E L

P T
INTRODUCTION TO MULTIMODAL URBAN TRANSPORTATION SYSTEMS

N
Arkopal K Goswami, PhD
Ranbir and Chitra Gupta School of Infrastructure Design and Management
IIT KHARAGPUR

Module 03: Non-motorised Transportation (NMT) Planning

Lecture 26 : NMT Characteristics
 E L
P T
 Basic NMT Characteristics
 Pedestrians

N
 Bicyclists
 Traffic flow equations
 Factors affecting pedestrian speed
 L
NMT Characteristics

E
Pedestrians

• Definition:

T
• Any person on foot

P
• People who walk, sit, stand in public
space—uses walking stick, crutches or

N
wheel chair—old or young—workers,
residents, shoppers etc. (IRC 103:2012)

• every motorist is a pedestrian in some

part of the trip
 L
NMT Characteristics

E
Pedestrians
• Space requirement: Body Ellipse Head

T
Shoulder
Buffer area

P
Body depth

Body width Luggage

N
0.5 m 0.35 m 0.52 m

0.6 m 0.51 m 0.51 m

US HCM, 2000 Indo HCM, 2018 Indo HCM, 2018
 L
NMT Characteristics

E
Cyclist

• Definition:

T
• Any person using physical power
driven cycles

P
• People who use bicycle, three-

N
wheeled rickshaw or four-wheeled
cars are all included as Cyclists

• Important mode of access along with

walking
 L
NMT Characteristics

E
Cyclist
• Space requirement: Bicycle Dimensions as per IRC 11-2015

P T
N
 L
NMT Characteristics

E
Cyclist
• Space requirement: Cycle Rikshaw Dimensions as per IRC 11-2015

P T
N
 L
NMT Characteristics

E
Cyclist
• Space requirement: Dimensions as per IRC 11-2015

P T
N
 L
NMT Characteristics

E
Traffic Flow and Capacity

• Similar to vehicular flow

T
as the flow be defined
using the same

P
variables– speed, flow,
density etc. (Virkler, 2002)

N
• Fundamental Equation Changes with NMT
of Flow characteristics and
• 𝑞 = 𝑢 ∗ 𝑘 Where, q= built environment
Flow; u=velocity or
speed; k=density
 L
Pedestrian Characteristics affecting speed

E
• Pedestrian Types
• Age

T
• Gender
• Trip Purpose

P
• Baggage and luggage carrying
• Platoon Groups of pedestrians

N
• Built environment characteristics
• Gradient
• Walkway width
• Kerb and street furniture
• Two-Way friction
• Location effect
 L
Pedestrian Characteristics affecting speed

E
Pedestrian types
• Pedestrian Types

T
• Age 1.2m/s (~72 m/min)
recommended  elderly 1m/s (30%
decrease)

P
• Gender—men vs women different
mean walking speed 81 m/min for Platoon
males; 76 m/min for females

N
• Trip Purpose—work trip generally
faster than leisure trips (like evening
stroll)
• Baggage and luggage carrying
walking speed affected by weight of • Platoon are generally socially
luggage, larger weight reduction in interaction, this decreases their
speed concentration and their speed
 L
Pedestrian Characteristics affecting speed

E
Built environment
• Built environment characteristics

T
• Gradient 10% (10 units of rise in every 100 units of length) on
uphill grades—0.1 m/s reduction in speed

P
No measurable effect on speed upto a grade of 3%
• Kerb and street furniture ped. tend to keep 0.30 to 0.45m lateral
clearance from the kerb line. Average space occupied by

N
pedestrians 72 to 79 m2/ped
• Two-Way friction speed and capacity decreases by 15% when
flow prop.10% 90:10 directional split Friction
50:50, capacity of 2-way= 1-way sidewalk
 L
Pedestrian Characteristics affecting speed

E
Built environment Location/Social-Cultural-Climatic effect

P T
N
(Source: Banerjee et
al., 2018)
 L
NMT Characteristics

E
How to identify variation in speed between two groups?—Independent sample T-Test

Step 1: Identify the groups you need to understand the difference between. For example—

T
Age<15 or >50, Gender M or F, gradient 10% or 15% etc.
Step 2: Collect speed data for the group For example—walking speed data (in m/min) for age

P
group <15 and >50
Step 3: Calculate the sum of difference σ x1−2 and the square of difference σ (x1−2 )2 between

N
speeds for the two group
Step 4: Calculate mod value T-statistic
x
σ 1−2
|t| = 𝑁
where N=no. of samples
x
σ( 1−2 )2−[
x
σ 1−2 2
]
𝑁
𝑁.(𝑁−1)
 L
NMT Characteristics

