USING COBB DOUGLAS PRODUCTION FUNCTION IN

CONTRUCTION INDUSTRY FOR TIME COST ANALYSIS

A Project report Submitted for the course:

ENGINEERING ECONOMICS AND DECISION ANALYSIS HUM1036

Submitted by
Raunak jain 16BCL0035
Harsh Kishore 16BCL0222

Under the guidance of

DR. SAVITHA N .
Assistant professor (Sr.)
Department of Social Sciences
School of Social Sciences and Languages
VIT ,VELLORE - 632014

APRIL, 2019
 DECLARATION

We hereby, Harsh Kishore - 16BCL0222, Raunak Jain -16BCL0035 declare that

the thesis entitled “Using Cobb Douglas Production Function in Construction Industry
for Time Cost Analysis”, submitted by us to VIT, Vellore in partial fulfilment of the for the
requirement of the course “Engineering Economics and Design Analysis”, with course code
HUM1036, of the degree of Bachelor of Technology is a bonafide work carried out by us
under the supervision of Dr. Savitha N., Assistant Professor (Sr.),Department of Social
Sciences, School of Social Sciences and Languages, VIT, Vellore – 632014.We further
declare that the work reported in this thesis has not been submitted and will not be submitted,
either in part or in full, for the award of any other degree or diploma in this institute or any
other institute or university.

Place : Vellore
Date : Signature
 CERTIFICATE

This is to certify that the in-plant training report entitled “Using Cobb Douglas
Production Function in Construction Industry for Time Cost Analysis”, submitted by
Harsh Kishore-16BCL0222, Raunak Jain-16BCL0035 to VIT Vellore, in partial fulfillment
of the requirement of the degree of Bachelor of Technology is a bonafied work carried out
by him/her under my supervision. The project fulfills the requirements as per the regulations
of this institute and in my opinion meets the necessary standards for submission. The contents
of this report have not been submitted and will not be submitted either in part or in full, for
the award of any other degree or diploma in this Institute or any other Institute or University.

Place : Vellore
Date : Signature of the Guide
 ACKNOWLEDGEMENT

At the outset, we thank the almighty God for his blessings for granting us the
knowledge and right aptitude to successfully complete our project work.

We would like to express our special gratitude thanks to our guide Dr. Savitha N.
Department of Social Science, School of Social Sciences and Languages where esteemed
guidance and immense support encouraged us to complete the project successfully.

We wish to express our sincere thanks to honourable Chancellor Dr.G.Viswanathan;

esteemed Vice President Shri Sankar Viswanathan; Shri Sekar Viswanathan and Shri
G.V. Selvam; respected Vice Chancellor Dr.Anand A.Samuel; respected Pro-Vice
Chancellor Dr. S Narayanan of this prestigious VIT, Vellore for providing us an excellent
world class academic environment and facilities for pursuing our B.Tech Programme.

We also thank all the faculty members of the Department of Social Sciences and
faculty of other departments of the School of Social Sciences and Languages and the non-
teaching staff for giving us the courage and strength that we need to achieve my goals.

Our Special thanks to our friends and their timely help and suggestions rendered for
the successful completion of this project.

This acknowledgement would be incomplete without expressing our whole hearted

thanks to our parents for their continuous support and guidance in all walks of our life.

Student’s Signature
 CHAPTER 1
INTRODUCTION
1.1 Introduction
Time-cost tradeoff problem (TCTP) has been extensively studied in construction
scheduling research to find out the best solutions in terms of minimizing the cost while
shortening the project duration. Many different algorithms and assumptions were used in
searching for the best solutions which are explained in detail later. Despite many existing
researches on TCTP, very little study can be found in exploring or explaining the source or
the origin of the cost increase during activity crashing. Instead, in many studies, the cost
functions associated with crashing were assumed, or based on historical data, or based on
simulation results. Being able to explain quantitatively where the increased crashing costs
come from is important to better understand the theoretical fundamental of TCTP.

