CECOS UNIVERSITY PESHAWAR

CONSTRUCTION PLANNING, SCHEDULING & CONTROL (CE-610)

LECTURE NO 9
Critical Path Method

Dr. Syed Azmat Ali Shah

[email protected]
CECOS University Peshawar

The Critical Path Method

The most widely used scheduling technique is the critical path method (CPM) for

scheduling. This method calculates the minimum completion time for a project

along with the possible start and finish times for the project activities. Many texts

and managers regard critical path scheduling as the only usable and practical

scheduling procedure.

Computer programs and algorithms for critical path scheduling are widely

available and can efficiently handle projects with thousands of activities.

Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 2
CECOS University Peshawar

The Critical Path Method

The critical path itself represents the set or sequence of activities which will take

the longest time to complete. The duration of the critical path is the sum of the

activities‘ durations along the path. Thus, the critical path can be defined as the

longest possible path through the "network" of project activities. The duration of

the critical path represents the minimum time required to complete a project. Any

delays along the critical path would imply that additional time would be required

to complete the project.

Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 3
CECOS University Peshawar

The Critical Path Method

There may be more than one critical path among all the project activities, so

completion of the entire project could be delayed by delaying activities along any

one of the critical paths. For example, a project consisting of two activities

performed in parallel that each requires three days would have each activity

critical for a completion in three days. Formally, critical path scheduling assumes

that a project has been divided into activities of fixed duration and well defined

predecessor relationships. A predecessor relationship implies that one activity

must come before another in the schedule.

Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 4
CECOS University Peshawar

The Critical Path Method

The CPM is a systematic scheduling method for a project network and involves

four main steps:

- A forward path to determine activities early-start times;

- A backward path to determine activities late-finish times;

- Float calculations; and

- Identifying critical activities.

Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 5
CECOS University Peshawar

Calculations for the Critical Path Method

The inputs to network scheduling of any project are simply the AOA or the AON
networks with the individual activity duration defined. The network scheduling
process for AOA and AON networks, however, is different. To demonstrate these
two techniques, let’s consider a simple 5-activity project, with activity A at the
start, followed by three parallel activities B, C, and D; which are then succeeded
by activity E. The AOA or the AON networks of this example are presented in
Figure. Detailed analysis of theses AOA or the AON networks are presented in
the following slides. It is noted that the example at hand involves only simple
finish-to-start relationships among activities.

Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 6
CECOS University Peshawar

Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 7
CECOS University Peshawar

Activity-On-Arrow Networks Calculations

The objective of arrow network analysis is to compute for each event in the
network its early and late timings. These times are defined as:
 Early event time (ET) is the earliest time at which an event can occur,
considering the duration of preceding activities.
 Late event time (LT) is the latest time at which an event can occur if the
project is to be completed on schedule.

Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 8
CECOS University Peshawar

Forward Path
The forward path determines the early-start times of activities. The forward path
proceeds from the most left node in the network (node 1 – Figure) and moves to
the right, putting the calculations inside the shaded boxes to the left.
Each node in the network, in fact, is a point at which some activities end (head
arrows coming into the node), as shown in Figure. That node is also a point at
which some activities start (tail arrows of successor activities). Certainly, all
successor activities can start only after the latest predecessor is finished.
Therefore, for the forward path to determine the early-start (ES) time of an
activity, we have to look at the head arrows coming into the start node of the
activity. We then have to set the activity ES time as the latest finish time of all
predecessors.

Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 9
CECOS University Peshawar

Preparation for the forward path

A node in an AOA network

Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 10
CECOS University Peshawar

In this example, the forward path calculations are as follows:

 Start at node 1, the first node of the project, and assign it an early-start time of
zero. Here, all activity times use an end-of-day notation. Therefore, the ES of
activity A is zero means that activity starts at end of day zero, or the
beginning of day 1 in the project.
 Then, move to node 3. This node receives one head arrow, and as such, it has
one predecessor, activity A. Since the predecessor started on time zero and
has 3 days duration, then, it ends early at time 3 (Early-Finish (EF) = Early-
Start (ES) + d). Accordingly, the ES time of all successor activities to node 3
(activities B, C, and D) is time 3. This value is therefore, put in the shaded
box on top of node 3, as shown in Figure.

Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 11
CECOS University Peshawar

Forward path calculations in AOA networks

Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 12
CECOS University Peshawar

 Now, move forward to successor nodes 5, 7, and 9. However, since node 9 is

linked to nodes 5 and 9 by dummy activities, we begin with nodes 5 and 7. Node
5 receives one head arrow from its predecessor activity B, we evaluate the EF
time of B as 6 (ES (3) + d (3)). Successor activities to node 5, therefore, can
have an ES time of 6. Similarly, the ES time at node 7 is calculated as time 9.
 Moving to node 9, the EF times of its 3 predecessors (d1, C, and d2) are time 6,
7, and 9, respectively. Accordingly, the ES time of successor activities is the
largest value 9. Notice that only the largest EF value of predecessor activities is
used to calculate the ES of successor activities and all other values not used. As
such, only ES values can be directly read from the calculations in Figure. EF
values, on the other hand can be calculated as EF = ES + d.

Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 13
CECOS University Peshawar

 The last node (11) receives one head arrow, activity E which has an ES value
of 9. The EF time of activity E, therefore =9 + 5 = time 14. Since node 11 is
the last node, the EF of this node becomes the end of the project, reaching
total project duration of 14 days.
Generally, for any activity x connecting between nodes i and j as shown in
Figure, the calculations as follows:

ETj = ETi + dx (1)

In case of more than one arrow terminating at node j, then consider the largest
value. Accordingly,
ESx = ETi (2)
EFx = ESx +dx (3)

Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 14
CECOS University Peshawar

Backward Path
The backward path determines the late-finish (LF) times of activities by
proceeding backward from the end node to the starting node of the AOA
network. We put the LF values in the right side boxes adjacent to the nodes, as
shown in Figure. For the example at hand, we do the following:

Backward path calculations in AOA networks

Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 15
CECOS University Peshawar

 Start from the last node of the network (node 11) and we transfer the early-
finish value from the left box to be the late-finish (LF) value at the right side
box.
 Then, move backward to node 9 which has only one tail arrow of activity E.
With the LF time of E being time 14, its LS time becomes LS = LF - d = 14 –
5 = time 9. At node 9, therefore, time 9 becomes the LF time of the
predecessor activities of this node.
 Moving backward to predecessor nodes 5, and 7. Node 5 has one tail arrow of
the dummy activity d1, and as such, the LF time value to be used at node 5
becomes 9. Similarly, the LF time value of node 7 becomes 9.

Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 16
CECOS University Peshawar

 Moving to node 3, we evaluate the LS time of its 3 successor activities B, C,

and D as 6, 5, and 3, respectively. The LF time at node 3, therefore, becomes
the smallest value 3. With other LS values not used, the values in the
calculation boxes, as such, directly show the LF times of activities. LS times
can be calculated as LS = LF – d.
 Now, proceed to the first node in the network (node 1). It connects to one tail
arrow of activity A. The LS time of A, therefore, is LS = LF – d = 3 – 3 = 0, a
necessary check to ensure the correctness of the calculation.
 Having Figure again in mind and to generalize the calculations, for any
activity x connecting between nodes i and j, the calculations as follows:
LTi = LTj - dx (4)
 In case of more than one arrow leaving node i, then consider the smallest
value.
 Accordingly, LFx = LTj (5)
LSx = LFx –dx (6)

Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 17
CECOS University Peshawar

Float Calculations
Once forward path and backward path calculations are complete, it is possible to
analyze the activity times. First, let's tabulate the information we have as shown
in Table. One important aspect is Total-Float (TF) calculations, which determine
the flexibility of an activity to be delayed. Notice in Table that some activities
such as activity A has ES time = LS time, and its EF time = LF time, indicating
no slack time for the activity. Other activities such as B can start early at time 3
and late at time 6, indicating a 3-day of total float. Float calculations can be
illustrated as shown in Figure for any activity.

Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 18
 CECOS University Peshawar

Early Start Late Finish Late Start Early Finish Total Float Critical
Activity Duration
(ES) (LF) (LS) (EF) (TF) Activity

A 3 0 3 0 3 0 Yes

B 3 3 9 6 6 3 No

C 4 3 9 5 7 2 No

D 6 3 9 3 9 0 Yes

E 5 9 14 9 14 0 Yes

Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 19
CECOS University Peshawar

Figure shows two ways of scheduling each activity using its activity times. One
way is to schedule it as early as possible (using its ES time). The other way is as
late as possible (using its LS time). The activity float can, therefore, be
represented by the following relationships:
Total Float (TF) = LF – EF (7)
= LS – ES (8)
Also, with the ES and LF times directly read from the boxes used in forward and
backward path calculations, the total float can also be calculated as; TF = LF –
ES – d. Using these relationships, activities total floats are calculated as shown in
Table.

Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 20
CECOS University Peshawar

Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 21
CECOS University Peshawar

Another type of float often used in network analysis is the Free Float, which can
be calculated as:
Free Float (FF) = ETj – ETi – d (9)
or FF = smallest ES (of succeeding activities) – EF (of current activity) (10)
The free float defines the amount of time that an activity can be delayed without
affecting any succeeding activity. With free float available for an activity, a
project manager knows that the float can be used without changes the status of
any non-critical activity to become critical.

Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 22
CECOS University Peshawar

Identifying the Critical Activities

Activities with zero total floats mean that they have to be constructed right at their
schedule times, without delays. These activities are considered to be critical. They
deserve the special attention of the project manager because any delay in critical
activities causes a delay in the project duration.
One interesting observation in the results of CPM analysis is that critical activities
form a continuous path of the critical activities that spans from the beginning to the
end of the network. In our example, activities A, D, and E (excluding dummy
activities) are critical and the critical path is indicated by bold lines on Figure 4.6.
Notice that among the 3 paths in this example (A-B-E; A-C-E; and A-D-e), the
critical path is the longest one, an important characteristic of the critical path. In real-
life projects with many activities, it is possible that more than one critical path are
formed. By definition, the length of these critical paths is the same.
Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 23
 CECOS University Peshawar

Precedence Diagram Method (PDM)

Precedence Diagram Method (PDM) is the CPM scheduling method used for AON
networks and it follows the same four steps of the CPM for AOA method.
Forward Path
Forward path can proceed from one activity to the other; the process is as follow
(Figure):
- At activity A. It is the first activity in the network. We give it an early-start (ES) of 0
in the left top box. Adding the activity duration, we determine the EF time of the
activity and we put it in the top right box.
- Then, move forward to the succeeding activities B, C, and D. These three activities
have only A as a predecessor with time 3 as its EF. As such, all the three activities
can start as early as time 3 (ES = 3). Each activity, accordingly, has its own EF time
based on its duration.
Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 24
CECOS University Peshawar

Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 25
CECOS University Peshawar

- Moving forward to activity E. This activity has 3 predecessors (3 head arrows)

of activities B, C, and D with their largest EF time being 9. The ES of activity E,
thus, with becomes time 9. Adding its duration, the EF becomes time 14.
To generalize the calculations consider Figure, of two activities i and j with
relationship finish to start and overlap between them. Overlaps will have a
positive sign, while lags will have a negative sign. The forward path calculations
are as follows:

Activities times in PDM Analysis

ESj = EFi - overlapij (11)
In case of more than one activity precedes activity j then consider the maximum. Then,
apply Equation 3 to calculate the early finish times.
Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 26
CECOS University Peshawar

Backward Path
Once the forward path is finished, the backward path can start, moving from the
last activity to the first, putting the calculations in the bottom two boxes of each
activity, as shown in Figure. The process is as follows:

Backward path in PDM analysis

Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 27
 CECOS University Peshawar

- Start at the last activity E and we transfer the early-finish value to become the
activity's late-finish (LF) time. Then, subtracting the activity's own duration, the
latestart (LS) time is calculated as time 9 and put in the bottom left box of the
activity.
- Moving backward to activities B, C, and D all have one successor (activity E) with
LS time of 9. The LF of all these activities becomes time 9. Each activity then has its
own LS time, as shown in Figure.
- Moving to activity A. The activity is linked to 3 tail arrows (i.e., has 3 successors)
of activities B, C, and D. The LF of activity A, thus, is the smallest of its successors'
LS times, or time 3. Activity A then has LS equals zero.
Considering Figure again, the backward path calculations are as follows:
LFi = LSj + overlapij (12)

Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 28
CECOS University Peshawar

In case of more than one activity succeeds activity j then consider the minimum.
Then, apply Equation 6 to calculate the late start times.
Notice that by the end of the backward path, all activity times can be read
directly from the boxes of information on the activity, without additional
calculations. This also, makes it simple to calculate the total float of each activity
using the same relationships used in the AOA analysis.

