum city in Kerala, India namely
Karyavattom-Chavadimukku, Chavadimukku-
Pongamoodu and Pongamoodu- Ulloor to gauge their
perception of LOS.
2.1 Data Collection
A self administrated questionnaire survey was
developed to test how road users perceive LOS of two
lane highways. Survey questions related to:
 Factors important to users on highways
 Users opinion on average speed
 Users opinion on quality of the road
 Delay experienced on the road
Free flow speed, peak and off peak hour volume count
were also collected for these road stretches.
2.2 Study Methodology
Fuzzy sets that represent the importance levels
and the performance ratings of identified attributes are
characterized by their membership functions.
Triangular membership functions are used in this
study.
The generation of membership functions for
importance levels of the identified attributes involves
the conversion of responses from the histogram to
membership functions. Numbers from 1 to 6 were
assigned to represent the following qualitative
assessments.
 Important = 1
 very less important = 2
 less important = 3
 Moderately important = 4
 highly important = 5
 very high important = 6
Each performance rating class is assigned a
linguistic term such as excellent, very good, good etc
as well as a letter grade A to F. The fuzzy
performance ratings are based on index numbers from
1 to 6, with 1 representing the worst and 6
representing the best.
 Excellent = A
 Very good = B
 Good = C
 Fair = D
 Bad = E
 Worse = F
2.3 Fuzzy Weighted Average (FWA)
The weight coefficients (or importance levels)
and the performance ratings of identified attributes are
treated as fuzzy numbers. The fuzzy weighted
Average method is used to combine all the identified
attributes to arrive at the LOS. Based on FWA, the
fuzzy based overall LOS is defined in (1)
LOS_R =
(1)
Where,
LOS_R = fuzzy overall LOS rating
Wi = fuzzy weight coefficient (or importance levels)
of attribute i
Ri = fuzzy performance rating of attribute i
N = number of attributes that define LOS
2.4 Fuzzy C means (fcm) Clustering
Fuzzy C means Fuzzy C means (fcm) is a data
clustering technique wherein each data point belongs
to a cluster to some degree that is specified by a
membership grade. The Fuzzy Logic Toolbox
command line function fcm starts with an initial guess
for the cluster centers. fcm assigns every data point a
membership grade for each cluster. By iteratively
updating the cluster centers and the membership
grades for each data point, fcm iteratively moves the
cluster centers to the right location within a data set.
This iteration is based on minimizing an objective
function that represents the distance from any given
data point to a cluster center weighted by that data
point‟s membership grade.
III. ANALYSIS
Data collected through questionnaire survey and
through primary surveys from the three road stretches
were analysed to get the level of service. The data
analysed from the questionnaire survey was used to
develop histograms, both for the importance level
and performance level. Fig 1, 2, 3 shows the
percentage of response of importance of speed,
quality of the road and delay at the three road
stretches respectively
Figure 1.Percentage response of importance of
speed



N
i
N
i
Wi
WiRi
1
1

Figure 2. Percentage response of importance of
quality of road
0
10
20
30
40
50
60
impt very
less
impt
less
impt
mod
impt
highly
impt
very
high
impt
%Response
Kariyavattom-
Chavadimukku
Chavadimukku-
Pongamoodu
Pongamoodu-Ulloor
Figure 3. Percentage response of importance of delay
From Fig 1, it is clear that about 34% of the
respondents consider speed as moderately important
for the road stretches Kariyavattom- Chavadimukku
and Chavadimukku- Pongamoodu. 36% of the
respondents in Pongamoodu-Ulloor road stretch also
consider speed as moderately important. From Fig 2,
about 30, 31 and 32% of the respondents consider
quality as highly important for the road stretches
Kariyavattom- Chavadimukku, Chavadimukku-
Pongamoodu and Pongamoodu-Ulloor road stretch.
Around 54% of the respondents consider delay as
moderately important for the road stretches
Kariyavattom- Chavadimukku as from Fig 3.
The histograms for the performance level for the
attributes were also developed. Fig 4, 5, 6 shows the
percentage of response of performance of speed of the
vehicles, quality of the road and delay for the three
selected road stretches respectively.
Figure 4. Percentage response of performance of
speed
Figure 5. Percentage Response of Performance of
Quality of Road
0
10
20
30
40
50
60
70
excellent very
good
good fair bad w orse
%Response
Kariyavattom-
Chavadimukku
Chavadimukku-
Pongamoodu
Pongamoodu-Ulloor
Figure 6. Percentage Response of Performance of
delay
From Fig 4, around 44% of the respondents
consider speed performance as bad for the road
stretches Pongamoodu-Ulloor. However, for the
Kariyavattom- Chavadimukku road stretch about 41%
of the respondents rated the performance of the speed
as fair. Fig 5 shows that around 44%, 42% and 38% of
the respondents consider quality of the road as good
for the road stretches Kariyavattom- Chavadimukku,
Chavadimukku-Pongamoodu and Pongamoodu-
Ulloor respectively. From Fig 6, it is clear that around
59% of the respondents consider the delay
performance in that road is good that is this delay may
cause not much congestion.
3.1Fuzzy Set Approach
Fuzzy sets that represent the importance levels
and performance rating of identified attributes such
as speed, delay and quality of the road section were
characterized by their membership functions.
Triangular membership functions were assumed for
this study. The membership functions were
developed using MATLAB Fuzzy logic toolbox. The
membership function for importance levels of
identified attributes are based on the survey while the
membership function for the performance rating were
derived from the assumptions and currently used
standards. Defuzzification was done for α value of 0,
0.5, and 1, for both importance and performance
levels. The Level of Service of each road stretches
was obtained by using Fuzzy Weighted Average