E
How to identify variation in speed between two groups?—Independent sample T-Test

Step 5: Calculate the degree of freedom, DOF= N-1, where N=no. of samples

T
Step 6: Find the p-value/critical t value 𝑡𝑐𝑟 from the Two Tail Distribution t-table using DOF. Take
p-value= 0.05 (5%) if nothing is specified

P
P-value
DOF

N
Step 7: If |t|<𝑡𝑐𝑟 , OR, if the p-value of the 𝑡 is > 0.05 (which could be identified from
the t-table)
then there is significant difference between the speeds of the two
groups
 L
Numerical Problem #1

E
S No. Speed of Speed of
Males Females
(m/min) (m/min)

T
Use the t-test to check if the speeds of the
1 72 69
males and the females are significantly
2 71 65
deferent or not at p-value= 5%?

P
3 69 68
4 73 67

N
5 68 65
Step 1: Identify groups males vs. females
6 71 64
7 73 63 Step 2: Data collection done
8 64 68
9 67 67
10 69 69
 L
Step 4: Calculate t
Numerical Problem #1—Solved σ x1−2
𝑁

E
S Speed of Males Speed of X1-2 (X1-2)2 |t| =
σ x1−2 2
No. (m/min) X1 Females σ(x1−2 − [
)2 ]
𝑁
(m/min) X2 𝑁. (𝑁 − 1)

T
1 72 69 3 9 32
10
2 71 65 6 36 |t| = =0.02004
32 2

P
3 69 68 256−[ 10 ]
1 1
10∗9
4 73 67 6 36
5 68 65

N
3 9
Step 5: Calculate DOF
6 71 64 7 49
7 73 63 10 100 DOF=N-1
8 64 68 -4 16 = 10-1
9 67 67 0 0 =9
10 69 69 0 0
Total Step 3: σ x1−2 =32 σ(x1−2 )2=256
 L
Numerical Problem #1—Solved

E
Step 6: Find 𝑡𝑐𝑟

P T
𝑝−𝑣𝑎𝑙𝑢𝑒 0.05
𝑡𝑐𝑟 𝑑𝑜𝑓 = 𝑡𝑐𝑟 9 = 2.262

Step 7: Decision
|t|<𝑡𝑐𝑟

N
There is significant difference between the speeds of males and females
 E L
P T
 Modelling Pedestrian for Planning and Design (2011)—Dr. Marwan AL-Azzawi. Lambert

N
Academic Publishing
 Traffic and Highway Engineering. 5th Edition—Garber & Hoel. Cengage Learning
 Transportation Engineering and Planning. 3rd Edition—Papacostas & Prevedouros. PHI
Learning
 E L
T
• Any person on foot—Pedestrian & person on any type of physically

P
powered cycle is a cyclists

• Different characteristics for each type of NMT—Variation in sizes and

N
dimensions

• Pedestrian speed is affected by individual characteristics as well as the

built environment

• Difference between speeds of two groups of sample can be

determined by a T-Test
 E L
P T
N
 E L
P T
INTRODUCTION TO MULTIMODAL URBAN TRANSPORTATION SYSTEMS

N
Arkopal K Goswami, PhD
Ranbir and Chitra Gupta School of Infrastructure Design and Management
IIT KHARAGPUR

Module 03: Non-motorised Transportation (NMT) Planning

Lecture 27 : Pedestrian Data Collection and Flow Characteristics
 E L
P T
 Data Collection on pedestrian flows
 Pedestrian flow characteristics

N
 Greenshields’ macroscopic flow model
 Regression modelling
 L
Pedestrian Data Collection

E
Manual Counts
• Count the flow of pedestrian through a junction, across a road,

T
or along a road section/footway manually using manual clicker
and tally marking sheet.
• Constraints:

P
• peak hour of the day
• special events should be avoided since they can result
in non-typical conditions

N
• locations need to be carefully selected

• Advantage: Simple and easy to perform

• Disadvantage: Labour Intensive
 L
Pedestrian Data Collection Sensor

E
Automatic Counts

T
Counts the number of people, direction and speed
automatically when a pedestrian passes by over or
near it.

the walkway

N
Slab type: Buried under P
Pyro-box: Counts
passerby people
Urban Post:
Fixed Counters
 L
Pedestrian Data Collection Sensor

E
Automatic Counts

T
Counts the number of people, direction and speed
automatically when a pedestrian passes by over or
near it.

