CHAPTER 12

Solutions to Chapter-End Problems

A. Key Concepts

Sensitivity Graphs:

12.1 (a) - First cost

- Annual operating and maintenance costs
- Annual savings or revenue
- Salvage value
- Service life

(b) - First cost

- Annual operating and maintenance costs
- Annual revenue
- Shipping cost
- Inflation rate
- Exchange rate
- Tariff

12.2 (a) For example, for Canada, as of March 2008: Base figure: 1.4%

Range of variation observed over last 10 years: 0.5% to 4%; this should
be reasonable for a relatively short time period

(b) For example, for Canada, as of May 2008: Base figure: $1US per
$1CAN

Range of variation observed in last two years: $1.05 to $0.95; this should
be reasonable for a relatively short time period

(c) Base figure: use the average annual savings of the equipment you
already have

Range of variation: a variation range of 5-10% should be reasonable since

the new equipment is similar to the old one; you can probably get good
estimate for the range of variation based on the information on the old one

(d) Base figure: use the average annual revenue of a similar internet-
based business (if it exists)

314
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

Range of variation: since the nature of internet-based business is highly

unpredictable at this stage, one should employ a large range of variation
such as 5-50%

(e) Base figure: use the book value computed by declining-balance

method with depreciation rate of 30%

Range of variation: this depends on the age of the computer, but this could
be anywhere between 0 and 50% of the purchase price

12.3 (a) Break-even analysis for multiple projects: to see the effect of the future
demand on the annual worth of leasing the trucks of different sizes

(b) Break-even analysis for a single project: to see the effect of the
uncertain heating expenses on the annual worth of the business

(c) Break-even analysis for a single project: to see the effect of the future
demand on the annual worth of the business

(d) Sensitivity graphs: to see the effect of the uncertain construction cost on the
total cost

(e) Break-even analysis for a single project or scenario analysis: break-

even analysis for examining the effect of the growth rate and the length of
growth separately; scenario analysis is appropriate to examine the growth
rate and the length of growth together

12.4 The number of years, N, to save F $ by putting aside A $ per

year at an annual interest rate i is N = ln[(iF + A)/A]/ln(1 + i).
With 5% and 10% decreases and increases in the savings per year and
the interest rate, Kelowna Go-Karts will take the following number of years
to save $50 000:

Parameter -10% -5% Base Case 5% 10%

Savings per year 6300 6650 7000 7350 7700
Interest rate 9.00% 9.50% 10.00% 10.50% 11.00%

Number of years to save €50 000: -10% -5% 0% 5% 10%

With changes to savings per year 6.13 5.88 5.66 5.44 5.25
With changes to interest rate 5.76 5.71 5.66 5.60 5.56

315
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

6.20
6.00
5.80

Years
5.60 Interest rate
5.40
Savings
5.20
-10% -5% 0% 5% 10%
% change from base case

Break Even Analysis:

12.5 (a) AW = −21 500(A/P, 10%, 10) + 10(1500) − 5(1500)

+ 21 500(1 − 0.2)10(A/F, 10%, 10)
= −21 500(0.16275) + 15 000 − 7500 + 21 500(0.8)10×(0.06275)
= $4145.74

(b) Break-even graph:

8000

6000
Annual worth ($)

4000

2000

0
0 500 1000 1500 2000
-2000

-4000
Operating hours

The break-even level of operating hours is 670.85 hours.

12.6 (a) First, the IRR for the incremental investment from “do nothing” to A is
found by solving for i in:
100 000(P/A, i, 5) = 50 000 fi (P/A, i, 5) = 2 fi IRR = 41.1%

316
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

A is the current best alternative. The IRR on the incremental investment

between A and B is found by solving for i in:

(400 000 – 100 000) + (150 000 – 50 000)(P/A, i, N) = 0

(P/A, i, N) = 300 000/100 000 = 3

This gives an IRR of 19.9%, thus the incremental investment in B is

justified.

(b) Preference for A over “do nothing” remains unchanged. The

incremental investment between A and B is of concern. If the savings per
year due to B are X, then the IRR in the incremental investment is found
by solving for i in:

(P/A, i%, N) = 300 000/(X – 50 000)

The IRR for various values of X is:

X (300 000)/(X-50 000) IRR

120 000 4.286 5.38%
125 000 4.000 8.00%
130 000 3.750 10.42%
135 000 3.529 12.86%
140 000 3.333 15.23%
145 000 3.158 17.57%
150 000 3.000 19.90%

With a MARR of 10%, A is preferred for X (annual savings of B) below

about $129 000 and B is preferred for X above $129 000. The amount
$129 000 is the break-even annual savings for B. A diagram of savings
due to B versus IRR on the incremental investment is:

20%
IRR on increment

15%

10%
120000 130000 140000 150000
5%
Annual savings of B

12.7 From trial and error with a spreadsheet, the break-even interest rate is
11.2% Model T is preferred for a MARR of 16%.
PW(T) = –100 000 + 50 000(P/A, i, 5) + 20 000(P/F, i, 5)

317
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

PW(A) = –150 000 + 62 000(P/A, i, 5) + 30 000(P/F, i, 5)

140000

Present worth ($)

120000
100000
80000
Model T
60000
Model A
40000
6% 8% 10% 12% 14% 16% 18% 20%

Interest rate

12.8 Since the two models have unequal lives, it is easiest to compare them
based on an annual worth computation.
The annual worth of each for a variety of interest rates is:

Depreciation AW AW
Rate (Model A) Model A Model B
0.3 3365 3809
0.4 3669 3809
0.5 3887 3809
0.6 4033 3809

Sample computation for d = 40% depreciation rate for Model A:

AW(model A) = 8000(A/P, 11%, 3) + 1000 – 8000(1 – d)3

= 8000(0.3982) + 1000 – 8000(0.6)3 = 3669

And for Model B with straight line depreciation at $2500 per year:

AW(model B)
= 10 000(A/P, 11%, 4) + 700 – (10 000 – 2500×4)(A/F, 11%, 4)
= 10 000 (0.3108) + 700 = 3809

Since the annual worth of Model A is lower than that of Model B over the
range of depreciation rates Julia has estimated for Model A, she should
pick Model A. From the above table, we can interpolate the break-even
depreciation rate to be 46%. Below this rate, Model A is preferred; above
this rate, Model B is preferred. She is indifferent with a depreciation rate of
46%. A break-even chart is as follows:

318
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

4200

Annual worth ($)

Model A
4000
3800 Model B
3600
3400
3200
0.3 0.4 0.5 0.6
Depreciation rate for A

Probability and Expected Value:

12.9 (a)
E(return on investment)
= Pr{7%}(10 000×0.07) + Pr{10%}(10 000×0.1) + Pr{15%}(10 000×0.15)
= 0.65(700) + 0.25(1000) + 0.1(1500) = 855

The expected return from this investment is $855.

12.10 (a)
E(loss) = Pr{capacity 30}(loss of 20) + Pr{capacity 40}(loss of 10)
+ Pr{capacity 50}(loss of 0) + Pr{capacity 60}(loss of 0)
= 0.2(20) + 0.4(10) + 0.3(0) + 0.1(0)
= 8 calls per hour

The expected loss of customers due to the lack of processing capacity is 8

per hour.

12.11 (a)
E(number of defects, A1)
= 0.3(0) + 0.28(1) + 0.15(2) + 0.15(3) + 0.1(4) + 0.02(5) = 1.53/100 units

E(number of defects, X1000)

= 0.25(0) + 0.33(1) + 0.26(2) + 0.1(3) + 0.05(4) + 0.01(5) = 1.4/100 units

According to the expected number of defects, X1000 seems to be slightly

better than A1.

12.12 (a)
E(cost in a summer month)
= 0.4(800) + 0.25(2×800) + 0.2(3×800) + 0.1(3×800 + 1500)
+ 0.05(3×800 + 2×1500)
= $1860 per month
319
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

E(cost in a non-summer month)

= 0.45(0) + 0.4(800) + 0.15(2×800) = $560 per month

CB Electronix should consider getting the complete coverage policy

because the expected cost is over $500 even in the non-summer months.

