CHAPTER THREE: METHODOLOGY

1.1. General
The objectives of the present work have been discussed in the first chapter. In order to
fulfill those objectives, a methodology needs to be formulated to achieve the required
results. This chapter describes the details of the site considered in the study, the
overall method adopted for ranking the safety hazardous locations in Kalanki
Satdobato road section.

1.2. Study Area

Kathmandu Ring Road (KRR) is an eight lane ring road circling around the cities of
Kathmandu and Lalitpur. KRR has been classified as National Highway Category
(Code H16). It serves as the main arterial road in Kathmandu valley. It has been
upgraded for the purpose of reducing the traffic congestion along radial road from
Central Business District (CBD). Therefore, it serves as one of the major transport
link in Kathmandu Valley. However, in the present scenario, this road link is facing
heavy traffic movement creating huge congestion at major intersections.

The Southern section of KRR from Kalanki (CH.10+...) to Satdobato (CH.20+...)

shown in Figure 3.1 was selected as study area.

1
Figure 3.1 Location of study area

The previous two lane KRR has been upgraded to the four lane with service road in
both sides from Kalanki to Satdobato as shown in the Figure 3.2 (DOR, 2019).

Figure 3.2 General cross-section of K-K section of KRR

Kalanki Satdobato road section has been built as the urban arterial road with
technical parameters as shown in Table 3.1 (Corporation, 2012).

Table 3.1: Technical parameters of K-K section of KRR

Parameters Main Road Service Road

Road Class Urban Arterial -
Design Speed 50 km/h 20 km/h
Number of lanes 4 lanes of dual 2 lanes of
carriageway single
Right of way 50 m
Setback 6 m both sides
Road Width 15.5 m 7.5 m
Road clearance (m) 5.0
Minimum circular curve radius for normal 400 70
crown section (m)
Recommended circular curve radius for 200 40
actual elevation (m)
Minimum circular curve radius for actual 100 20
elevation (m)
Minimum length of easement curve (m) 45 20

Maximum longitudinal grade 5.5 8

recommended (%)

2
 Maximum longitudinal grade limitation 7 9
(%)
Minimum length of longitudinal slope 130 60
section (m)
Parameters Main Road Service Road
Convexity vertical curve limit minimum 900 100
radius (m)
Convexity vertical curve general 1350 150
minimum radius (m)
Concave vertical curve limit minimum 700 100
radius (m)
Concave vertical curve general minimum 1050 150
radius (m)
Length of minimum vertical curve (m) 40 20
Pavement structure Asphalt Concrete Asphalt
pavement Concrete
Design load of bridge and culvert Maintain the China Road
original bridge Class – I
Seismic peak ground acceleration 0.3 g 0.3g
Chainage from/to 10+600 (Kalanki) 20+994
(Koteshwor)

1.3. Overview of Methodology

The proposed framework for achieving the objectives of the research “Ranking of
Safety Hazardous Locations” is divided into six stages as shown in Figure 3.3.

3
Figure 3.3 Methodology Flow Chart

1.3.1. Stage 1 and 2: Identification of road elements and safety factors for each
element
On the basis of literature review related to AHP and road safety, study of “Design of
Construction Drawing of The Improvement Project of Kathmandu Ring Road Project
in Nepal” (Corporation, 2012), field visit and experiences, four road elements
(straight, curve, bridge & merge and intersection) were identified in the selected road
section for which factors responsible for the safety of road are also assigned as
tabulated below Table 3.2.

Table 3.2 Factors in each element

A. Speed limit signs and no overtaking signs

1. Straight
Segments: B. Lighting poles and reflective signs
C. Road marking
D. Shoulder width
E. Pavement maintenance condition
F. Drainage