P
• Constraints:
• Protection from disruption—weather,
theft, etc.

N
• Locations need to be carefully selected
• Skilled labour required

• Advantage: High quality data free from error

• Disadvantage: Costly and fragile
 L
Pedestrian Data Collection

E
Videographic Survey
• Cameras are setup at the selected sites and video recording taken of the

T
pedestrians during the selected observation periods
• Data Extraction Technique: Can be manual or through computer vision
techniques

P
• Advantage: More efficient than manual count
• Disadvantage: Requires more skill

N Videographic Counting through Computer Vision & Image Processing

 L
Pedestrian Flow Characteristics

E
Flow parameters

• Fundamental Equation

T
of Flow

P
• 𝑞 = 𝑢 ∗ 𝑘 Where, q=
Flow; u=velocity or

N
speed; k=density

𝑢
• Or, 𝑞 = Where,
𝑠
s=spacing=1/k
 L
Pedestrian Flow Characteristics

E
Flow parameters
• Terminologies: k or s

T
• Pedestrian speed (u) is the average pedestrian
walking speed, units m/sec or m/min

P
• Pedestrian flow rate (q) is the number of
pedestrians passing a point per unit of time,
unit ped/15 min or ped/min

N
• Pedestrian density (k) is the average number of q
pedestrians per unit of area, unit ped/m2
• Pedestrian space (s) is the average area
provided for each pedestrian, unit m2/ped
This is the inverse of density
 L
Pedestrian Flow Characteristics

E
Typical flow model

• Greenshield’s macroscopic stream model

T
Linear relationship between u vs. k

P
• Greenberg’s logarithmic model
Logarithmic relationship between u vs. k

N
• Underwood’s exponential model
Exponential relationship between u vs. k
 L
Pedestrian Flow Characteristics

E
Greenshield’s macroscopic stream model
•Linear relationship between speed and density

T
• Mean Free Speed uf: Speed when
Mean-free
there is no friction from other Speed
objects, i.e. k=0

P
• Jam Density kjam: Density at which

N
no movement is possible (traffic
jam), i.e., u=0

• Intercept of this straight line: uf

• Slope of this straight line: - (uf/ kjam) Jam

Density

10
Indian Institute of Technology Kharagpur
 L
Pedestrian Flow Characteristics

E
Greenshield’s macroscopic stream model
Since, q=uk

T
the linear relationship between u & k changes to a parabolic one between q and k
Max. Flow
Replacing u=q/k in

P
Gives us

N
Equation of a parabola
By the property of parabola:

• Jam Density kjam: Density at which no movement is possible

(traffic jam), i.e., u=0 also q=0
Jam
• Max Density kmax: Occurs at the midway, when k= kjam /2 Density
• Max Flow qmax: Max flow occurring at kmax
 L
Pedestrian Flow Characteristics

E
Greenshield’s macroscopic stream model
Similarly again rewriting the equation of the linear relationship in terms of u and q

T
Mean-free
Speed
By the property of parabola:

P
• Mean Free Speed uf: Speed at
q=0
Jam Density

N
(constant)
• Max Flow qmax: Occurs at the
Max. Flow
midway, when u= uf /2
 L
Numerical Problem #1

TE
u
Consider this hypothetical data, where, k= density (ped/m2) and
u=speed (m/min) of pedestrians on a footpath

P
(a) Determine the parameters of the flow model?
(b) Find the flow and density corresponding to 30 m/min?

N
Use Greenshield’s macroscopic model, where the relationship between u and k
is a straight line, with equation y=ax + b, x= speed and y=density

13
 L
Numerical Problem #1—Solved

E
u

T
This can be solved firstly by calibrating the data to a linear regression model

The technique is called “curve fitting”

P
Using (y=ax + b ) and then solving for the coefficients (a and b) using the method
of least squares:

N
xi and yi are data points in the given table

14
 L
Numerical Problem #1—Solved

E
(a) The basic parameters of Greenshield’s model

T
(u)

N P
Since the speed (u) is dependent on the density (k); considering x=k and y=u; we
solve to get the values of a and b in y=ax+b
Thus, b= -0.2 and a= 40.8
u= 40.8 - 0.2k

15
Indian Institute of Technology Kharagpur
 L
Numerical Problem #1—Solved
204/2= 102 ped/min

E
(a) The basic parameters of Greenshield’s model

T
As per the u vs. k relationship; u= 40.8 - 0.2k
204

P
Thus, Mean free speed (uf) = 40.8 km/hr
And, for the q vs. k relation ship q= 40.8k - 0.2k2