Decision Trees:

12.13 (a) p(customer getting a bad board) = 2/5 = 0.4

(b) E(send stock) = p(good board)(0) + p(bad board)(–10 000)

= 0.6(0) + 0.4(–10 000) = –4000

The expected cost of sending the customer one of the five boards is
$4000.

(c) Tree diagram:

Incremental
Cost ($)
E(2) Good Board
= 4000 0
0.6
Send Stock
E(1) 2
= 4000
Bad Board
10 000
1 0.4

Send from Other Plant

5000

E(1) = E(2) = 4000 because E(2) < 5000

Randall should send the customer one from stock.

320
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

12.14 Decision Tree diagram:

Payoff ($)
E(2) Demand High
= 447 500 750 000
Increase 0.4
Production Demand Same
2 350 000
0.35
E(1)
= 447 500 Demand Low
100 000
0.25
1
Demand High
Do Not 500 000
Increase 0.4
Production Demand Same
3 400 000
0.35
E(3) Demand Low
200 000
= 390 000 0.25

E(2) = Pr{high demand}(750 000) + Pr{medium demand}(350 000)

+ Pr{low demand}(100 000)
= 0.4(750 000) + 0.35(350 000) + 0.25(100 000)
= 447 500
E(3) = 0.4(500 000) + 0.35(400 000) + 0.25(200 000) = 390 000
E(1) = E(2) = 447 500 because E(2) > E(3)

SJCF should increase their production.

(b) The cumulative risk profile for the two decision alternatives are in the
figure below.

Outcome dominance does not exist because one decision is not better
than the other for all outcomes (e.g. for the outcome demand low, “do not
increase production” is better, but for the outcome demand high, “increase
production” is better.)

First degree stochastic dominance does not exist. This can be seen from
the cumulative risk profiles below, where Pr(demand ≤ X) is greater for the
“increase production” decision for all values of X up to $500, but not for all
values of X.

321
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

1.0

0.9

0.8

0.7

0.6
P(x)

0.5
Increase

0.4 Do Not Increase

0.3

0.2

0.1

0.0
0 50 100 150 200 250 300 350 400 450 500 550 600 650 700 750 800
X , Payoff ($000's)

12.15 Tree diagram:

Expected Gain in
Market Share (%)

E(2) Rapid Growth

30
= 15.9 0.3
Combined
Rate Steady Growth
2 15
0.4
E(1)
= 15.9 Slow Growth
3
0.3
1
Rapid Growth
15
Add-On 0.3
Rates Steady Growth
3 10
0.4
E(3) Slow Growth
5
= 10 0.3

E(2) = Pr{rapid growth}(30) + Pr{steady growth}(15) + Pr{slow growth}(3)

= 0.3(30) + 0.4(15) + 0.3(3) = 15.9

E(3) = 0.3(15) + 0.4(10) + 0.3(5) = 10

E(1) = E(2) = 15.9 because E(2) > E(3)

LOTell should introduce the combined rate.

322
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

(b) Several types of dominance reasoning can be applied: mean-variance,

outcome or stochastic dominance. Each is covered below:

Mean-variance dominance:

Var(combined rate) = Pr{rapid growth}(30 - 6.9)2

+ Pr{steady growth}(15 -6.9)2 + Pr{slow growth}(3 - 6.9)2
= 0.3(533.61) + 0.4(65.61) + 0.3(15.21) = 31 $2

Var(add-on rates) = Pr{rapid growth}(15 – 5.5)2

+ Pr{steady growth}(10 -5.5)2 + Pr{slow growth}(5 – 5.5)2
= 8 $2

One of the two decisions cannot be eliminated with mean-variance

reasoning because the combined rate decision has the higher expected
gain in market share, but the add-on rates decision has the lower (better)
variance.

Outcome dominance:

Observe that the “add-on rate” decision has a better outcome for slow
growth, but the “combined rate” decision has a better outcome for both
steady and rapid growth. Since one of the decisions is not better than the
other for all possible outcomes, outcome dominance cannot be used to
eliminate either decision.

Stochastic Dominance:

The Cumulative risk profiles (CDFs) for the two decisions are shown in the
figure below. The “combined rate” decision is dominated for market share
outcomes up to 5% (i.e. has a higher probability that demand is less than
or equal to X for X ≤ 5%), but the “add-on rate” decision is dominated for
outcomes over 5% market share. Thus, neither decision dominates the
other in a first degree stochastic dominance sense.

323
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

1.0

0.9

0.8

0.7

0.6
P(x)

0.5
Add-on rates Combined rates

0.4

0.3

0.2

0.1

0.0
0 5 10 15 20 25 30 35 40
X , Gain in market share (%)

B. Applications

12.16 The present worth of the software when the base annual savings are
= 10 000, the gradient is G = 1000, and interest rate is i = 15%:
PW = [10 000 + 1000(A/G, 15%, 6)](P/A, 15%, 6)(P/F, 15%, 2) = 34 617

The other computations are as follows:

Parameter -15% -7.5% Base Case 7.5% 15%

Base savings 8500 9250 10000 10750 11500
Savings gradient 850 925 1000 1075 1150
Interest rate 12.75% 13.88% 15.00% 16.13% 17.25%

Present Worth: -15% -7.5% 0% 7.5% 15%

Changes to base savings per year 30325 32471 34617 36764 38910
Changes to savings gradient 33717 34167 34617 35068 35518
Changes to interest rate 38484 36485 34617 32872 31238

324
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

The sensitivity graph is:

40000

Present worth ($)

Base savings
37500
35000 Savings gradient

32500
Interest rate
30000
-15.0 -7.5% 0.0% 7.5% 15.0%
%
% change from base case

12.17 First, note that all cost figures are given in real dollars, as they do not take
into account the effect of inflation.
The municipality uses an actual interest rate of 7% when inflation is
expected to be 3%, and hence, their real MARR is:

MARRR = (1 + MARRA)/(1 + f) – 1 = 1.07/1.03 – 1 = 0.0388 or 3.88%

Since all costs are based on current estimates, it is simplest to work with
real dollars and the real MARR for the analysis. Note that even if the City’s
estimates of inflation change by 5% or 10%, the real MARR they use will
not change and hence the present worth of the project (in real dollars) will
be unaffected by the inflation rate.

(a) The present worth of the project is $360 204.

Parameter -10% -5% Base Case 5% 10%

Annual Construction Costs (in $000s) 18000 19000 20000 21000 22000
Annual Maint. and Rep. costs (in $000s) 1800 1900 2000 2100 2200
Inflation Rate 3.30% 3.15% 3.00% 3.15% 3.30%

Present Worth (in $000s) -10% -5% 0% 5% 10%

Changes to annual construction costs 95949 99784 103619 107454 111289
Changes to annual maint. costs 100927 102273 103619 104965 106310
Changes to inflation rate 103619 103619 103619 103619 103619

325
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

(b) The sensitivity graph is as follows:

115000

Present worth (in $000s)

110000

105000

100000

95000

90000
-10% -5% 0% 5% 10%
% change to base case

The present worth of the project is most sensitive to changes in the annual
maintenance and repair costs.

12.18 The present worth of the pool (in $000s) with the “base case” costs and
MARR is:
PW = 6000 + [400(A/F, 5%, 10)]/0.05

With one at a time 5% and 10% variations in the first cost, maintenance
costs and the MARR, the present worth of costs are:

Parameter -10% -5% Base Case 5% 10%

Construction costs (in $000s) 5400 5700 6000 6300 6600
Refinishing costs (in $000s) 360 380 400 420 440
MARR 4.50% 4.75% 5.00% 5.25% 5.50%

Present worth of costs (in $000s) -10% -5% 0% 5% 10%

Changes to construction costs 6036 6336 6636 6936 7236
Changes to refinishing costs 6572 6604 6636 6668 6700
Changes to MARR 6723 6677 6636 6599 6565

The sensitivity graph shows that the present worth is most sensitive to the
first cost of the pool:

7500
Construction costs
Present worth (in

7000
Refinishing costs
!000s)

6500
MARR

6000
-10% -5% 0% 5% 10%
% change from base case

326
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

12.19 Sample Equivalent Monthly Worth (EMW) Computation (for base case):
EMW = (Monthly Revenues) – (Monthly Costs)
= 300×2 – 100 – [(6000 – 3000)(A/P, 8%/12, 24) + 3000(0.08/12)]
– 600(A/P, 8%/12, 6)
= 242

Parameter -20% -10% Base Case 10% 20%

Monthly Advertising Costs 80 90 100 110 120
Number of Customers per month 240 270 300 330 360
Interest rate (% per month) 0.533% 0.600% 0.667% 0.733% 0.800%

Sensitivity Graph Information -10% -5% 0.0% 5% 10%

Changes to Advertising Costs 262 252 242 232 222
Changes to Number of Deliveries 122 182 242 302 362
Changes in Interest Rate 245 243 242 241 239
Equivalent monthly worth

400
Number of deliveries

300
Interest rate
($)

200 Advertising costs

100
-10% -5% 0% 5% 10%
% change from base case

The monthly worth is most sensitive to changes in the number of deliveries

per month.