Identifying Critical Activities

Critical activities can also be easily determined as the ones having zero float
times, activities A, D, and E. The critical path is then shown in bold as Figure.
The PDM analysis, as explained, is a straight forward process in which each
activity is considered as an entity that stores its own information.
Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 29
CECOS University Peshawar

Time-Scaled Diagrams

Time-scaled diagrams are used extensively in the construction industry. Such

diagrams enable one to determine immediately which activities are scheduled to

proceed at any point in time and to monitor field progress. Also, it can be used to

determine resources need. The time scale used in time-scaled diagrams can be

either the calendar dates or the working periods (ordinary dates), or using both at

the same time.

Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 30
CECOS University Peshawar

Time-Scaled Diagrams

The activities are represented as arrows that drawn to scale to reflect the activity

duration it represents. The horizontal dashed lines represent total float for groups

of activities and free float for the immediate activity to the left of the dashed line.

The precedence of an activity is the immediate activities before it or that linked

to it through vertical dashed lines. The name and the duration of an activity are

written above and below the arrow representing it respectively (Figure). The ES,

EF, and FF times of the activities can be easily read directly from the diagram.

The TF for an activity is the smallest sum of succeeding FF on all paths.

Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 31
CECOS University Peshawar

Time-Scaled Diagrams

Accordingly, the LS and LF times can be easily calculated as follows:

LSi = ESi + TFi (13)

LFi = LSi + Di (14)

The critical path can be easily determined as the continuous lines from the

beginning to the end of the network with any dashed lines. The main advantage

of this diagram is its simple representation and it can be used directly for

determining resources need.

Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 32
CECOS University Peshawar

However, its disadvantage is that it needs a great effort to be modified or

updated. Also, it cannot be used to represent overlapping activities. Figure shows

the time-scaled diagram for the same 5-activities project solved previously using

AOA and AON networks.

Time-scaled diagram

Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 33
CECOS University Peshawar

The TF for activity A equals the smallest of the sum of the floats along all paths

from the end of activity A to the end of the project. The float on path ABE = 3,

path ACE = 2 and path ADE = 0, then the TF of activity A = 0. The calculations

are shown in Table.

Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 34
CECOS University Peshawar

Schedule Presentation

After the AOA and AON calculations are made, it is important to present their

results in a format that is clear and understandable to all the parties involved in

the project. The simplest form is the Bar chart or Gantt chart, named after the

person who first used it. A bar chart is a time versus activity chart in which

activities are plotted using their early or late times, as shown in Figures a and b.

Early bar chart is drawn using the ES times of activities, while the late bar chart

is drawn using the LS times.

Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 35
CECOS University Peshawar

a) Early bar chat b) Late bar chart

Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 36
CECOS University Peshawar

The bar chart representation, in fact, shows various details. Float times of

activities, critical activities can be shown in a different color, or bold borders, as

shown in Figure. The bar chart can also be used for accumulating total daily

resources and / or costs, as shown at the bottom part of Figure. In this figure, the

numbers on each activity represent the number of labors needed.

One additional benefit of the bar chart is its use on site to plot and compare the

actual progress in the various activities to their scheduled times. An example is

shown on Figure, showing actual bars plotted at the bottom of the original bars of

the schedule.
Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 37
CECOS University Peshawar

Using bar chart to accumulate resources

Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 38
CECOS University Peshawar

Criticisms to Network Techniques

The CPM and PDM analyses for network scheduling provide very important
information that can be used to bring the project to success. Both methods,
however, share some drawbacks that require special attention from the project
manager. These drawbacks are:
 Assume all required resources are available: The CPM calculations do not
incorporate resources into their formulation. Also, as they deal with activity
durations only, it can result in large resource fluctuations. Dealing with
limited resources and resource leveling, therefore, has to be done separately
after the analysis;

Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 39
CECOS University Peshawar

 Ignore project deadline: The formulations of CPM and PDM methods do not
incorporate a deadline duration to constrain project duration;

 Ignore project costs: Since CPM and PDM methods deal mainly with
activities durations, they do not deal with any aspects related to minimize
project cost;

 Use deterministic durations: The basic assumption in CPM and PDM

formulations is that activity durations are deterministic. In reality, however,
activity durations take certain probability distribution that reflect the effect of
project conditions on resource productivity and the level of uncertainty
involved in the project.
Dr. Syed Azmat Ali Shah Construction Planning, Scheduling & Control (CE-610) 40