CE 402 Example Problems

1) You are going to decide about the contingency, amount for four projects; Project-1, Project-2,
Project-3, and Project-4. Project-1 is contracted on unit price basis , Project-2 on cost-plus-fee
basis, and Project-3 and Project-4 on lump sum basis. Project-1, Project-2, Project-4 are all in
Turkey, and Project-3 is in Syria. Write down (rank) the contingency amounts of projects (C1,
C2, C3, C4) from lowest to highest.

C2 < C1 < C4 < C3

2) A contractor has developed: parametric estimation level-1, parametric estimation level-2, and
simulation level-1 models for conceptual cost estimation of a building project. Which model will
you expect to have the worse accuracy, and which model do you expect to have the best accuracy.
Describe briefly why?

Simulation level-1 is expected to have the worst accuracy, and parametric estimation level-2 is
expected to have the best accuracy. In parametric estimation the information of parameters are
included for estimation, whereas in simulation this information is not included. In parametric
estimation level-2 more information at more detailed cost level (system level) is included. The
inclusion of more information in cost estimation techniques is expected to improve the accuracy
of cost estimates.

3) An owner is planning to have a new automobile plant and as a cost engineer you are responsible
for the bidding stage of the project. You are going to prepare the bid documents and will
determine the requirements for the bid. You have decided to award the project to the lowest
bidder. But you know that there is a chance that the lowest bidder may not sign the contract.
What can you do to enforce the lowest bidder to sign the contract?

Get a bid bond.

4) While bidding on a project, a subcontractor determines the quantity of structural steel to be 1,000
tons, and estimates the total direct and indirect cost of the structural steel works as $1,500,000.
After the project is completed, the actual quantities end up being 1,100 tons, and total project
actual cost becomes $1,650,000. What will be the profit (or loss) of the subcontractor if:
a) The contract is lump sum, for which the bid amount of subcontractor is equal to $1,600,000
and there are no changes in the scope of steel works.

Loss = 1,650,000 - 1,600,000 = $ 50,000

b) The contract is unit price for which the bid amount of subcontractor is equal to $1,600,000.

Unit price of steel in the bid = 1,600,000/1,000 = $1,600/ton

Total amount the contractor will receive = 1,600 x 1,100 = $ 1,760,000
Profit = $ 1,760,000 - 1,650,000 = $110,000

c) The contract is cost-plus-fee with a percentage fee of 10% of the subcontractor’s actual cost
Profit = 1,650,000 x 0.1 = $165,000
 6) A pipeline contractor is developing parametric cost models to estimate cost of a new 16” pipeline
project. The contractor is using data of 7 previous 16” pipeline projects given below to develop the
models. The contractor is considering three regression models.
a) Among the three models given which one should be selected for estimating ? Explain why?
b) Using the model selected estimate the total cost of a 120 km., 16” pipeline project with
Year Index=900 and Location Index = 106.

DATA OF PREVIOUS PROJECTS MODEL-1 SUMMARY OUTPUT

Year Index Location Index Cost (TL/m)
100 100 13.2 Regression Statistics
200 105 25.8 R Square 0.9117
300 115 44.7
400 102 69.2 Coefficients t Stat P-value
500 110 114.2 Intercept -33.86 -0.221 0.841
600 106 140.0 Year Index 0.34 5.652 0.005
700 112 231.0 Location Index 0.09 0.037 0.972

MODEL-2 SUMMARY OUTPUT MODEL-3 SUMMARY OUTPUT

Regression Statistics Regression Statistics
R Square 0.9115 R Square 0.9115

Coefficients t Stat P-value Coefficients t Stat P-value

Intercept -24.72 -2.110 0.089 Intercept -29.09 - 1.913 0.090
Year Index 0.34 7.170 0.001 (Year Index)2 0.0005 7.009 0.001

a) For the first project (with Year Index = 100) the third model predicts the cost as -24.09 TL/m which is
not reasonable and acceptable, hence this model is eliminated. In the first model the variable Location
Index has a regression coefficient with a P-value of 0.972 that is significantly larger than 0.1. Hence, this
variable should be dropped from the model since it does not have a significant contribution and Model-1
may make inaccurate predictions. Thus, Model-2 is selected since the variable(s) included in the model has
significant contribution and the model makes reasonable predictions for the existing and future projects.