12.20 (a) A summary of the costs for Vendor A’s device is below at the base
value and for 5% and 10% increases and decreases.
Also given is the annual worth of costs for Vendor A’s device. A sample
computation for the base case is:
AW = 200 000(A/P, 15%, 7) + 10 000 + 6500
+ (0.05 + 0.95 + 1.25)(50 000) – 5000(A/F, 15%, 7)
= 176 620

Parameter -10% -5% Base Case 5% 10%

First Cost 180 000 190 000 200 000 210 000 220 000
Annual Maintenance Cost 9 000 9 500 10 000 10 500 11 000
Maintenance cost/unit 0.045 0.048 0.050 0.053 0.055
Labour cost/unit 1.13 1.19 1.25 1.31 1.38
Annual other costs 5 850 6 175 6 500 6 825 7 150
Other Costs/unit 0.86 0.90 0.95 1.00 1.05
Salvage Value 4 500 4 750 5 000 5 250 5 500

Annual costs -10% -5% 0% 5% 10%

327
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

Changes in First Cost 171 813 174 217 176 620 179 024 181 427
Changes in Fixed Costs 175 620 176 120 176 620 177 120 177 620
Changes in Variable costs 176 370 176 495 176 620 176 745 176 870
Changes in Salvage Value 176 665 176 643 176 620 176 598 176 575

A sensitivity graph indicating the sensitivity of the annual worth of Vendor

A’s device to changes in the costs shows that the annual worth is most
sensitive to changes in the first cost and then to changes to the fixed costs
(maintenance and other).

182 000
First cost
180 000
Annual cost ($)

Fixed costs
178 000
Variable costs
176 000 Salvage value

174 000

172 000

170 000
-10% -5% 0% 5% 10%
% change from base case

(b) Vendor B’s computations and sensitivity graph are below. As with
Vendor A, the annual worth is most sensitive to the first cost and to the
fixed annual costs. Also observe that at expected annual production
levels, Vendor B’s device has a lower annual worth.

Parameter -10% -5% Base Case 5% 10%

First Cost 315 000 332 500 350 000 367 500 385 000
Annual Maintenance Cost 18 000 19 000 20 000 21 000 22 000
Maintenance cost/unit 0.009 0.010 0.010 0.011 0.011
Labour cost/unit 0.45 0.48 0.50 0.53 0.55
Annual other costs 13 950 14 725 15 500 16 275 17 050
Other Costs/unit 0.50 0.52 0.55 0.58 0.61
Salvage Value 18 000 19 000 20 000 21 000 22 000

Annual cost -10% -5% 0% 5% 10%

Changes in First Cost 150 279 153 766 157 253 160 740 164 227
Changes in Fixed Costs 155 253 156 253 157 253 158 253 159 253
Changes in Variable costs 157 203 157 228 157 253 157 278 157 303
Changes in Salvage Value 157 352 157 302 157 253 157 204 157 155

328
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

165 000 First cost

Annual cost ($)

160 000 Fixed costs

Variable costs
Salvage value
155 000

150 000
-10% -5% 0% 5% 10%
% change from base case

12.21 (a) Let D = the number of deliveries per month. Solve for D in:
0 = EMW = (Monthly Revenues) – (Monthly Costs)
= 2D – 100 – [(6000 – 3000)(A/P, 8%/12, 24) + 3000(0.08/12)]
– 600(A/P, 8%/12, 6)
D = 179

The break-even number of deliveries is 179.

300
200
Monthly worth ($)

100
0
- 100 50 100 150 200 250 300
- 200
- 300
Number of deliveries per month

(b) Solving for the interest rate is most easily done with a trial and error
approach on a spreadsheet. The break-even interest rate is 10.29% per
month, or 123.5% per year.

329
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

400

Monthly worth ($)

200
0
- 200 0% 5% 10% 15% 20% 25%

- 400

- 600
Interest rate per month

12.22 (a) A summary of the costs and benefits and the annual worth
computations for Alternative A is as follows:
Costs and Benefits: -10% -5% Base Case 5% 10%
Initital hardware and installation costs 83 250 87 875 92 500 97 125 101 750
Annual Benefits 58 500 61 750 65 000 68 250 71 500

Annual Worths: -10% -5% 0% 5% 10%

Initital hardware and installation costs 11 516 10 595 9 673 8 751 7 830
Annual Benefits 3 173 6 423 9 673 12 923 16 173

A sample computation for the annual worth of the base case is:

AW(base case)
= 65 000 – [(138 750 + 92 500)(A/P, 15%, 10) + 9250] = 9673

and for the annual worth if the initial hardware and installation costs are
10% higher than the base case:

AW(base case)
= 65 000 – [(138 750 + 101 750)(A/P, 15%, 10) + 9250] = 7830

A sensitivity graph of these results is:

20 000
Annual worth ($)

Annual benefits
15 000

10 000
Initial hardw are
5 000 & installation costs

0
-10% -5% 0% 5% 10%
% change from base case

330
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

The annual worth is most sensitive to changes in the annual benefits of

the network.

(b) A summary of the costs and benefits and the annual worth
computations for Alternative B is as follows:

Costs and Benefits ($): -10% -5% 5% 10%

Base Case
Initital hardware and installation costs 94 950 100 225 105 500 110 775 116 050
Annual Benefits 66 600 70 300 74 000 77 700 81 400

Annual Worths ($) -10% -5% 0% 5% 10%

Initital hardware and installation costs 12 999 11 948 10 897 9 846 8 795
Annual Benefits 3 497 7 197 10 897 14 597 18 297

And the corresponding sensitivity graph is:

20 000 Annual benefits

Annual worth ($)

15 000

10 000 Hardw are &

installation costs
5 000

0
-10% -5% 0% 5% 10%
% change from base case

12.23 A graph showing the number of years required to save $50 000 as a
function of the amount saved per year is below.
With annual savings of A and an interest rate i = 0.15, the number of
years, N, to save F = $50 000 is N = ln[(iF + A)/A]/ln(1 + i).

8
Number of years to
save !50 000

5
5000 6000 7000 8000 9000
4
Amount saved per year (!)

The break-even annual savings amount is $8190 (this was obtained with
trial and error with the spreadsheet table using the amount saved per year
331
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

as the input variable). The Go-Kart Klub will need to raise $1190 each
year (in addition to the $7000) if they wish to save $50 000 in 5 years.

12.24 First the PW of acquiring the molder and keeping it over its 6-year life is
the sum of its capital costs, and operating and maintenance costs:
PW(capital) = 27 000 – BVdb(6)(P/F, 15%, 6)
= 27 000 – 20 000(1 – 0.4)6(0.4323) = 26 596.61

The present worth of the operating and maintenance costs over the 6-year
life can be obtained with the use of a geometric gradient series to present
worth conversion factor:

PW(op. and maint.) = 30 000(P/A, iϒ, 6)/(1 + g)

where iϒ = (1 + i)/(1 + g) – 1 = 1.15/1.05 – 1 = 0.09524

PW(op. and maint.) = 30 000(4.4167)/1.05 = 126 191.43

This gives a present worth of cost for the 6-year planned life of the
molding equipment of $150 788 = 126 191.43 + 24 596.61.