(FWA) method. The LOS obtained will have a
maximum value and a minimum value, which is
plotted on a standard fuzzy set to obtain the LOS of
the selected two lane road stretches. The values of
LOS obtained for the selected stretches are shown in
the Table 1.
Table 1. Values obtained for level of service
Road Stretches LOS Values
Karyavattom-
Chavadimukku
(4.04,4.62)
Chavadimukku-
Pongamoodu
(3.80,4.60)
Pongamoodu-Ulloor (3.40,4.31)
The obtained value from the fuzzy set approach
was plotted on a standard fuzzy set to obtain the LOS
of the selected road stretches. Fig 7, 8, 9 shows the
LOS plot for the selected road stretches respectively.
Figure 7. LOS of Karyavattom-Chavadimukku road
stretch
Figure 8. LOS of Chavadimukku-Pongamoodu road
stretch
Figure 9. LOS of Pongamoodu-Ulloor road stretch
Levels of service of the three selected road stretches
were obtained as „C‟.
3.2 Fuzzy Clustering
Coding of each variable involved assigning
suitable weights to various levels of selected variables
to achieve the desired result. For the easy analysis, all
the variables obtained from the detailed questionnaire
were coded in a standard manner. In this, the various
factors were assigned the weights on a scale of 1-6.
The membership values of labeled data were
calculated using fuzzy cluster analysis.
Data entered as a matrix form, which contains m
rows and n columns. Rows contain the number of
data points and columns contain the characteristics of
each variable. The data obtained from the
questionnaire survey was used as input. All variables
were coded and these labeled data were used for the
analysis. The data from three road stretches were
analysed using fcm function in MATLAB. The data
were clustered into 6, which gave the six LOS. The
position of the maximum membership grade gave the
value of cluster in which the LOS of the desired road
included. Table 2 gives the maximum membership
grade and position of the cluster for the selected road
stretch
Table 2. LOS of the selected road stretches by
clustering methods
Road Stretches Members
hip Grade
Obtaine
d LOS
Karyavattom-
Chavadimukku
0.1672
B
Chavadimukku-
Pongamoodu
0.6419
C
Pongamoodu-Ulloor 0.4347 C
3.3 LOS Determination Using HCM method
In HCM, level of service is based on percent time
spent following and average travel speed. Percent
time spent following was estimated from (2).
PTSF=100(1-e-0.000879Vp
)+fd/np (2)
Where

PTSF = percent time spent following f d/np =
adjustments for the combined effect of the directional
distribution of traffic and of the percentage of non-
passing zones on percent time spent following
Percent time spent following were calculated for
three road stretches and the obtained values are shown
in Table 3.
Table 3. Percent time spent following for the selected
road stretches
Road Stretches Percent Time
Spent Following
(PTSF)
Karyavattom-
Chavadimukku
52.8
Chavadimukku-
Pongamoodu
54.54
Pongamoodu-Ulloor 56.06
On comparison with the values suggested by
HCM for the percent time spent following, the level
of service for the Kariyavattom- Chavadimukku,
Chavadimukku-Pongamoodu and Pongamoodu-
Ulloor road stretches were found to be B , B and C.
IV. SUMMARY AND CONCLUSION
The level of service is a qualitative measure
which needs to reflect road user perception. Several
attributes contributing to the perception of level of
service were identified by conducting a questionnaire
survey and LOS of the selected road stretches were
assessed using fuzzy set approach and fuzzy
clustering technique. Fuzzy set approach only the
importance and performance levels of identified
attributes such as speed, delay and quality of the road
section can be considered. However, in fuzzy
clustering technique, all the attributes such as age,
gender, educational level, purpose of trip etc can be
considered. Hence fuzzy clustering technique gave
more accurate prediction.
The levels of service were also determined using
field procedure recommended in HCM 2000. LOS
obtained from the user perception methods was found
to be lower level than that obtained from the field
measurement. The user perception attributes
influences the determination of LOS of two lane
highways significantly. This may be the reason why
the obtained LOS values are not in true match with
that of HCM values.
REFERENCES
[1] Highway Capacity Manual, Transportation
Research Board, National Research Council,
Washington D.C,2000
[2] Timothy,J,Ross.,“Fuzzy Logic With
Engineering Applications”, University of
Mexico, USA,2004
[3] Fang Clara Fang, and Lily Elefteriadou,
“Using Fuzzy Clustering of User Perception
to Define Levels of Service at Signalized
Intersections”, Journal of Transportation
Engineering, 129(6), 2003, 657-663
[4] K. Ramesh Kumar, Level of Service Indices
for Urban Roads Through User’s
Perception-Fuzzy Set Approach, doctoral
diss., Indian Institute of Technology,
Madras,1993.
[5] Partha Chakroborty and Shinya Kikuchi,
“Application of Fuzzy Set Theory to the
Analysis of Capacity and Level of Service
of Highways”, IEEE Journal,1990.
[6] Lin Zhang and Panos D.
Prevedouros,“Signalized Intersection Level
of Service that Accounts for User
Perception”, Transportation Research Board,
Washington D.C,2004,1-26.
[7] IRC 106:1990, “Guidelines for Capacity of
Urban Roads in Plain Areas” , Indian Road
Congress, New Delhi.

Assessing Level of Service of Two Lane Highways Using User Perception and Its Relationship With Field Measurements

More Related Content

What's hot (20)

Viewers also liked (20)

Similar to Assessing Level of Service of Two Lane Highways Using User Perception and Its Relationship With Field Measurements (20)

Recently uploaded (20)

Assessing Level of Service of Two Lane Highways Using User Perception and Its Relationship With Field Measurements