N
At q=0, k= kjam we get 0= 40.8kjam - 0.2kjam 2 therefore solving this quadratic
equation kjam =0 or kjam = 204 ped/m2 , this is when the footpath is completely
crowded
At q=qmax, k= kjam /2, using the equation, qmax= 40.8(kjam/2) - 0.2(kjam/2)2
Thus, qmax= 2081 ped/min

16
Indian Institute of Technology Kharagpur
 L
Numerical Problem #1—Solved

E
(b) q and k at u=30 m/min

T
for k, use relationship u= 40.8 - 0.2k

At u=30, k=54 ped/m2

P
for q, use relationship q= 40.8k - 0.2k2

N
At u=30, k=54 ped/m2
Thus, q= (40.8*54)-(0.2*542)

q= 1604 ped/min
 L
Modelling Pedestrian Flow Characteristics

E
Greenberg’s logarithmic model
•Logarithmic relationship between speed and density

Underwood’s Exponential model

P T
•Exponential relationship between speed and density

N
 L
Modelling Pedestrian Flow Characteristics

E
State-of-the-art
Type of Modelling Technique/Software

T
* Does not *Does not consider
focus on crowd human factors/
dynamics Behavioral Movement behaviour

P
Fluid/Particle Matrix-based
Conceptual Computer

N
System system
Assume fluid HCM,2000 results implicitly
Observed, empirical and Simulates behaviour behaviour as per incorporated—evacuation
reported actions using
of individual Boltzmann Gas Eqn. models majorly
questionnaire
LEGION EXODUS PFES
 Social force model
PEDROUTE RAMPAGE EGRESS
SIMPED PEDROUTE/PAXPORT
 E L
P T
 Modelling Pedestrian for Planning and Design (2011)—Dr. Marwan AL-Azzawi. Lambert

N
Academic Publishing
 Traffic and Highway Engineering. 5th Edition—Garber & Hoel. Cengage Learning
 Transportation Engineering and Planning. 3rd Edition—Papacostas & Prevedouros. PHI
Learning
 E L
T
• Various types of data collection each has its advantages and

P
disadvantages

• Collection technique is dependent upon the time and monetary

N
constraints

• Assumed relationship between u and q could be—linear, exponential

and logarithmic

• Linear regression is an effective method in modelling statistical

relationships
 E L
P T
N
 E L
P T
INTRODUCTION TO MULTIMODAL URBAN

N
TRANSPORTATION SYSTEMS
Arkopal K Goswami, PhD
RCG School of Infrastructure, Design and Management , IIT KHARAGPUR
Module 03: Non-motorised Transportation (NMT) Planning
Lecture 28 : Pedestrian flow models
 E L
P T
 Pedestrian flow models—macroscopic
 Types of trip purpose

N
 Types of land uses
 Two-regime models
 Fit of a model
 L
Pedestrian Flow Models

E
Macroscopic models
• This approach uses empirical equation to quantify the flow of pedestrian crowd and the

T
capacity of walking facilities.

• The relationship among density, speed and directional flow for pedestrians is similar to that

P
for vehicular traffic streams

Qp = Vp × Kp Qp = Vp /s

N
Where, Where,
Qp = Unit flow rate (ped/min/m) S= Pedestrian Space (m²/ped)
Vp = Pedestrian speed (m/min), and
Kp = Pedestrian density (ped/m²)
 L
Pedestrian Flow Models

E
As per trip makers

Speed in m/min
T
• As density increases and
pedestrian space decreases
degree of mobility afforded to

P
the individual pedestrian
declines

N
• Therefore pedestrian space
decreases

• Students walks faster Density in ped/m2

 L
Pedestrian Flow Models

E
Pedestrian space inverse of density

T
• q=u*k and also k=1/s, since

P
Space Module is inverse of
density

• Thus, q=u/s

N
• q vs. s relationship also
parabolic
 L
Pedestrian Flow Models

E
Pedestrian space inverse of density

P T
Can be transformed
in terms of Space
Module (M)

N
• Space is an important assessment parameter for pedestrian studies
• Analogous to density in vehicle studies
 L
Pedestrian Flow Models

E
Macroscopic models
General relationships for analysis amongst various pedestrian flow parameters have been

T
evolved based on single-regime approach
• single-regime based on the assumption that the same speed–density relation is valid
for the entire range of densities seen in traffic streams

P
• Pedestrian speed (Vp) and density (Kp): Vp = a – b × Kp
• Pedestrian flow (Qp) and density (Kp): Qp =a x Kp – b x Kp2
• Pedestrian speed (Vp) and flow (Qp): Qp = Vp (a – Vp)/b

N
𝑎 𝑏
• Pedestrian flow (Qp) and area module (S): Qp = − 2
𝑠 𝑠

• multiple-regime speed–density relation will also be different in different zones of

densities
• Since, human behavior will be different at different densities
• simplest one two-regime model, separate for congested and uncongested traffic
 L
Pedestrian Flow Models

E
Macroscopic models
• Two-regime macroscopic models

P T
N
 L
Pedestrian Flow Models

E
Why is there a need for such models?