There has been a lack of theoretical base to model the cost functions associated with
activity crashing. Evensmo and Karlsen (2008) were among the few researchers tried to
explain the origin of cost increase during activity crashing. However, in their study, they only
discussed the causes of cost increasing due to labor input changes. A significant limitation in
their approach is the lack of consideration of changes of equipment inputs during crashing
activities

Cobb-Douglas production function (CDPF)has been widely used in research on

economics (Meeusen and van den Broeck 1977, Dennis et al. 2010), technology progress and
productivity.

Where: Q = production rate; L = labor input; K = capital/equipment input; A =

technology; α and β are the output elasticity of labor and capital respectively.

In particular, CDPF models production rate, labor inputs, equipment/capital inputs

and technology efficiency in a very elegant formation, which can be used to explain many
types of production activities. Some important features of CDPF (Fig. 1) are very useful in
explaining the origin of the crashing cost under many different situations
 Fig 1.2

An Isoquant is a contour line drawn through the set of points at which the same
quantity of output is produced while changing the quantities of inputs. Fig. 1 shows a set of
isoquants for a production function with two inputs of capital (K) and labor (L). K is
equivalent to or interchangeable with equipment in this study. One important feature of
CDPF is reflected by the summation value of α and β. When the summation of α and β is less
than 1 which is called Decrease Return to Scale (DRTS), the double inputs of L and K will
generate less than double output of Q. This is illustrated by case from point C to D in Fig. 1.
When the summation of α and β is equal to 1, Constant Return to Scale (CRTS), the double
inputs of L and K will generate double output of Q, which is illustrated by case from point B
to C. When the summation of β and α is greater than 1, Increase Return to Scale (IRTS), the
double inputs of L and K will generate more than double output of Q, which is illustrated by
case from point A to B

CDPF provides a feasible tool to model construction schedule crashing activities

especially modeling the time-cost trade-off problem. The origin or source of the cost

increase in construction TCTP can be theoretically explained using CDPF. The duration
crashing is achieved through increasing either L or K or both, so to increase Q in a given time.
 Applying CDPF with cost functions of both labor and equipment provide a potential
way to incorporate detailed labor and equipment costs and utilization information into the
time-cost optimization model in construction TCTP.

1.2 Statement of the Problem

In the statement of the problem we tried to justify how the efficiency of labor and
machines can be effected and what are the factors that can effect the time and cost of the
project

1.3 Objectives of the Study

1. To study about the company L&T in detail

2. To study the decrease return to scale and inflation effect on the company

3. To study the financial and cost analysis of L&T

1.4 Methodology

The various methodology was adopted for the analysis purpose of the project are data
collection method and the statistical tools used for the analysis. A. Data Collection The
secondary data was collected for ten years from 2008 to 2018. The dataset collected for the
analysis in this project is Amigobulls.
 CHAPTER 2
REVIEW OF LITERATURE

Existing researches on construction TCTP can be classified based on cost functions,

solution methods, objective functions, and models.

2.1. Construction Crashing Time-Cost Function

The relation between cost and the time has been well studied in various studies. Cost
functions such as linear (Bazaraa and Shetty 1979, Fulkerson 1961), nonlinear (Moussourakis
and Haksever 2010), discrete (Kelly, 1961, this research), convex (Lamberson and Hocking
1970, Demeulemeester et al. 1993), and concave (Berman, 1964) have been implemented in
the studies on TCTP hitherto.

Objective function in construction crashing cost analysis may vary significantly.Some

researchers consider multi-objective function and assume priority for time and cost, and
based on that, they try to find the optimum solution. Some consider minimizing total project
cost. Some try to minimize total project cost or duration subject to predefined constraints.
Moussourakis and Haksever (2010) considered three objective functions including
minimization of project completion subject to a crash budget constraint, minimization of the
total project cost, and minimization of the total cost under late completion penalties.
Leu et al. (2001) tried to determine the project completion time regarding project
total cost which includes both direct and indirect costs. Some other authors consider a limited
budget for the project and try to minimize the project duration regarding the budget
constraints.
Jaskowski and Sobotka (2006) proposed a multi-criteria objective function which
considers both time and cost together, and they end up with a Pareto set. They used an
evolutionary algorithm to compare different solutions based on fitness values.
One of the main assumptions or constraints, which are commonly considered in the literature,
is the limitation of resources. Jaskowski and Sobotka (2004) try to minimizethe project
completion time regarding the limitation on resources. They also proposed an evolutionary
algorithm to assign recourses to activities in a proper time.
Other authors chose a different approach to cope with this problem. To avoid delay in
projects, Lin et al. (2011) predict construction project completion time using historical data.
They used regression model to forecast the future projects completion time. Chen et al. (2011)
developed a cash-payment model for forecasting the cash flow. They evaluated their model
by comparing two historical real data set.