The equivalent annual cost can be found from:

EAC(molder) = 152 788(A/P, 15%, 6) = 152 788(0.2642) = 40 366.59

The cost per piece is then EAC(molder)/production volume. Using this, we

can construct the break-even analysis. The break-even quantity is
212 500. Below this annual production quantity, Bountiful should continue
to purchase from outside. Above this quantity, they should buy the molding
equipment.
Cost per unit (£)

0.23
0.21
0.19 Purchase cost
0.17 In-house cost
0.15
180 200 220 240
Production volume
(1000s of pieces/yr)

12.25 Under Canadian tax rules, assuming a CCA rate of 30%, first, we need to
calculate the Tax Benefit Factor:
TBF = 1 – td/(i + d) = 1 – (0.35)(0.3)/(0.12 + 0.30) = 0.75

332
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

AW(first cost) = [–65 000

+ (65 000/2xTBF + 65 000/2(P/F, 12%, 1))(A/P, 12%, 5)
= –13 758
AW(savings) = (Annual Savings)(1 – 0.35)
AW(salvage) = 20 000(1 - TBF)(A/F, 12%, 5)
= 20 000(0.75)(0.15741) = 2361
AW(truck) = –13 758 + 2361 + (Annual Savings)(0.65)

With a spreadsheet, the annual worth of the truck for annual revenues
from $12 000 to $22 000 can be calculated. The break-even revenue per
year is $17 535.

4000
Annual worth ($)

2000

0
12 14 16 18 20 22
-2000

-4000
Revenues (in $000s/yr)

12.26 (a) EAC calculation summary:

6 7 8 9 10
Service
Life
Year EAC(total) EAC(total) EAC(total) EAC(total) EAC(total)
1 20 067 19 114 18 400 17 844 17 400
2 17 519 16 621 15 947 15 423 15 004
3 16 733 15 887 15 252 14 758 14 363
4 16 385 15 588 14 990 14 525 14 153
5 16 209 15 459 14 897 14 460 14 110
6 16 117 15 413 14 885 14 474 14 146
7 15 412 14 916 14 531 14 222
8 14 972 14 610 14 321
9 14 703 14 432
10 14 549

333
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

(b) Sensitivity graph:

Economic life (years)

6

0
6 7 8 9 10
Service life (years)

The sensitivity graph indicates that the economic life is somewhat

sensitive to the length of service life. For a service life of 6 or 7 years, it is
economical to keep the bottle capping machine until the end of its service
life. For a service life longer than 7 years, it is more economical to keep
the machine for a shorter period than the service life.

12.27 The annual worth of the Clip Job for annual maintenance costs varying
from $30 to $80 is as follows:
Annual Annual
Maintenance Worth
30 103.84
40 113.84
50 123.84
60 133.84
70 143.84
80 153.84

For example, with annual maintenance costs of $50, the annual worth of
the Clip Job is:
AW(clip job) = 120(A/P, 5%, 4) + 40 + 50 = 120(0.282) + 90 = 123.84

The annual worth of the Lawn Guy is:

AW(lawn guy) = 350(A/P, 5%, 10) + 60 + 30 = 350(0.1295) + 90 = 135.33

334
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

And the maintenance cost for the Clip Job which makes Sam indifferent
between the two lawnmowers is $61.50. A break-even graph illustrates the
break-even maintenance costs:

160

Annual worth ($)

Clip Job

140
Lawn Guy
120

100
30 40 50 60 70 80
Clip Job
annual maintenance costs

Since Sam estimates the maintenance costs to be about $60 per year,
and the break-even maintenance cost is close to this amount, we would
recommend that Sam purchase the Lawn Guy if he is risk averse and
wants to avoid the possibility of $80 maintenance costs for the Clip Job.

12.28 The annual worth of the car for salvage values varying from $6000 to
$24 000 are:
Salvage Value Annual Worth
6000 10647
9000 10165
12000 9684
15000 9202
18000 8720
21000 8239
24000 7757

For example, the annual worth with a salvage value for $9000 is:

AW(car) = 24 000(A/P, 11%, 5) + (2000 + 600 + 800 + 1000)

+ 400(A/G, 11%, 5) – 9000(A/F, 11%, 5)
= 24 000(0.27057) + 4400 + 400(1.7923) – 9000(0.16057)
= 10 165

The annual costs of taking taxis are $8000 (=$7000 + $1000). Based on
an annual worth comparison, Ganesh should not buy the car. The salvage
value of the car would have to exceed the break-even value of $22 486
before the car will have equal or lower costs than that of taking taxis.
Taking into account only the financial aspects of the decision, Ganesh
should not buy the car unless he feels there are other benefits (e.g.
convenience of having a car) that have not been taken into account.
335
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

11000
Car
10000

Annual worth ($)

9000 Taxi

8000

7000

6000
6000 9000 12000 15000 18000 21000 24000
Salvage value of car

12.29 E(monthly savings)

= 0.25(800 000) + 0.25(1 000 000) + 0.25(1 200 000) + 0.25(1 400 000) =
1 100 000

PW = 1 100 000(P/A, 1%, 24) = 1 100 000(21.243) = 23 367 000

The present worth of the expected monthly savings is about $23 367 000.

12.30 E(revenue)
= 0.1(2.95) + 0.35(3.25) + 0.3(3.50) + 0.15(4.00) + 0.1(5.00)
= $3.58 per parcel

Regional’s monthly capacity is 60 000 parcels. So the present worth of the

expected revenue over 12 months is:

PW = (3.58×60 000 − 8000)(P/A, 1%, 12)

= (206 800)(11.255) = $2 327 534

12.31 E(life) = p(4 years)(48 months) + p(5 years)(60 months)

+ p(6 years)(72 months)
= 0.4(48) + 0.4(60) + 0.2(72) = 57.6 months

Using the expected life of 57.6 months, the present worth of the monthly
expenses is computed as follows:

PW = (90 + 30 + 20)(P/A, 10/12%, 57.6)

= 140[(1 + 0.008333)57.6 – 1]/[0.008333(1 + 0.008333)57.6]
= (140)(45.599) = 6383.86

The present worth of the monthly expenses is about €6384 before a major
repair.

336
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

12.32 First, find the annual cost for each scenario:

AC(optimistic) = 75 000
AC(expected) = 240 000(A/F, 15%, 2) = 240 000(0.46512) = 111 628.80
AC(pessimistic) = 500 000(A/F, 15%, 3) = 500 000(0.28798) = 143 990

E(annual cost)
= p(optimistic)(75 000) + p(expected)(111 628.80)
+ p(pessimistic)(143 990)
= 0.15(75 000) + 0.5(111 628.80) + 0.35(143990)
= 117 460.90

The expected annual cost of the vitamin C project is $117 461.

12.33 Tree diagram:

Payoff ($)
E(2) Win House
= 150 249 900
0.001
Buy Ticket
E(1) 2
= 150
Win Nothing
−100
1 0.999

Do Not Buy Ticket

0

E(2) = p(not win)(−100) + p(win)(250 000 − 100)

= 0.999(−100) + 0.001(249 900)
= 150

E(1) = E(2) = 150 because E(2) > 0

Buying a ticket has a greater expected value than not buying a ticket.

12.34 (a) If E(2) = 0.999X + 0.001(250 000 − X) = 0.998X + 250 < 0, then not
buying is preferred.

0.998X + 250 < 0

0.998X < −250
X < −250.50

Hence, if the ticket costs more than $250.50, not buying is the preferred
option.

337
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

(b) If E(2) = (−100)(1 − X) + 249 900X < 0, then not buying is preferred.

(−100)(1 − X) + 249 900X < 0

−100 + 100X + 249 900X = −100 + 250 000X < 0
250 000X < 100
X < 0.00039984

Hence, if the probability of winning is less than 0.04%, not buying is the
preferred option.