• To forecast future flow parameters using the

T
model equations—i.e. u vs. k, q vs. k or u vs. q

P
• To design the capacity of pedestrian facilities
like—Sidewalk, crosswalk, corridor, stairway,

N
etc.

(Image source: ITDP)

• To understand the performance of the
facility—Level of Service (LOS); more
specifically Pedestrian Level of Service (PLOS)
 L
Numerical Example #1

E
Given the following equation of a u-k relationship for a commercial street with no sidewalk.
Determine the :

T
(a) average speed of pedestrians when density=0.9, 1.1 and 3 ped/m2
(b) the graphical representation of the model

P
(c) derive the other flow parameter models

Vp = 65-15 Kp

N
Where,
Qp = Unit flow rate (ped/min/m)
Vp = Pedestrian speed (m/min), and
Kp = Pedestrian density (ped/m²)
 L
Numerical Example #1

E
(a) Using the equation and using Kp =0.9,1.1 and 3 respectively
Vp = 65-(15 *0.9)

T
= 51.5 m/min

P
Vp = 65-(15 *1.1)
=48.5 m/min

N
Vp = 65-(15 *3)
= 20 m/min

Higher density will accommodate less average pedestrian speed

 L
Numerical Example #1

E
(b) From the equation 65 m/min is the intercept of the straight line
Vp = 65-15 Kp

T
Therefore mean free pedestrian speed= at uncongested condition= Vpf = 65m/min

P
Vp
Slope of the line= 15 (minus indicate
negative direction of slope) 65

N
Thus, density at Vp=0= Kjam
Slope= 65/Kjam
4.33
Or, 15=65/Kjam
Or, Kjam= 4.33 ped/m2
Kp
Kjam
 L
Numerical Example #1

E
(c) Using the fundamental equation of traffic flow
Qp = Vp × Kp

T
Or Vp=Qp/Kp
Vp = 65-15 Kp Similarly,
or, Qp/Kp = 65-15 Kp Qp =(65 Vp - Vp2)/15

P
or, Qp = 65Kp-15 (Kp )2 Flow vs. Speed—
Flow vs. Density— Parabolic

N
Parabolic
Also, using Kp = 1/s
Qp = 65Kp-15 (Kp )2
Qp = 65(1/s)-15 (1/s)2
Flow vs. Space—
Inverse Parabolic
 L
Pedestrian Flow Models

E
Models estimated by IndoHCM as per Landuse

P T
N
 L
Pedestrian Flow Models

E
Models estimated by IndoHCM as per Landuse

P T
N
 L
Pedestrian Flow Models

E
Strength of models—R2 Goodness-of-fit index

• R2 determines how well the data fits to

T
the actual relationship model Model relationship
Actual Data
• Ranges between 0 to 1 Points

P
• Higher the value more is the fit

N
• A very high value is not always good—
indicates bias/distortion in relationship
 E L
P T
 Traffic and Highway Engineering. 5th Edition—Garber & Hoel. Cengage Learning

N
 Indian Highway Capacity Manual—Indo HCM (2011). CRRI-CSIR, GoI
 E L
P T
• Microscopic modelling of pedestrian flow models using space

• Space is an important performance parameter

N
• IndoHCM estimates flow models for different landuse conditions

• R2 is used the judge the fit of the model to the actual data
 E L
P T
N
 E L
P
INTRODUCTION TO MULTIMODAL URBAN
T
N
TRANSPORTATION SYSTEMS
Arkopal K Goswami, PhD
RCG School of Infrastructure, Design and Management , IIT KHARAGPUR
Module 03: Non-motorised Transportation (NMT) Planning
Lecture 29 : Pedestrian flow characteristics on facilities
 E L
P T
 Pedestrian Facilities
 Sidewalk/Footpath

N
 Crosswalk
 Stairway
 Foot Over bridges/ Skywalk
 Fundamental flow parameters of the pedestrian facilities
 L
Pedestrian Facilities