2.2. Models

There are three main categories in existing literature regarding models or assumptions
in time-cost trade-off problem. 1) Deterministic relationship between time and cost was
assumed in Gerk and Qassim (2008), Moussourakis and Haksever (2010), Pollack- Johnson
and Liberatore (2006). 2) Stochastic relation between time and cost was assumed in Aghaie
and Mokhtari (2009), Cohen et al. (2007), Ke et al. (2009). 3) Fuzzy relation between time
and cost was assumed by Ghazanfari et al. (2009).
Aghaie and Mokhtari (2009) proposed a nonlinear mix integer programming to
increase the probability of completion of the project in a given due date. They also assume
that each activity duration follow an exponential distribution. Ke et al. (2009) proposed
integrating stochastic simulation and genetic algorithm to increase the probability of
completion of a project by a specific due date. Cohen et al. (2007) wanted to minimize the
expected cost related by the project.
Ghazanfari et al. (2009) assumed fuzzy variables. Via Possibility Goal Programming,
the cost was minimized while considering the minimum duration. The main contribution of
this fuzzy approach is the use of vagueness in the cost function during the project execution.
Zheng and Ng (2005) also presented fuzzy set theory regarding the uncertainty included in
TCTP problem. They also used GA as a meta-heuristic algorithm to develop a Pareto set of
time and cost.
 CHAPTER 3
PROFLE OF THE INDUSTRY

3.1. About the Company

L&T Technology Services Limited (LTTS) is a global leader in Engineering and

R&D (ER&D) services. With 350+ patents filed for 51 of the Global Top 100 ER&D
spenders, LTTS lives and breathes engineering. Our innovations speak for itself – World’s
1st Autonomous Welding Robot, Solar ‘Connectivity’ Drone, and the Smartest Campus in the
World, to name a few.

LTTS’ expertise in engineering design, product development, smart manufacturing,

and digitalization touches every area of our lives - from the moment we wake up to when we
go to bed. With 48 Innovation and R&D design centers globally, we specialize in disruptive
technology spaces such as 5G, Artificial Intelligence, Collaborative Robots, Digital Factory,
and Autonomous Transport.

3.2 L&T Technology Services

L&T Technology Services.., a subsidiary of Larsen & Toubro, is an engineering

services company that operates in the global Engineering, Research and Development
(“ER&D”) space. L&T Technology Services offers design, development and testing services
for the industrial products, medical devices, transportation, aerospace, telecom and process
industries. The company serves customers across the product engineering life cycle from
product conceptualization to implementation. Services include consulting, design,
development, testing, maintenance and to-market integration services.L&T Technology hits
the Indian Capital Markets with its IPO offering 10.4 million share at a price band of Rs.850
to Rs.860.
 CHAPTER 4
RESULTS AND ANALYSIS

Table 4.1

Fig 4.2
 Fig 4.3

Fig 4.4

Fig 4.5
 Fig 4.6

Fig 4.7
Fig 4.9

Fig 4.10
 Fig 4.11

Fig 4.12

In this section, we do sensitivity analysis on labor cost ( ), and capital cost The main
reason is that in the real world, in most cases, we are not sure about the future prices of labor
and capital (regarding Cobb-Douglas production function). Moreover, efficiency of the labor
and capital is different from a project to the other one.

Sensitivity analysis is applied to be ensuring of risk associated with labor and

equipment costs and efficiencies. That is to say, in the real world construction projects, we
cannot guarantee a specific value for the summation of labor and equipment efficiencies ( ).
The same story is true about the labor and equipment costs ( ). So, in the planning phase,
managers or decision makers to avoid the associated risks which are arisen from uncertainties
in determining labor and equipment cost and efficiencies, it is reasonable to analyze different
scenarios for those.