12.35 (a) PW(high performance) = −550 000 + 500 000(P/A, 12%, 5)

= −550 000 + 500 000(3.6048)
= 1 252 400
PW(medium performance) = −550 000 + 250 000(P/A, 12%, 5) = 351 200
PW(low performance) = −550 000 + 150 000(P/A, 12%, 5) = −9280

(b) Tree diagram:

Annual Present
Savings ($) Worth ($)

E(2) Performance Low

150 000 −9280
= 252 532 0.35
New
Design Performance Med.
2 250 000 351 200
0.55
E(1)
= 252 532 Performance High
500 000 1 252 400
0.05
1

No New Design
0 0

E(2) = p(high performance)(1 252 400)

+ p(medium performance)(351 200) + p(low performance)(−9280)
= 0.05(1 252 400) + 0.55(351 200) + 0.35(−9280)
= 252 532

E(1) = E(2) = 252 532 because E(2) > 0

BB should approve the development of the new robot.

338
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

12.36 Tree diagram:

Gain in
Annual Profit
E(3) (in $000s)
= 156.25 Acceptable Quality
Shipping 200
No Delay 0.75
E(2) 3
= 113.75 0.6
Poor Quality
Partnership 25
with China 0.25
2
Acceptable Quality
100
Delay 0.75
E(1) 4
0.4
= 113.75 Poor Quality −100
E(4) 0.25
1 = 50

No Partnership
0

E(3) = p(acceptable quality)(200 000) + p(poor quality)(25 000)

= 0.75(200 000) + 0.25(25 000)
= 156 250

E(4) = p(acceptable quality)(100 000) + p(poor quality)(−100 000)

= 0.75(100 000) + 0.25(−100 000)
= 50 000

E(2) = p(no shipping delay)E(3) + p(shipping delay)E(4)

= 0.6(156 250) + 0.4(50 000)
= 113 750

E(1) = E(2) = 113 750 because E(2) > 0

The partnership with China is still recommended as a result of analyzing

shipping delay and quality control possibility.

339
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

12.37 Tree diagram:

Cost ($)

E(5) Severe Failure

205 000
= 93 250 0.15
Minor
Modification Minor Failure 90 000
5
E(4) 0.45
= 93 250 No Failure 55 000
Not Favourable 0.4
4
0.66
E(2) E(6) Severe Failure 230 000
= 78 545 = 98 000 0.05

Test Minor Failure 115 000

2 6
Major 0.3
Modification
E(1) No Failure 80 000
= 78 545 0.65
Favourable Result
1 50 000
0.34

E(3) Severe Failure 150 000

= 94 750 0.55
No Test Minor Failure 35 000
3
0.35
No Failure
0
0.1

E(5)
= p(severe failure)(test, modification, and failure costs)
+ p(minor failure)(test, modification, and failure costs)
+ p(no failure)(test and modification costs)
= 0.15(50 000 + 5000 + 150 000) + 0.45(50 000 + 5000 + 35 000)
+0.4(50 000 + 5000)
= 93 250

E(6)
= 0.05(50 000 + 30 000 + 150 000) + 0.3(50 000 + 30 000 + 35 000)
+0.65(50 000 + 30 000)
= 98 000

E(4) = E(5) = 93 250 because E(5) < E(6)

E(2) = p(not favourable)E(4) + p(favourable)(test costs)

= 0.66(93 250) + 0.34(50 000)
= 78 545

E(3) = p(severe failure)(failure costs) + p(minor failure)(failure costs)

+ p(no failure)(no costs)
= 0.55(150 000) + 0.35(35 000) + 0.1(0)

340
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

= 94 750

E(1) = E(2) = 78 545 because E(2) < E(3)

According to decision tree analysis, Rockies should test the upgraded

system, and if the result is not favourable, Rockies should apply the minor
modification to the system.

C. More Challenging Problems

12.38 (a) The break-even installation cost for Alternative A is $141 045. A break-
even chart is shown below:

15000
Annual worth ($)

10000

5000
0
70000 90000 110000 130000 150000 170000
-5000

-10000
Installation cost ($)

The break-even installation cost is outside the range Merry Metalworks

has estimated.

(b) The break-even annual cost for Alternative A is $55 327. A break-even
chart is shown below:

30000
Annual worth ($)

20000

10000

0
50000 60000 70000 80000
-10000
Annual benefits ($)

341
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

12.39 (a) The break-even installation cost for Alternative B is $160 190, which is
well above the range of estimated costs.
A break-even chart showing various annual worths is below:

20000

Annual worth ($)

15000
10000
5000
0
80000 100000 120000 140000 160000 180000
-5000
Installation cost ($)

(b) The break-even annual benefits for Alternative B are $63 103. This is
within the range of benefits Merry Metalworks has specified, so there is
some risk that Alternative B will yield a negative present worth. The break-
even chart showing various annual benefits is below:

30000
Annual worth ($)

15000

0
40000 50000 60000 70000 80000 90000
-15000

-30000
Annual benefits ($)

12.40 (a) The break-even production volume with a per-unit revenue of $3.25 for
Vendor A is 64 120 (obtained by interpolation).
This can be seen in the following break-even graph:

342
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

Annual cost or revenue ($)

Annual sales revenue
240000
Annual cost
200000
160000
120000
80000
30000 40000 50000 60000 70000 80000
Production volume (boards/yr)

(b) For Vendor B, the break-even volume is 47 375 boards/year.

Annual cost or revenue ($)

280000
Annual sales revenue
240000
200000 Annual cost
160000
120000
80000
30000 40000 50000 60000 70000 80000
Production volume (boards/yr)

12.41 Since the lease is an annual amount, it is easiest to compare the two
alternatives based on annual costs. This solution assumes Canadian tax
rules and a 20% CCA rate.
First, the annual worth of the lease decision is $5500 per year, after taxes
this is $5500(1 – t) where t is the tax rate. The annual worth of costs for
the buy decision depends on Kelly’s tax rate.

Tax rate AW(buy) TBF AW(lease)

0.20 4180 0.143 4400
0.25 4060 0.152 4125
0.30 3949 0.158 3850
0.35 3843 0.163 3575
0.40 3742 0.167 3300

For example, at a tax rate of 30% the annual worth of costs for the two
alternatives, taking taxes into account is:

AW(buy)
= [15 000 - (15 000/2xTBF + 15 000/2xTBF(P/F, 8%, 1))](A/P, 8%, 5) +
[1000 + 400(A/G, 8%, 5)](1 – t) – 2500(1-TBF)
343
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

= [15000 - (750x0.158 + 750x0.158x0.54027](0.25046) ) + [1000 +

400(1.846 47)](1 – 0.3) – 2500(0.842) = 3949

AW(lease) = 5500(1 – t) = 3850

We can see from the table that Kelly should lease if his tax rate is below
27% and buy if he expects his tax rate to be above 27%.

4500
Annual worth ($)

4000
Buy
3500
Lease

3000
0.20 0.25 0.30 0.35 0.40
Tax rate

12.42 Break-even graph:

AW(Small) = –(6000 + 1500)(A/P, 12%, 5)
+ (annual savings per employee)×(#employees)
AW(Large) = –(10 000 + 3500)(A/P, 12%, 5)
+ (annual savings per employee)×(#employees)

30000

25000 Large-scale
Annual worth ($)

20000

15000

10000 Small-scale
5000

0
40 60 80 100
Number of Employees

The small-scale version would be a better choice if Western Insurance

wasn’t expecting much growth in the next 5 years. However, they are
growing. If the growth continues steadily for the next 5 years, the large-
scale version seems to be a better choice since it has a greater annual
worth than the small-scale version does as long as the average number of
employees is greater than 60.
344
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

12.43 A summary of the scenarios and the annual worths of the two alternatives
is below:
Pessimistic Expected Optimistic
Alternative A: Outcome Outcome Outcome
Initital hardware and inst. Costs 115000 92500 70000
Initial software cost 138750 138750 138750
Annual maintenance costs 9250 9250 9250
Annual Benefits 50000 65000 80000
Annual Worth -9810 9673 29156
Alternative B:
Initital hardware and inst. Costs 125000 105500 86000
Initial software cost 158250 158250 158250
Annual maintenance costs 10550 10550 10550
Annual Benefits 54000 74000 94000
Annual Worth -12988 10897 34783

From the scenario analysis, we can see that alternative B has the higher
annual worth using the expected outcome figures. Both alternative A and
alternative B are quite sensitive to changes in the annual benefits of the
LAN alternative and give negative annual worths if the pessimistic
outcome is to occur. Alternative B has a greater spread in its annual
benefits and accompanying that, a lower annual worth than alternative A if
the pessimistic outcome should occur.