E
Sidewalk/Footpath
• Paved or unpaved

T
• Separated from
motorised traffic Chandpole bazar, Jaipur ITO, New Delhi JM Road, Pune

P
• Ped volume varies
with location/

N
landuse, purpose, day
and time
Landuse code:
Residential
Institutional MG Marg, Gangtok Charminar Bazar, Hyderabad
Commercial
Mixed (resi. +
(Image source: UAY Project, IIT Kgp)
comm.)
 L
Pedestrian Facilities

E
Sidewalk/Footpath Two peaks

• Paved or unpaved

T
Weekday

• Separated from motorised traffic Weekend

P
• Ped volume varies with location/
landuse, purpose, day and time

N
• Two peaks on weekdays—morning and
evening peak (Source: IndoHCM, 2018)

• Morning peak work trips

mid-day and evening trips shopping
and leisure trips
 L
Pedestrian Facilities

E
Crosswalk

• For walking across

T
street/road
• India Zebra and

P
signalized crossing only at * ITO, New Delhi * Chandpole bazar, Jaipur
grade facilities
• Fixed signals Phases of Signalized *

N
signal predefined— Unsignalized *
irrespective of ped. or veh. Landuse code:
flow Residential
Institutional
• Ped volume varies with
Commercial
location/ landuse, Mixed (resi. +
purpose, day and time comm.)
* JM Road, Pune
(Image source: UAY Project, IIT Kgp)
 L
Pedestrian Facilities

E
Crosswalk—other types of pedestrian signalization

P T
N
Button for Button+ Sensor Both cyclist and
activating the for activating the Pedestrians can
pedestrian green pedestrian green cross ”Too-Can”
in the traffic signal in the traffic signal or “Toucan”
cycle cycle
(Animation source: www.ageas.co.uk )
 L
Pedestrian Facilities

E
Crosswalk—Terminologies related to performance
Time taken by car to reach zebra crossing
• Gap: time diff. between tail of leading vehicle

T
and head of following vehicles (in sec.) Gap

• Ped. critical gap (tc): min. gap size in traffic

P
stream which will allow the entry of a ped. to
cross the road (in sec.) Time taken by ped.
• If gap< tc  Rejected gap by ped.

N
to cross the road
• If gap> tc Accepted gap by ped.
Wait time
• Ped. Wait time: Time lost between arrival of
pedestrian at crossing point and the starting
of the crossing manoeuvre/ accepting the
gap (in sec.)
 L
Pedestrian Facilities

E
Crosswalk—Estimating critical gap
Cumulative Frequency (CF) distribution—Example

T
• Cumulative addition
• Adding subsequent frequency values

P
Spending (INR) Frequency CF (“<“ type) CF (“>” type)
<1000 22 22 614
+
1001-2000 45 67 592

N
+
2001-3000 57 124 547
+
3001-4000 97 221 490
+
4001-5000 152 373 393
5001-6000 241 +
614 241
 L
Pedestrian Facilities

E
Crosswalk—Estimating critical gap
Cumulative Frequency (CF) distribution—Ogive

T
• Graphical representation sigmoid
shape

P
• Raff’s definition: tc is the size of gap

Frequency
whose number of accepted gaps
shorter than it is equal to the number

N
of rejected gaps larger than it

Intersection point: Median

Items
 L
Numerical Example #1

E
Determine the pedestrian critical gap from the following data

T
Length of Gap Frequency No. of No. of rejected
(t sec.) accepted Gaps gaps
(< t sec) (> t sec)

P
0 0 0 0
0-1 2 2 116

N
1-2 10 12 114
2-3 20 32 104
3-4 25 57 84
4-5 27 84 59
5-6 32 116 32
 L
Numerical Example #1—Solved

E
From the table we can see—the data is that of a cumulative frequency distribution

So we can draw less than ogives for accepted gaps and greater than ogives for rejected gaps

T
140
120

P
100 The intersection of the two
Frequency

80 ogives give tc=3.2 sec

N
60
40
20
0
0 1 2 3 4 5 6 7
Gap (sec)
 L
Pedestrian Facilities

E
Stairway
• Connection to foot over

T
bridges or skywalks

• Hilly area connection

P
to different levels
Typical stairway in Indian sub-
• Ped volume depends on urban train stations

N
the location—for (Image source: IndoHCM, 2018)
suburban station
depends on train arrivals
and departures
Gangtok connecting two
grades of street
(Image source: UAY Project, IIT Kgp)
 L
Pedestrian Facilities