So according to the Cobb-Douglas production, which is the main approach of this

study, time and cost of the project can be varied in two ways considering the impact of labor
and equipment factors:

1) In the first category, we analyze the impact of the labor and equipment efficiency. In
reality, we are not sure about the labor and equipment efficiency which is presented as
elasticity in Cobb-Douglas production function. This means various internal sources may
affect the workers or equipment efficiency. Hiring unskillful workers or other equipment that
are not that much compatible with the specific project can decrease the efficiency. For
example, we may use new and sometimes more technical equipment which the labors were
not taught to use. Inthe literature, authors like Mateescu (2010) did regression analysis to
have a forecast of what are labor and equipment efficiencies ( ). So here, we do sensitivity
analysis for better understating how they impact the duration and total cost. We consider that
the summation of DRTS which is more common in reality. So, we will be aware of how
would be the allocation of labor and equipment when they are varied.
2) The other scenario that can be assumed for the sensitivity analysis is theimpact of
labor and equipment costs. One of the main issues that construction project managers are
coped with is cost inflation from the planning date to the date that start constructing. In the
literature it is known as market influence.

4.1. Decrease Return to Scale (DRTS) Efficiency

In the first case, we just consider internal sources that decrease the efficiency of labor
and equipments associated with construction projects. The reason for this assumption is that
in the real world mostly adding more labors and equipments do not exactly add that much
value to the production rate (Q). So, it is reasonable to assume that is DRTSTo do so, we

assume uncertainty of . In this case, we assume that which is more

compatible with the real world. In fact, in most cases in the real world the production
function is decreasing return to scale. For example, let assume a case that there is not enough
space to do a task. If we add a labor to workers not only he may not add value to the work;
but also, he may cause problems, and other feels they cannot work convenient. Moreover we
assume that of the . The rest of tha would be .

Fig. 4.13

According to Figure 15, total cost is presented based on the summation of . The result
of the simulation confirms our expectation. As expected, as the summation decreased, the
total cost increased; but the interesting point is that the total cost can be increased to 50 times
more than the case that the summation is equal to 1. The reason is the specific conditions of
Cobb-Douglas production function. That is, since elasticity factors are the power of labor and
equipment, even a little change I them, can cause a huge difference to the total production
value and so the cost.

4.2. Market Influence (Cost Inflation)

As stated earlier, market can influence the labor and equipment cost. We consider this
uncertainty as an external source of uncertainty to do analysis. So, according to equation ,Cl
and Ck and are considered as stochastic ones. We assume three different scenarios for this
condition.
In the first case, based on expert point of view, we assume that there would be an
inflation rate of 5 to 15% for labor cost and 3 to 5% for capital cost. For the second case it is
vice-versa. And in the last scenario, we assume higher inflation rate for both of labor and
capital, 5 to 15%. All of these scenarios are realistic, and in some economic conditions they
may be applied.
One of the most interesting points that can be achieved from this section is that,
regardless what the labor and capital cost is, the optimum duration based on the objective
function (equation 11) is 83 days. Now, it may not be that much surprising according to
previous section analysis. As stated, when the summation of tends to be 1, the duration
becomes 83 days. So, in these cases, since this summation is equal to 1, the optimum
duration would not be changed.

4.3 Considering Both Inflation rate and DRTS Efficiency

In this section, we again assume the same previous three scenarios; but this time

is again randomly distributed between 0.6 and 1.

In these cases, we consider both internal and external sources of uncertainties. That is,

efficiency ( ) is assumed DRTS which is more compatible with the real world, and
also we consider inflation for both labor and equipment costs.