If Merry Metalworks is not averse to the risk of having a pessimistic

outcome, they should choose alternative B as it has the higher annual
worth for both the expected and optimistic outcomes. If they are risk
averse, they might choose alternative A, and thus avoid the possibility of a
lower annual worth if the pessimistic outcome occurs. Note that this will
also rule out the possibility of having B's higher annual worth should either
the expected or optimistic outcomes occur.

12.44 (a) From Problem 12.32:

AC(optimistic) = 75 000
AC(expected) = 111 628.80
AC(pessimistic) = 143 990

AW(optimistic, public accept)

= 1 000 000(A/F, 15%, 1) − AC(optimistic)
= 1 000 000(1) − 75 000
= 925 000

AW(optimistic, not accept)

= 200 000(A/F, 15%, 1) − AC(optimistic)
= 125 000

345
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

AW(expected, public accept)

= 1 000 000(A/F, 15%, 2) − AC(expected)
= 1 000 000(0.46512) − 111 628.80
= 353 491.20

AW(expected, not accept)

= 200 000(A/F, 15%, 2) − AC(expected)
= −18 604.80

AW(pessimistic, public accept)

= 1 000 000(A/F, 15%, 3) − AC(pessimistic)
= 1 000 000(0.28798) − 143 990
= 143 990

AW(pessimistic, not accept)

= 200 000(A/F, 15%, 3) − AC(pessimistic)
= −86 394

(b) Tree diagram:

Annual
E(3) Worth ($)
= 391 400
Public Accept
Research 925 000
Optimistic 0.333
3
0.15
Public Not Accept
125 000
0.667
E(2)
= 107 975 Public Accept
Develop 353 491
New Product Expected 0.333
2 4
0.5
Public Not Accept
E(4) −18 605
0.667
= 105 303
E(1)
= 107 975 Public Accept
143 990
0.333
1 5
Pessimistic
Public Not Accept −86 394
E(5) 0.667
= −9676

Do Not Develop New Product

0

E(3) = p(public accept)(925 000) + p(not accept)(125 000)

= 0.333(925 000) + 0.667(125 000)
= 391 400

E(4) = 0.333(353 491.20) + 0.667(−18 604.80) = 105 303.17

346
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

E(5) = 0.333(143 990) + 0.667(−86 394) = −9676.13

E(2) = p(optimistic)E(3) + p(expected)E(4) + p(pessimistic)E(5)

= 0.15(391 400) + 0.5(105 303.17 + 0.35(−9676.13)
= 107 974.94

E(1) = E(2) = 107 974.94 because E(2) > 0

Pharma-Excel should proceed with the development of the new product.

12.45 Tree diagram:

Net
E(4) Value ($)
= 65 000 No Future Need
No Extra −40 000
Expenses 0.3
E(2) 4
0.75
= 40 000 Future Need
110 000
Repair 0.7
2 E(5)
= −35 000 No Future Need
Extra −140 000
Expenses 0.3
5
0.25 Future Need 10 000
E(1) 0.7
1 = 40 000
E(8)
Adopt
E(6) No Future = −18 000 0
= 36 600 Need 0.8
No 8
Modification 0.3
6 Not Adopt
E(3) 0.6 −90 000
= 30 520 Future Need 0.2
60 000
0.7 E(9)
Replace
3 Adopt
E(7) No Future = −22 000 0
= 21 400 Need 0.8
9
Modification 0.3
7 Not Adopt −110 000
0.4 0.2
Future Need
40 000
0.7

Do Nothing
0

Based on the tree diagram, the net value for each terminal position (from
top to bottom of the tree) is calculated as follows:

Terminal position 1 = cost of repair = −40 000

Terminal position 2 = PW(benefit) − (cost of repair) = 110 000
Terminal position 3 = (cost of repair) + (extra expenses) = −140 000
Terminal position 4 = PW(benefit) − (costs of repair & extra expenses)

347
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

= 10 000
Terminal position 5 = complete recovery of investment = 0
Terminal position 6 = cost of replacement = −90 000
Terminal position 7 = PW(benefit) − (cost of replacement) = 60 000
Terminal position 8 = complete recovery of investment = 0
Terminal position 9 = (cost of replacement) + (cost of modification)
= −110 000
Terminal position 10 = PW(benefit) − (replacement&modification costs)
= 40 000
Terminal position 11 = do nothing = 0

E(8) = p(adopt)(0) + p(not adopt)(−90 000) = −18 000

E(9) = p(adopt}(0) + p(not adopt)(−110 000) = −22 000

E(4) = p(not needed)(−40 000) + p(needed)(110 000) = 65 000

E(5) = p(not needed)(−140 000) + p(needed)(10 000) = −35 000
E(6) = p(not needed)E(8) + p(needed)(60 000) = 36 600
E(7) = p(not needed)E(9) + p(needed)(40 000) = 21 400

E(2) = p(no extra expenses)E(4) + p(extra expenses)E(5) = 40 000

E(3) = p(no modification)E(6) + p(modification)E(7) = 30 520

E(1) = E(2) = 40 000 because E(2) > E(3) and E(2) > 0

BBB should repair the production line.

12.46 The Result of Sensitivity Analysis

Probability Asian demand ≠ 0.1 0.2 0.3 0.4 0.5

Probability Asian demand ↓ 0.9 0.8 0.7 0.6 0.5
E(4) 0.8755 1 1.125 1.25 1.375
E(5) −0.85 −0.7 −0.55 −0.4 −0.25
E(6) 0.525 0.55 0.575 0.6 0.725
E(7) 0.28 0.23 0.24 0.22 0.2
E(2) = 0.6E(4) + 0.4E(5) 0.185 0.32 0.455 0.59 0.725
E(3) = 0.6E(6) + 0.4E(7) 0.427 0.422 0.441 0.448 0.515
E(1) = max{E(2), E(3)} 0.427 0.422 0.455 0.59 0.725
Partnership or no partnership? no no partner partner partner

As long as the probability that Asian demand increases is greater than or

equal to 0.3, then the partnership with China seems to be the preferred
option. Also, the higher the probability, the more confidence in the
partnership option.

348
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

12.47 (a) E(2) = p(rapid growth)E(3) + p(steady growth)E(4)

+ p(slow growth)E(5)
= 16p1 + 8p2 + 0p3
= 16p1 + 8p2

(b) Graphically, all possible values of p1 and p2 are represented by the

area that satisfy p1 + p2 ≤ 1, p1 ≥ 0, and p2 ≥ 0. Furthermore, the values of
p1 and p2 that also satisfy 16p1 + 8p2 < 10 are represented by the shaded
region in the graph.

p2

1.25
16p1 + 8p2 = 10
1
(0.25, 0.75)

p1 + p2 = 1
Area Area
= 0.5-0.14 = 1/2(0.375)(0.75)
= 0.36 = 0.14

0 p1
0.625 1

From the graph, we can identify the estimate of probabilities that imply
introducing the package now is the preferred option (see the shaded
region). For example, (p1, p2) = (0.25, 0.5) would lead to the decision to
introduce now. Note that whenever the values of p1 and p2 do not add up
to 1, then it implies that p3 = 1 − p1 − p2. Also from the graph, we have the
information that roughly 0.36/0.5 = 72% of all possible (p1, p2) values
indicate that introducing the package now is the preferred option. Another
observation is that, for (p1, p2) values with high p1 value (e.g., p1 > 0.625),
the option to wait for the survey result is preferred. The possible
interpretation for this decision is, if the market growth is likely to be rapid,
then waiting for 3 months would not jeopardize LOTell’s opportunity to
gain more market share even in the presence of the competitors.

349
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

Appendix 12A Solutions

12A.1 (a) The decision matrix:

Criteria Weight A B C D E F G
C1 8190 3650 3180 3110 2820 1610 1150
0 0 0 0 0 0 0
Normalized 1.5 10.0 3.6 2.9 2.8 2.4 0.7 0.0
C2 2550 1030 1160 1050 7400 3500 7100
0 0 0 0
Normalized 2.0 10.0 3.1 3.7 3.2 1.8 0.0 1.6
C3 65 12.2 45 35 30 10 28
Normalized 2.5 0.0 9.6 3.6 5.5 6.4 10.0 6.7
C4 8.6 3 6.3 14.1 6 11.8 4.5
Normalized 3.0 5.0 0.0 3.0 10.0 2.7 7.9 1.4
C5 1.0 3 6 7 6 10 4 4
Score 53.1 41.5 36.7 60.2 41.1 53.8 28.1

The best subway route from this data is D.