4
 G. Pedestrian Crossing facilities
A. Speed advisory signs, sharp bend, steep up/down grade
warning signs
2. Horizontal and
B. Lighting poles and reflective signs
Vertical Curves:
C. Road marking before and in the curve
D. Shoulder width
E. Combination of horizontal and vertical curves
F. Pavement maintenance condition
G. Drainage
H. Sight distance provision
I. Superelevation in horizontal curves
J. Road Safety Intervention
3. Bridges: A. Speed limit, no overtaking, and load limit signs
B. Lighting poles and reflective signs
C. Road marking
D. Reduction in the pavement and shoulder width
E. Pavement maintenance condition
F. Drainage
G. Guardrails and bridge approach protection
A. Speed limit and warning signs
B. Lighting poles and reflective signs
4. Merge &
C. Road marking
Intersections:
D. Shoulder width
E. Pavement maintenance condition
F. Drainage
G. Visibilty (Sight distance)/ turning radius
H. Distance to the previous
I. Traffic Calming measures/ Appropriate geometry to
reducing speed
J. Pedestrain Crossing Facilities

5
1.3.2. Stage 3: Allocation of weights to factors using AHP
Once the criteria have been identified, and the concepts of establishing priorities and
consistency were clearly understood, the relative weights were allocated to the
selected criteria at each hierarchy level. For this, a scale needs to be established.
Many studies have been conducted for finding the most appropriate scale of
measurement. The 1-to-9 scale has been preferred over other scales since it most
closely resembles our natural ability to distinguish strengths of dominance or
preferences between objects. Table 2.1 shows Saaty’s Intensity of 1-to-9 Importance
Scale.

After selecting the scale of measurement, the pair-wise comparisons were performed
with the help of experts. This concept allows evaluating different criteria by
comparing two criteria at a time as simply described by Figure 3.4. This approach
simplifies the evaluation process by focusing the evaluator’s attention to the two
alternatives at hand.

Therefore, the pair-wise comparisons need to be performed for all combinations of

criteria. Equation 1 provides the number of pair-wise comparisons for n number of
criteria or alternatives. For example, for 10 numbers of criteria, 45 pair-wise
comparisons would be performed by the evaluator or expert. During the evaluation,
the evaluator need to decide the intensity of his or her preference between two
criteria, one pair at a time, ignoring other criteria during the process. The process was
repeated for all combinations of the criteria. In this way, the evaluator’s judgments are
kept in consistent check and a ranking of all the criteria based on pair-wise
comparison is generated.

No. of pairwise comparisons = n × (n − 1)/2..........................Equation 1

Where,

n = Number of criteria. Also, represents size of square matrix.

6
Figure 3.4 Illustration of Pair-wise Comparison
After completing pairwise comparison, the relative weight matrixes (RWM) are
constructed and the matrixes would be:

C1 C2 ….. Cn Aij
C1 1 w1/w2 ….. w1/wn w1
C2 w2/w1 1 ….. w2/wn w2
: : : : : =
: : : : :
Cn wn/w1 wn/w2 ….. 1 Wn
Then the process is followed by calculation of matrix eigenvector, Aij and consistency
index test (CI) of the criterion. For matrix eigenvector, A ij multiply the n elements in
each row, take the nth root, and prepare a new column for the resulting values. Then
divide each number by the sum of resulting values of the new column.
∑𝑛 𝑤1 𝑤1 1
𝑖=1( ∗
𝑤1
∗… )^( )
Eigen vector, Aij =∑[∑𝑛 𝑤2 𝑤2∗ 𝑤𝑛
𝑤1
𝑛
𝑤1 1
𝑤1
𝑖=1( ∗ ∗…∗ )^( )]
𝑤2 𝑤2 𝑤𝑛 𝑛

∑𝑛(∑𝑛 𝐴𝑖𝑗)𝑤𝑗
Eigen value, ƛ𝒊 =
𝑗 𝑖=1
𝐴𝑖𝑗

Consistency test, CI =
𝜆𝑚𝑎𝑥−𝑛

𝑛−1

The consistency index was then compared with random index (RI) which is shown in
Table 3.3. The ratio of consistency index to the random index is called Consistency
7
ratio (CR). If the CR is greater than 10%, the judgment is considered inconsistent and
should be excluded or repeated again.