As illustrated, like the case that we just assume uncertainty in efficiency of labor and
equipment (internal sources of uncertainties), duration varied. So it can be understood from
these figures that when there is an internal source of uncertainty which causes DRTS
efficiency, model chooses different duration based on the summation of efficiency.
 Chapter 5
Summary and Conclusion

A framework using Cobb-Douglas production function (CDPF) to solve construction time-

cost trade-off problem (TCTP) is proposed in this paper. Within the framework, a multi-
objective optimization method utilizing Hybrid Genetic Algorithm is presented with a case
application. The proposed algorithm tries to improve some deficiencies which were available
in the previous studies such as improving the randomly generation of the initial solutions via
2-opt procedure, and using roulette wheel selection method.

Then, sensitivity analysis on budget and duration needed for completion of the project was
conducted. A significant advantage of introducing CDPF into TCTP is that CDPF can be
used to quantitatively explain the origin of the crashing costs from both labor and equipment
perspective, which was a fundamental gap in previous research on TCTP.

The results suggested that, by tying CDPF to TCTP, the proposed approach is capable to
identify optimal labor and equipment allocation solution effectively to satisfy the need for
duration reduction. Indeed, more than the combination of total cost and duration of a project,
now we can analyze what combination of labor and equipment could minimize the total cost.

In this study, in three different categories, sensitivity of the model to the parameters are
analyzed. First, only the internal sources of uncertainty are assumed. In doing so, it is
assumed that the summation of and is DRTS (it is varied between 0.6 to 1). According to the
fact that these parameters are have impact as power of labor and equipment regarding
equation 1, they highly affect the total cost even to 50 times more than the case the
summation of and is equal to 1.

In order to analyze the market influence (fluctuation in labor and equipment costs), three
different scenarios, high inflation rate for labor cost and low inflation rate for capital cost, the
vice versa condition, and high inflation rate for both of them are considered. In doing so, the
total cost is influenced with this inflations, but duration stay remains.

Finally, with combining both internal and external sources of uncertainty, the case is
analyzed. In this case, the total cost is much more than the previous case even the first case
since more than DRTS efficiency, we consider an inflation rate for each of labor and
equipment costs.
The conclusion is that introducing Cobb-Douglas function into time-cost tradeoff problem
provides us extra capacity to further identify the optimal allocations of labor and equipment
resources during crashing. Until now, no one in the literature consider the impact of these
factors in construction crashing cost. Considering these factors, managers can determine their
priority to invest on labor and equipment in different projects for crashing. It means that,
some managers may prefer to invest more in equipment which can be used in future projects,
while sometimes, they just want to take care of the current projects. In doing so, they may
prefer to add more labor to do their work in a shorter time.
 BIBILOGRAPHY

1) Aghaie, A.; Mokhtari, H. 2009. Ant colony optimization algorithm for stochastic
project crashing problem in PERT network using MC simulation. International
Journal of Advanced Manufacturing Technology 45 (11/12): 1051-1067.
2) Banker, R.D.; Natarajan, R. 2008. Evaluating contextual variables affecting
productivity using data envelopment analysis. Operations Research 56 : 48-58.
3) Bazaraa, M.; Shetty, B. 1979. Nonlinear programming theory and algorithms, Fifth
Ed., York: Wiley.
4) Berman, E. 1964. Resource allocation in a PERT network under continuous activity
time–cost function. Management Science 10 : 734–745.
5) Bozejko, W.; Hejducki, Z.; Wodecki, M. 2012. Applying metaheuristic strategies in
construction project management. Journal of Civil Engineering and Management 18 :
621-630
6) Buddhakulsomsiri, J.; Kim, D.S. 2006. Properties of multi-mode resource-
constrained project scheduling problems with resource vacations and activity splitting.
Europea Journal of Operational Research 175 : 279-295.
7) Pendharkar, P.C.; Rodger, J.A.; Subramanian, G.H. 2008. An empirical study of
the Cobb-Douglas production function properties of software development effort.
Information and Software Technology 50 : 1181-1188.
8) Lin, M.C.; Tserng H.P.; Ho, S.P.; Young, D.L. 2011. Developing a construction-
duration model based on a historical dataset for building project. Journal of Civil
Engineering and Management 17 : 529-539.
9) Zheng, D.X.M.; Ng, S.T. 2005. Stochastic time-cost optimization model
incorporating fuzzy sets theory and non-replaceable front. Journal of Construction
Engineering and Management 131 : 176-186.