(b) The revised decision matrix:

Criteria Weight A B C D E F G
C1 8190 3650 3180 3110 2820 1610 1150
0 0 0 0 0 0 0
Normalized 2.0 10.0 3.6 2.9 2.8 2.4 0.7 0.0
C2 2550 1030 1160 1050 7400 3500 7100
0 0 0 0
Normalized 2.0 10.0 3.1 3.7 3.2 1.8 0.0 1.6
C3 65 12.2 45 35 30 10 28
Normalized 2.0 0.0 9.6 3.6 5.5 6.4 10.0 6.7
C4 8.6 3 6.3 14.1 6 11.8 4.5
Normalized 2.0 5.0 0.0 3.0 10.0 2.7 7.9 1.4
C5 2.0 3 6 7 6 10 4 4
Score 56.1 44.5 40.3 54.8 46.4 45.2 27.4

Yes. In this case, A is slightly better.

12A.2 (a) Only Jobs 1, 3, 7, and 8 can not be dominated.

#1 #3 #7 #8
Criterion Weight Spinoff Soutel Ring Jones
Pay 4 1700 2200 2200 2700

350
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

Home 2.5 2 80 250 500

Studies 2 3 4 5 3
Size 1.5 5 150 300 20

(b) The normalized decision matrix:

#1 #3 #7 #8
Criterion Weight Spinoff Soutel Ring Jones
Pay 4 0.0 5.0 5.0 10.0
Home 2.5 10.0 8.4 5.0 0.0
Studies 2 6.0 8.0 10.0 6.0
Size 1.5 10.0 5.1 0.0 9.5
Scores: 52.0 64.7 52.6 66.2

With the given information, Job 8 scores the highest and is therefore the
best.

351
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

Notes for Mini-Case 12.1

1) Examples: small projects, projects that are similar to the past projects, and
projects without major complexities.

2) Complex projects for which a wide experience base does not exist.
Examples: Introducing new technology into a manufacturing line,
experimenting with a variety of daily staff scheduling patterns when
demand varies over the day, determining a good emergency response
strategy for infrequent events such as an outbreak of a communicable
disease or a toxic chemical spill.

3) Other benefits include its use as a communication tool, its ability to do

what-if analysis without having to experiment with the real system, and its
use for training and education.

4) Since the output is a random variable, relying on a single trial could be

very inaccurate as it may not reflect the overall behaviour of the system.
Running many trials will give a much better picture of how the
performance measure behaves as a function of its many stochastic inputs.

5) Ji-Ye would first need to assess how uncertain she is of the plant’s
performance. If she has a great deal of experience with other such plants,
the simulation might not be of much use. If she is uncertain of the plant’s
performance and could use some guidance, then she should consider how
close a fit the simulation being offered is to the environment they are
considering. Is the simulation flexible? Is it useable? Has it been reliable
in the past? How much time will it take to adapt it to her application?

How much she would be willing to pay for it will depend on these factors
as well as how much the uncertainty could impact the plant’s performance,
how long the simulation would take to adapt and use, how much budget
she has for such a project.

352
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

Solutions to All Additional Problems

Note: Solutions to odd-numbered problems are provided on the Student CD-ROM.

12S.1

Suppose you deposit $100 with the bank. The six equally likely possible
outcomes are that you get back nothing, $110, $120,…, $150. So your expected
return is:
(110 + 120 + 130 + 140 + 150)/6 = 108.3

This is equivalent to an interest rate of 8.3%.

12S.2

Consider the fate of a single ship. To build and provision it required 100 000 gold
pieces. The three possible outcomes are that it is lost, that it returns empty—
which is an outcome worth 50 000 gold pieces, since that is what it cost to
build—or that it returns laden with merchandise—which outcome is worth 300
000 gold pieces, the value of the merchandise plus the ship itself.

So the expected value at the end of the year is

0 × 0.25 + 50 000 × 0.25 + 300 000 × 0.5 = 162 500

This represents a 62.5% return on your original investment.

As you increase the number of ships in your fleet, the rate of return remains the
same, but the expected variance is reduced. There is less chance of finding
yourself at the extreme ends of the distribution and either losing everything or
getting a return much greater than 62.5%.

12S.3

You have a 50% chance of getting the first question right, so your expected
income from the first round is
¥100 000 × 0.5 = ¥50 000

To get the ¥200 000 prize for the second round, you must have both passed the
first round and got the second question right. The combined probability of this is
0.5 × 0.5 = 0.25, so your expected income in the second round is
¥200 000 ×0.25 = ¥50 000

353
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

By similar reasoning, your expected income in each subsequent round is ¥50

000, so your total expected income after ten rounds is ¥500 000, and this is what
it would be rational to pay for a chance to play the game.
In the variant where there is no upper limit to the series of questions, it would at
first appear that you have an infinite series of payouts, each of expected value
¥50 000, and that it would therefore be worth paying an infinite amount of money
for the privilege of playing. However, this reasoning is not correct. Even
Japanese game shows have an upper limit to their budget. For example, the
maximum feasible payout is not likely to exceed ¥450 000 000 000, since this
was Japan’s Gross Domestic Product in 2006. But you would reach this payout
after just 22 rounds of the game, so your maximum expected winnings would still
only be 22 × ¥50 000 = ¥1 100 000, and it would not be rational to pay more than
this to play.

12S.4

The first thing to do is to calculate the value of a launch now, three months from
now, and so on. The value of a launch now is an infinite series of payments of
$100 000/month, where the interest rate is 2%. The capitalized cost of this series
is $100 000 / 0.02 = $5 000 000. The capitalized cost of the same series, starting
3 months later, is

$5 000 000(P/F,0.02,3) = 5 000 000 × 0.9423.

A six-month delay gives a capitalized cost of 5 000 000 × 0.888; nine months,
5 000 000 × 0.8368; a year, 5 000 000 × 0.7885.

(Alternatively, we could observe that all options lead to the same outcome at the
end of one year. At that time, the satellite is up and earning; a one-year study
period would be equally reasonable.)

So the expected value of a launch with the U.S. rocket is

−1 200 000 + 5 000 000(0.5 + 0.2 × 0.9423 + 0.1 × 0.888 + 0.05 × 0.8368 + 0.15
× 0.7885)
= −1 200 000 + 5 000 000 × 0.937 = 3.487 million.
Variance = $867 000

Using the Russian rocket gives us

−900 000 + 5 000 000(0.5 + 0.1 × 0.9423 + 0.4 × 0.7885)
= −900 000 + 5 000 000 × 0.9096= 3.648 million.
Variance = $1 009 000

Using the Chinese rocket gives us

−750 000 + 5 000 000(0.4 + 0.15 × 0.9423 + 0.1 × 0.888 + 0.05 × 0.8368 + 0.3 ×
0.7885)

354
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

= −750 000 + 5 000 000 × 0.9085 = 3.792 million.

Variance = $1 084 000

Using the European rocket gives us

−1 000 000 + 5 000 000(0.7 + 0.3 × 0.7885)
= −1 000 000 + 5 000 000 × 0.9365 = 3.682 million.
Variance = $960 000

Using the Japanese rocket gives us

−1 500 000 + 5 000 000(0.9 + 0.05 × 0.9423 + 0.05 × 0.7885)
= −1 500 000 + 5 000 000 × 0.9865 = 3.492 million.
Variance = $735 000

So the Chinese launcher looks like the best bet from the viewpoint of expected
value.

Should we also take variance into account?

One way to think about this question is to consider that, if we change our choice
in order to get a lower variance, we are paying a premium of several hundred
thousand dollars to reduce our uncertainty. Why might we want to do this? Well,
if the company is short of cash, and a twelve-month delay might cause the
company to fail, we might reasonably attach a higher importance to staying in
business than to maximizing expected profit. Secondly, in making plans for the
year ahead, it would be convenient to be fairly sure what our cash flows are
going to be. Are either of these considerations worth losing $300 000 in expected
profit? There is no algorithm that will tell us.