Consistency ratio, CR=

CI

RI

Table 3.3: Random Index for different dimensions of RWM (Saaty and Wong 1983)
Dimension 1 2 3 4 5 6 7 8 9
RI NA NA 0.58 0.90 1.12 1.24 1.32 1.41 1.45

1.3.3. Stage 4: Field Survey

Entire route of survey was divided into parts of 500m such that the priority index
could be prepared and improvements could be made accordingly. The section of road
to be analyzed (Kalanki – Koteshwor) contained different segments such as horizontal
curves, vertical curves and other facilities such that bridge and pedestrian facilities.
These sections were analyzed separately as severity of problem differs from one to
another.

This was carried out in following steps:

• Reconnaissance Survey
• Facilities Check
• Comparison with guidelines
• Severity Analysis/ Rating

Reconnaissance Survey
Entire section of road was traversed by motorbike. This was carried out to classify the
section into straight section, curves, bridges, merge & intersections and make
necessary changes in predefined section of analysis. Section of road where service
way to be analyzed were noted.

Facilities Check
After classification and completion of first phase of survey, this survey was carried
out to note every road facility throughout the section. Photographs were taken and
precise locations were noted with the help of mobile application and GPS. Every
bridge site, intersections and curves were thoroughly observed. In every 500m
facilities of road section were aggregated. Survey of service lane and main lane was
carried out separately. Qualitative description of pavement, drainage and other
8
facilities were noted. Quantitative descriptions of facility such as speed limit signs
were noted.

Comparison with guidelines

Safety guidelines from government and other standards were studied and relevant
information was noted. This was compared to data extracted from field. Quantitative
and qualitative comparison was done.

Condition Rating
This was carried out with guidance of Road Safety Experts. Safety conditions were
rated according to necessity of facility in the section and severity of improvement
from safety point of view. Condition rating is assigned between zeros to one as
tabulated in Table 3.4, zero is assigned for no deviation with the standard condition
and its value increases up to one for very poor condition of safety factors.

Overall condition was rated by qualitative and quantitative analysis. Some of the
facilities such as pedestrian crossing were adequate in number but they were not safer
enough due to unsuitable and inappropriate placing of the facility.

Table 3.4: Condition rating of a road safety factors

SN State of condition Value

1 Excellent condition 0
2 Good condition 0.10-0.24
3 Average condition 0.25-0.49
4 Poor condition 0.50-0.74
5 Very poor condition 0.75-1.00
(Agarwal, et al., 2013)

1.3.4. Stage 5: Ranking the roads

Combining the weight of safety factors & condition rating of each factor obtained
from stage 3 & 4, Safety hazardous Index was developed using formulas as below.
Safety hazardous Index at straight sections:
SHIs= ∑ (Wsfs x Rsfs)
Safety hazardous Index at curve sections:
SHIc= ∑ (Wsfc x Rsfc)

9
Safety hazardous Index at bridge sections:
SHIb= ∑ (Wsfb x Rsfb)
Safety hazardous Index at intersections:
SHIi= ∑ (Wsfi x Rsfi)
Where,

SHIs, SHIc, SHIb, SHIi = Safety Hazardous Index for straight, curve, bridge and
intersections respectively.

Wsfs, Wsfc, Wsfb, Wsfi = Weight of safety factors at straight, straight, curve, bridge and
intersections respectively.

Rsfs, Rsfc, Rsfb, Rsfi = Condition rating of safety factors at straight, straight, curve,
bridge and intersections respectively.

Further, Safety hazardous index for entire road section (SHIrs) of 2km was obtained
by summation of SHI of all elements.

SHIrs = SHIs + SHIc+ SHIb + SHIi

This was performed for every 2 km road segments of Kalanki- Koteshwor Road
Section. It is noted that higher the SHI at a particular location, higher safety hazardous
condition at that particular location. In this way, ranking the hazardous locations of 2
km stretch each in Kalanki- Koteshwor Road allows the road safety authorities to
implement road safety infrastructure for atleast 2 km stretch in eight lane ring road
keeping limited available road safety budget in mind.

1.3.5. Stage 6: Recommendations for preventive actions

In this stage, the required countermeasures or preventive measures are to be suggested
after identification of the hazardous locations along the study area based on the
available budget and prioritization of countermeasures (which is not in the scope of
this work).

10