E
Foot Over Bridge/skywalk

• Eliminates all potential vehicle

T
conflicts in comparison with at
grade crossings

P
Bandra Skywalk, Mumbai
• Skywalks in Mumbai aims to
dissipate pedestrian crowd from

N
suburban train stations Butterfly bridge (Ped.
Interchange), Gangtok
• Butterfly bridge connects to the
Gangtok cable cars

Bandra Skywalk, Mumbai

(Image source: UAY Project, IIT Kgp)
 L
Fundamental flow parameters of pedestrian facilities

E
Speed observed by IndoHCM at footpath
• Speed most important component

T
in flow model calibrations

• Maximum speed= 107 m/min and

P
Minimum speed= 38m/min
Average=66 m/min (~1.1m/s)

N
• Speed also varies by landuse—
highest for residential (mean= 82
m/min) followed by transportation
terminal (mean= 76 m/min)

• Lowest for recreational (58.05) and

commercial (62.72)
 L
Fundamental flow parameters of pedestrian facilities

E
Speed observed by IndoHCM at crosswalk

T
• Avg. speed at which ped. cross the
road

P
• Males walk faster than females

• Crossing speed increases with the

N
increase in the no. of lanes

• Speed on footpath < crosswalk

 L
Fundamental flow parameters of pedestrian facilities

E
tc observed by IndoHCM at crosswalk

• Critical gap is estimated based on

T
the vehicular gaps accepted and
rejected by ped.

P
• A pedestrian who wishes to cross
the road is subjected to vehicular

N
gaps

• Most popular method for estimating

critical gap is using Raff’s method

• tc increases with number of lanes

 L
Fundamental flow parameters of pedestrian facilities

E
Waiting time observed by IndoHCM at crosswalk

• Important factor in the pedestrian crossing decision process

T
• Waiting time increases ped. do not get adequate acceptable (safe) gaps to
cross the traffic stream
• Increasing wait time crash risk increases due to impatience of peds.

P
• Wait time increases with number of lanes
• Females are observed to wait longer than males

N
 E L
P T
 Traffic and Highway Engineering. 5th Edition—Garber & Hoel. Cengage Learning

N
 Indian Highway Capacity Manual—Indo HCM (2011). CRRI-CSIR, GoI
 E L
P T
• 4 types of pedestrian facilities—footpath, crosswalk, stairs and
skywalk

N
• 2 types of crossing facility in India—Zebra and fixed signal type

• Critical pedestrian gap can be estimated using Raff’s method

• IndoHCM provides the essential traffic engineering details for NMTs

 E L
P T
N
 E L
P T
INTRODUCTION TO MULTIMODAL URBAN

N
TRANSPORTATION SYSTEMS
Arkopal K Goswami, PhD
RCG School of Infrastructure, Design and Management , IIT KHARAGPUR
Module 03: Non-motorised Transportation (NMT) Planning
Lecture 30 : Pedestrian Level of Service (PLOS) based on Flow models
 E L
P T
 What is PLOS?
 PLOS of Pedestrian Facilities

N
 Sidewalk/Footpath (as per Indo HCM)
 Crosswalk (as per US HCM)
 L
Pedestrian Level of Service (PLOS)

E
Definition
Huge and varied nos. of definition

T
• PLOS is “qualitative measure of pedestrian traffic flow, along
with environmental factors that might affect perceived level of
comfort, convenience, safety, security and the economy of

P
walkway systems” (HCM, 2000) crowded

• PLOS has been defined as “a measure for explaining current

N
facilities, situations, equipment and infrastructures in streets,
and it also evaluates the quality of service” (Asadi-Shekari & Zaly
Shah, 2011)
encroached
• “PLOS is a measure for assessing the operating condition of
facility in a quantitative manner. It denotes the level of comfort
provided by the facility to pedestrians” (IndoHCM, 2018)
 L
Pedestrian Level of Service (PLOS)

E
Based on fundamental traffic flow equations

T
• PLOS classified  on
Measures of A B
Effectiveness (MOEs) C

P
D
• MOEs used  E
• Speed

N
• Flow
• Density (or space) F
 L
Pedestrian Level of Service (PLOS)

E
Based on fundamental traffic flow equations

T
PS >1.3-1.9 m2/p ;
Ped Space (PS) >4.9 m2/ped ; FR <= 21-27 p/ min/m

P
Flow Rate (FR) <= 12 ped/ min/m

PS >0.6-1.3 m2/p ;

N
PS >3.3-4.9 m2/p ; FR <= 27-45 p/ min/m
FR <= 12-15 p/ min/m

PS <=0.6 m2/p ;
PS >1.9-3.3 m2/p ; FR varies
FR <= 15-21 p/ min/m
 L
Pedestrian Level of Service (PLOS) for footpaths

