This module is about estimation of times,
cost, and resources for activities.
In order to arrive at a schedule for the activities,
we need to estimate the activity durations and the resources required for each activity.
The resources required may be men, material,
equipment and funds, and may not be uniform throughout the duration of an activity.
While we would like accurate estimates,
it should be noted there already is some uncertainty as
to how long the activity would actually take to complete,
especially if the activity is being undertaken for the first time,
which is often the case in projects.
One should note that there's a trade-off between
getting more accurate estimates and the cost of obtaining increased accuracy.
Time estimates would be better after the individual work packages
are identified and for tasks that are to be done soon.
Long-duration projects increase the uncertainty and may have
only approximate time and cost estimates for activities that are to be done much later.
Organizational structure can also play a role in estimating the duration of activities.
A dedicated team may take less time for an activity because of
focused attention but the cost of resources of the dedicated team may be more.
Organization culture plays an important role
in what may be referred to as estimating gamesmanship.
If in an organization,
the project manager has a tendency to cut the estimates made by,
say, the person who has to do the task,
the next day I'm around,
that person may increase his or her estimate more than normally done so that after that,
the accepted duration is still high.
If the estimates of cost and duration are often padded,
we would expect most of the projects to be completed within budget and on time,
but this however is not the case.
As pointed out by Eliyahu Goldratt,
this may be due to one of several reasons.
We will briefly summarize these reasons next.
One possibility is the Murphy's Law,
which says that the work expands to fill that available time maybe operating.
The other possibility is the prevalence of what may be referred to as student syndrome,
which is to postpone starting the activity until the last minute.
This of course would mean that the buffer implied
by the inflated estimates may disappear quickly,
and any unforeseen events happening may trigger
a delay in the completion time of an activity and/or an increase in cost,
thereby affecting the completion time and/or the cost of the project.
The other possibility is that a person completing the task
early may not want to report the early completion of the task,
possibly fearing that his or her colleagues may not approve of
such actions given their own tardiness in completing the task assigned to them.
Even if an early completion of activities is reported,
the person who is responsible for the succeeding activity may not be ready to
start that activity early due to the other prior commitments made,
considering a planned start date for the activity.
It's also possible that persons responsible for
performing an activity may be doing excessive multitasking.
Finally, the required resources may not be available as per the schedule.
At that end of the spectrum is the time and cost estimates may be
deliberately made low so as to win the contract in a competitive bidding for the project.
At a later date but much before the completion date of the project,
the person or the organization that got the contract may start negotiating
for additional budget and/or delay in completing the project.
Although this may not be thought of as being unethical,
it is the dark side of competition.
The person or the organization sponsoring the project may stipulate
a fixed-price contract with penalties for the delay in the completion of the project.
But unfortunately, at least in India,
legally enforcing the contract is a time-consuming effort which is often better avoided,
resulting in an out-of-court settlement.
Next, we will look at some important factors
that need to be considered while making the estimates.
Clearly, persons familiar with the activities should make the estimates.
The estimate should be based on normal conditions and normal use of resources,
and it is better to avoid including contingencies while making the estimates.
The estimate should be made independently for each task,
and aggregation of all packages for estimation should be avoided.
Some of the other factors that should be taken into
account are holidays and vacation periods,
if any, price and wage increases,
wastage and spoilage, if appropriate.
Maintenance cost of resources,
such as machinery and equipment, etc.
should also be taken into account.
Risk assessment may lead to different estimates for the same activity.
This may sometimes bring out
the inherent contradictions and assumptions made while making the estimates.
Reconciliation of the different assumptions and
contradictions may lead to better estimates.
Estimates made by top or senior management is often referred to as top-down estimates,
while those made by the persons responsible for
work packages are referred to as bottom-up estimates.
Presumably, top-down estimates are for the project
and may be included in the project charter.
Top-down estimates are typically arrived at by
a consensus approach involving senior and/or middle-level managers.
If there are managers who have experience with similar projects,
top-down estimates can be reasonably accurate.
Bottom up estimates may form the basis for inclusion in
the project scope after some reconciliation between the top-down and bottom-up estimates.
Bottom up estimates are presumably more
accurate since they are made by persons responsible for the work packages,
and they may have some experience with similar,
if not identical, activities.