b) Cost = -24.72 + 0.34 x 900 = 281. 28 TL/m

Total Cost = 120,000 x 281.28 = 33, 753,600 TL.
7) As an estimator you have developed three different methods for conceptual cost estimation of site
development costs. Range estimates of the three methods are given below. Based on the range
estimates given which method should you select for conceptual cost estimation of site
development costs (1 pts)? Explain why (7 pts)?
Lower(10%) Expected Higher(10%)
Method-1 2,210,000 2,600,000 2,990,000
Method-2 1,911,000 2,450,000 2,989,000
Method-3 2,400,000 2,750,000 3,100,000
If the magnitude of difference between expected and higher estimates are compared for three mothods:
Method-1 = (2,990,000 - 2,600,000)/ 2,600,000 = 0.153 or % 15.3
Method-2 = (2,989,000 - 2,450,000)/ 2,450,000 = 0.216 or % 21.6
Method-3 = (2,750,000 – 2,400,000)/2,400,000 = 0.127 or % 12.7

Hence Method-3 has the narrowest range and should be selected as it is expected to make more accurate
estimates.
 8) A regression model was developed to estimate cost of building projects as;
C = 830A + 

Where C is the predicted cost in $; A is the total building area in m2.; and  is a random error with a
mean value of 0 and a standard deviation value of 200,000. Using the model, and assuming that
the error  has a normal distribution; calculate the probability of cost being between $4,350,000
and $4,550,000 (P (4,350,000 ≤ C ≤ 4,550,000)) for a new project with a total building area of
5,000 m2.

Note: For a variable (X) that has a normal distribution with mean (µ) and standard deviation(σ)
probability of X being within one standard deviation from mean µ is approximately 0.68 and
probability of X being within two standard deviations from mean µ is approximately 0.95.

C = 830 x 5,000 +  = 4,150,000 + 

P (3,750,000 ≤ C ≤ 4,550,000) = 0.95 and P (3,950,000 ≤ C ≤ 4,350,000) = 0.68
P (4,350,000 ≤ C ≤ 4,550,000) = (0.95 – 0.68) / 2 = 0.135

9) A contractor is preparing a bid for a unit price goverment project which has a scheduled duration
of 48 months. The escalation formula included in the contract to determine the rate of increase in
unit prices is given below. What is the main disadvantage of this escalation formula (2 pts)?
What should the contractor do during bidding to offset (compensate) this disadvantage
(6 pts)?
 In  Where; In is TEFE for the month that the cost is escalated and
P  P0  0.2  0.8 
 I0  IO is TEFE for the month that the contract was signed.
a) The formula does not give full escalation of the total project amount.
b) The contractor should estimate the amount that the formula does not escalate by predicting the the
current value of the difference between predicted monthly costs and the escalated amounts according
to the formula and add this difference to the bid price.

10) For a crew formed of 3 welder and 2 helper the man-hours required to weld 1 m. of 20-inch
pipeline is 0.4 man-hours. Estimate the duration and labor cost for welding 500 m. of 20-inch
pipeline using the given crew. Use daily working schedule of 8 hours. (Hourly wage for welder is
$3 and hourly wage for helper is $2.)

To complete 500 meter of 20-inch pipeline 500 x 0.4 = 200 man-hours are required.
In one day the crew provides 3 x 8 + 2 x 8 = 40 man-hours.
Hence for this crew to complete 500 meter of 20-inch pipeline 200/40 = 5 days are needed.
Then the labor cost is estimated as 5 x 8 (3 x 3 + 2 x 2) = $ 520.