E
As per IndoHCM

Type of landuse: Residential,

T
Commercial etc.
Effective width= Available

P
width after subtracting
shy distance

N
Flow Rate= Ped Flow/
Effective width
 L
Pedestrian Level of Service (PLOS) for footpaths

E
As per IndoHCM
Determining effective width

P T
N (Source: HCM, 2000)
 L
Pedestrian Level of Service (PLOS) for footpaths

E
As per IndoHCM

• Shy distance

T
• Buffer that pedestrian give themselves to avoid accidently
stepping off the curb, brushing against a building face, or
getting too close to other pedestrians standing under

P
awnings or window shopping
• The lower value of shy distance indicates moderate value of
obstruction along the linear length of the footpath

N
• Higher value indicates high degrees of obstruction

• Fixed-object effective width

• includes the object’s physical width, any functionally unusable
space, and the
buffer given to the object by pedestrians
 L
Pedestrian Level of Service (PLOS) for footpaths

E
As per IndoHCM

P T
N
More crowded as we move to letter grade F
 L
Numerical Example #1

E
A 2.5 m wide footpath around transport terminal barricaded with guard rail on both sides.
The existing peak flow rate is 1000 pedestrians per 15 min. What will be the PLOS considering

T
pedestrians at transport terminal areas? Use IndoHCM PLOS definition.

N P
 L
Numerical Example #1—Solved

E
Terminal

T
2.5m width

N P
 L
Numerical Example #1—Solved

TE
distance
= 2.5 – (2*0.5)
= 2.5 – 1.00
= 1.5 m
N
Effective width of Footpath =
Total width – shy away

P
 L
Numerical Example #1—Solved

Ped flow= 1000 per 15 mins

TE
P
So,
In 1000/15= 66.67 per min

N
Ped flow rate= 66.67 per
min/1.5= 44.44 ped/min/m
 L
Numerical Example #1—Solved

E
Step 6– Determine PLOS from Chart

P T
N Ped flow rate= 44.44 ped/m/min

The LOS of the footpath as per IndoHCM is ‘D’

 L
PLOS for Crosswalks at Signalized Intersections

E
As per HCM 2000
Here MOE is delay for waiting at the intersection, calculated by,

T
0.5 ∗ (𝐶 − 𝐺)2 Where, 𝑑𝑝 = average delay; C= Cycle Length
𝑑𝑝 = (s); G= effective green-time (s)
2∗𝐶

P
Cycle length (sec) is the time that it
takes a signal to complete one
full cycle of indications

N
Effective green time is the
actual time available for the
More wait-time vehicles to cross the
as we move to intersection, i.e. sum of all
green and yellow time minus
letter grade F lost time
 L
Numerical Example #2

E
Calculate the PLOS of pedestrian crossings at both major and minor streets of a signalized
intersection operating with, 80.0-s cycle length, and no pedestrian signals. Major street: Phase

T
green time, Gmaj = 44.0 s; Crosswalk length, Lmaj = 14.0 m; Minor street: Crosswalk length, Lmin =
8.5 m; Phase green time, Gmin = 28.0 s. Use HCM 2000 PLOS definition.

P
0.5 ∗ (𝐶 − 𝐺)2
𝑑𝑝 =
2∗𝐶

N
 L
Numerical Example #2—Solved

Major street
0.5∗(𝐶−𝐺𝑚𝑎𝑗)2 0.5∗(80−28)2

TE
P
𝑑𝑝,𝑚𝑎𝑗 = = = 16.9𝑠
2∗𝐶 2∗80
LOS B

N
Minor street
0.5∗(𝐶−𝐺𝑚𝑖𝑛)2 0.5∗(80−44)2
𝑑𝑝,𝑚𝑖𝑛 = = = 8.1 𝑠
2∗𝐶 2∗80
LOS A
 E L
P T
 Traffic and Highway Engineering. 5th Edition—Garber & Hoel. Cengage Learning

N
 Indian Highway Capacity Manual—Indo HCM (2011). CRRI-CSIR, GoI

 Highway Capacity Manual (HCM)– HCM 2000, Transportation Research Board USA
 E L
P T
• Lots of PLOS definition and ways to determine PLOS

• Most common PLOS type is using Flow Characteristics

N
• PLOS is classified based on MOEs

• IndoHCM determines PLOS based on the MOE Flow rate

• US HCM determines PLOS based on the MOE average delay

 E L
P T
N