Three popular estimating procedures are: the ratio or the parametric method,
as it is referred to,
the template method, and the apportion method.
The ratio or the parametric approach is to first identify the different parameters
and their respective impacts on the total cost and/or the duration of the project.
The next step is to estimate the value of
these different parameters for the project and their consideration.
Depending upon the values of these parameters and the respective impacts,
the total cost and time are estimated,
assuming essentially a linear relationship.
For instance, the project is to build a factory.
The cost may be estimated considering the plant size,
the type for the plant, etc.
For a software development project,
the effort required may be estimated by parameters or just lines of code,
function points, object points,
test points, use case point, etc.
The template method, as the name suggest,
uses templates that have been prepared
based on the past data obtained from experience with similar projects.
Differences in the new project as compared to
the projects executed in the past are noted,
and adjustments in the cost estimates are made depending upon the differences.
Clearly, this method is useful only if the project under
consideration is similar to the projects executed in the past.
The apportion method can give good estimates if the features
and costs of the project that is being undertaken are
similar to projects executed in the past.
First, the total cost of the project is estimated based on
past experience or some knowledge of
the cost of similar projects done in the past by other organizations.
Then the estimated total cost of the project is apportioned
to various deliverables, sub-deliverable, etc.
using the work break down structure.
Finally, adjustments in some of the costs including possibly the total cost of
the project are made if any of the allocated costs are viewed as unreasonable.
For instance, in the diagram,
which is adapted from Figure 5.1 in "Project
Management: The Managerial Process" by Clifford Gray,
Erik Larson and Galton Desai,
fourth edition, special Indian edition,
Tata McGraw-Hill Education Private Limited, in 2010,
the total software development cost is first estimated to be,
let us say, $200,000.
Thus, design specifications, software development, testing,
and documentation have been estimated to cost $60,000,
$80,000, $50,000, and $10,000, respectively.
The design specifications are further broken down into three work packages denoted as D1,
D2, and D3 respectively.
Typically, work packages D1,
D2, and D3 cost 30%, 40%, and 30% percent,
respectively, of the total cost of design specifications.
Thus, D1 is estimated to cost 30% of $60,000, that is,
$18,000 while D2 and D3 are estimated to cost $24,000 and $18,000, respectively.
Software development is further broken down
into two work packages denoted as S1 and S2, respectively.
Typically, work packages S1 and S2 cost 40% and 60%,
respectively, of the total cost of software development.
Thus, S1 is estimated to cost 40% of $80,000,
that is, $32,000 while S2 is estimated to cost $48,000.
Testing is further broken down into two work packages denoted as T1 and T2, respectively.
Typically, work packages T1 and T2 cost an equal amount,
which in this case would be $25,000 each.
Documentation is treated as a single work package costing $10,000.
After these estimates are made,
an assessment is made as to whether the cost of any of
the work packages is unreasonably high or low.
If so, adjustments are made in the cost of some of the work packages which, in turn,
would necessitate a change in one or more of the costs of the four deliverables which,
in turn, may change the estimated total cost of the project.
For software development projects,
some models such as COCOMO (which stands for
"Constructive Cost Models"), neural networks, etc.
have been attempted, but these models have been found
to be of limited use because of the difficulty of not
only getting the appropriate data but also getting
good estimates of the necessary parameters which may,
in turn, affect the accuracy of the estimated cost and/or the times.
Resources that are required for an activity may be people,
material, equipment, and working capital.
Material or equipment requirements depend upon
the type of activity and are estimated from experience.
Requirements of people with the necessary skills is
estimated based on the type of activity and the effort required.
The working capital requirements are determined from
the estimated total cost and time for the activity.
The total cost for completing an activity is broken down
into working capital requirements for each unit of time,
which may be a day, a week, or a month,
as the case may be for the duration of the activity.
The working capital requirements,
when incorporated into the schedule – that is,
the start time and end date for each activity of
the project – gives the so-called time-phased budget for the project.
The time-phased budget is used for monitoring the project while it is in progress.
The monitoring and control of a project are covered in a subsequent module.
This completes the module on estimation of cost,
times, and resources required.