If our company’s resources are large compared with the launch costs, and if we
launch satellites on a regular basis, we will probably consider that variance in
individual cases will average out in the long term, and hence make a decision
based solely on expected value. However, if the company’s entire resources are
just sufficient for one launch, it may be worth paying a premium in exchange for a
lower probability of going out of business.

This is one example of a case where it is worthwhile to distinguish the utility of a

cash flow from its value. To explain this distinction, a comparison may be helpful.
You are offered a chance to play one of two games. It costs ten cents to play
either of them. Game 1 involves tossing a coin once; you get $1 if it comes up
heads. Game 2 involves tossing a coin twice; you get $4 if there are 2 heads,
otherwise you get nothing. Which game would you rather play?

Now consider two different games. It costs all you possess to play either of them.
Game 3 involves tossing a coin once; you get ten times what you possess if it
comes up heads. Game 4 involves tossing a coin twice; you get forty times what
you possess if there are 2 heads, otherwise you get nothing. Which game would
you rather play?
355
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

The idea here is that, while the value of a dollar is always a dollar, the utility of
having ten dollars rather than nothing is much greater than the utility of having a
million and ten dollars rather than a million; the first ten dollars may save you
from starving, whereas ten dollars on top of a million just allows you to leave a
larger tip. There is a branch of economics, utility theory, which provides a basis
for including these considerations in decision-making; however, it is beyond the
scope of this text.

12S.5

To answer this question, we need to choose a study period. How long do we

expect the second company to be making this model of communications device?
Mobile communications is a fast-changing field, so it may not be that long—
certainly not as long as ten years. On the other hand, there is also the question
of our reputation with the other company. If we maintain our reputation, they may
use our batteries in the successor to the current device. Knowledge of the market
would be useful here. Are there many other customers for our batteries, or does
this customer dominate the market?

It might be reasonable to get a high and a low estimate. For the low estimate,
assume that the other company is only one of a number of possible customers,
and that they are only going to be making this model for another three years. For
the high estimate, assume that we really need to maintain a good reputation with
this company, and that fulfilling this order successfully may lead to repeat
business for many years to come—ten years, for example.

Three-Year Study

Suppose the cost of inspecting a battery is c. Then our expected profit on

inspected batteries is
2000 × (5−c) × (P/A,0.25,3) = 2000 × (5−c) × 1.952 = 3904 × (5−c)

If on the other hand we choose not to do the inspection, we definitely get the first
year’s profit:
Profit in First Year, no inspection: 2000 × 5 × (P/F,0.25,1).

The chance that the first lot of batteries contains no defectives is 0.99992000 =
0.819. So we have a 0.819 chance of getting an additional 2000 × 5 ×
(P/F,0.25,2). And we have a 0.819 × 0.819 chance of getting an additional 2000
× 5 × (P/F,0.25,3). So our expected profit is

2000 × 5 × ((P/F,0.25,1) + 0.819(P/F,0.25,2) + 0.671(P/F,0.25,3))

= 2000 × 5 × (0.8 + 0.819 × 0.64 + 0.671 × 0.512)
= 10 000 × 1.67 = 16 700
This gives the equation 3904 × (5−c) = 16 700

356
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

which can be solved to give c = 0.73. So it is worth spending about $0.73 per
battery on inspection.

Ten-Year Study

If we do a ten-year study, the expected profit on inspected batteries is

2000 × (5−c) × (P/A, 0.25, 10) = 2000 × (5−c) × 3.571 = 7142(5−c)

With no inspection, our expected profit is

2000 × 5 × 0.75 × (1 + ∑9i=1 (0.819 × 0.75)i)
= 7500 × (1 + (1.6149−1) / (0.614 × 1.6149))
=7500 × 2.607 = 19 550
Solving for c, we get $2.26.

So our overall conclusion is that it is worth spending at least $0.75 per battery. If
we are concerned about establishing a long-term relationship as a supplier to the
other company, it may be worth spending up to $2.25 per battery.

12S.6

The following C source code is a working example of the Monte Carlo method,
applied to the problem of calculating π.

main()
{
int i, j;
float pi, piEst, realI, realJ, xSquared, ySquared;
double x,y;
FILE *outdata;

pi = 4.0 * atan(1.0);
srand48(1.0);

if ((outdata=fopen("monty3","w")) == NULL)
{
printf("failed to open the write file\n");
exit(1);
}

j = 0;

for (i=1; i<= 1000000; i++)

{
x = 2 * drand48() - 1;
y = 2 * drand48() - 1;

xSquared = x * x;
ySquared = y * y;

if ((xSquared + ySquared) < 1.0)j++;

357
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

realI = i;
realJ = j;

piEst = 4.0 * realJ/realI;

if (i%1000==0)fprintf(outdata,"%d %f %f
\n",i,pi,piEst);
}

You can copy this code to your own machine. To run it, you may need to replace
the calls to srand48 and drand48 with a call to your local random-number
generator.

To apply this method to an economics problem, you would replace x and y with
inflation rate, selling price, or other variables. The current version of the code will
give you a random distribution of values between −1 and 1, which will probably
not match the possible values of the variables. If the actual range for selling price
of a product is $4.00 to $6.00, the expression
price = 4.00 + 2 * drand48();
will give you a uniform distribution of prices over this range.

But what if you know that the distribution is not uniform, which it may not be?
This can be handled by approximating the probability function with a histogram,
and mapping the interval (0,1) to the cumulative value of the histogram. Thus, for
example, if there is a 10% probability that the price will be in the range ($4.00,
4.50), you can use the expression
if (0 < drand48() < 0.1) price = $4.25
and so on for the rest of the range.

Other Applications

The Monte Carlo method has applications to many other areas of engineering.
For example, it has been used to simulate the turbulent combustion of a diesel
engine spray injection; the spray and the compressed air are represented as
made up of a large number of “packets,” each of which has a weighted chance of
mixing with another packet of fuel, air, or burned gases. Many runs of this model
give a reasonable approximation to actual performance.

12S.7

It is possible to use the “Goldilocks Strategy” of examining the best case, worst
case, and median case for each strategy. However, this is not as good as using
the Monte Carlo method—for some of the input parameters, the PW of each
option is a nonlinear function of the parameter value, so looking at three points
does not give you an adequate idea of the overall distribution of PW.

358
Copyright © 2013 Pearson Canada Inc.
 Chapter 12 - Dealing With Risk

It is absolutely essential to use the same time period, starting and ending at the
same times, for both strategies. It seems tempting to start the time period for the
lightweight-display strategy later, perhaps when it goes into production, in order
to make the comparison “fair.” But you cannot escape the fact that money has a
time value so easily.

Anything common to both the strategies can be left out. For example, the nine
months between the present and the heavyweight-display strategy going in to
production can be omitted, since it is the same for both strategies. You could also
omit the fixed costs of $30 000/month, since they are the same in both cases.

Once you have set up the MC method, it’s as easy to run 1 000 000 trials as it is
to run 100. You should run at least 10 000 to get an adequate sample.

After you have 10 000 or more present-worths for each strategy, how should you
best display them?

One possibility is to just calculate the mean value of each strategy. (Plotting a
graph of the mean value versus number of trials will show the evolution of this
mean value with number of trials; this is one way of deciding whether you’ve
done enough trials.) This choice involves throwing away almost all the
information you have generated. You may still have enough—if one PW is clearly
higher than the other, that would be the sensible choice—but you can make
better use of the additional values you calculated.

The first improvement would be to calculate the variance as well as the mean for
each strategy. If the strategy with the higher mean also has a very high variance,
it might not be the best choice. Suppose, for example, that it gives a 30%
possibility of going out of business, while the strategy with the lower mean value
gives no possibility of going out of business.

A more informative picture of the distribution of present value can be obtained by

plotting a histogram of the results for each strategy, grouping present value into
about ten intervals and counting the number of cases that fall into each interval.

359
Copyright © 2013 Pearson Canada Inc.