ESI 6323 Sample Questions for Second Test

Problem 1.

A manufacturing firm in Calgary produces an item using a 3-month time supply rule (i.e.,
each production lot covers 3 months of demand). An analyst, attempting to introduce a
more logical approach to selecting run quantities, has obtained the following estimates of
characteristics of the item:
Annual demand rate D = 4000 units/year
Fixed order cost S = $5
Variable order cost c = $4 per 100 units
Holding cost rate h = 25% (annually)
Assume the ordered amount is delivered instantaneously upon ordering.

a) What is the EOQ of the item?

EOQ = sqrt(2*5*4000/0.01) = 2000

b) What is the time between consecutive replenishments of the item when the EOQ is
used?

T* = EOQ/D = 2000/4000 = ½ year

c) The production manager insists that the S = $5 figure was pulled out of the air, that is,
it is only a guess. Therefore, she insists on using her simple 3-month supply rule.
Determine the range of S values for which the EOQ (based on S = $5) would be
preferable (in terms of lower total ordering and holding costs) to the 3-month supply.

3-month rule: Q = 1000

Cost of 3-month rule = S * 4000/1000 + 0.01*1000/2 = 4S + 5
EOQ at S = 5: 2000
Cost of Q = 2000: S*4000/2000 + 0.01*2000/2 = 2S + 10.
We need 2S + 10 ≤ 4S + 5, i.e., 2S ≥ 5, or S ≥ $2.50

1
Problem 2.

The Wagner Company supplies electric motors to Electronic Distributors, Inc. on a

delivered-price basis. Wagner has the responsibility for providing transportation. The
traffic manager has three transportation service choices for delivery—rail, piggyback, and
truck. He has compiled the following information:

Transport Mode Transit Time, Days Rate, $/Unit Shipment Size, Units
Rail 16 25.00 10,000
Piggyback 10 44.00 7,000
Truck 4 88.00 5,000

Electronic Distributors purchases 50,000 units per year at a delivered contract price of
$500 per unit. Inventory carrying costs for both companies are 25 percent per year.
Which mode of transportation should Wagner select?

Rail
Cycle stock cost = (0.25)*(500)*(10,000/2) = 625,000
Average pipeline inventory = (16/365)*(500)*(0.25)*(50,000) = 273,972.60
Transportation cost = 25*50,000 = 1,250,000
Total cost = $2,148,972.60

Piggyback
Cycle stock cost = (0.25)*(500)*(7,000/2) = 437,500
Average pipeline inventory = (10/365)*(500)*(0.25)*(50,000) = 171,232.88
Transportation cost = 44*50,000 = 2,200,000
Total cost = $2,808,732.88

Truck
Cycle stock cost = (0.25)*(500)*(5,000/2) = 312,500
Average pipeline inventory = (4/365)*(500)*(0.25)*(50,000) = 68,493.15
Transportation cost = 44*50,000 = 4,400,000
Total cost = $4,780,993.15

2
Problem 3.

The weighty trash bag company has the following price schedule for its large trash can
liners. For orders of less than 500 bags, the company charges 30 cents per bag; for orders
of 500 or more but fewer than 1,000 bags, it charges 29 cents per bag; and for orders of
1,000 or more, it charges 28 cents per bag (an all-units discount structure).

The weighty trash bag company’s only customer is Publix supermarkets, who consume
the bags at a steady rate of 600 per year. It costs Publix $8 to place an order, and the
Publix holding cost rate is 20%.

(a) What is the optimal trash bag order size for Publix?

( 2 ) ( 8) ( 600 ) = 400
Q(0) =
( 0.2 ) ( 0.3)
( 2 ) ( 8) ( 600 ) = 406
Q(1) =
( 0.2 ) ( 0.29 )
( 2 ) ( 8) ( 600 ) = 414
Q(2) =
( 0.2 ) ( 0.28)

The only realizable EOQ value is Q(0) = 400. We need to check cost at Q(0) = 400, and
at the breakpoints of 500 and 1000.

G0(400) = (600)(0.3) + ( 2 ) ( 8 ) ( 600 ) ( 0.2 ) ( 0.3) = $204

G1(500) = (500)(0.29) + (600)(8)/500 + (0.2)(0.29)(500)/2 = $198.10
G2(1000) = (1000)(0.28) + (600)(8)/1000 + (0.2)(0.28)(1000)/2 = $200.80

So the optimal order quantity is 500.

3
(b) It costs the weighty trash bag company $24 each time Publix places an order, it costs
them 20 cents to produce a bag, and they also use a holding cost rate of 20%. The
weighty trash bag company produces at a steady rate of 600 per year (the same rate as
demand) and each time an order is placed by Publix, it is immediately transported to
Publix (with negligible lead time; that is, the bag producer builds up inventory at a
steady rate until the exact order/delivery quantity is on hand, at which point it is
immediately shipped to Publix, and this process repeats itself infinitely many times).
What is the Publix order quantity that would maximize the profit of the weighty trash
bag company?

( 2 ) ( 24 ) ( 600 )
Q* = = 848.5
( 0.2 ) ( 0.2 )

(c) What order quantity would minimize total supply chain costs? (assume that the 20
cent production cost includes the per unit transportation cost to Publix.)

The optimal order quantity for the system must be between 500 and 848.53,
which implies a selling price of 0.29. Let us consider the system costs at this
price.

Setup Costs = supplier + retailer setup costs = (24 + 8)(600)/Q

Holding Costs = supplier + retailer holding costs = (0.2)(0.2 + 0.29)Q/2

( 2 ) ( 24 + 8) ( 600 )
Q* = = 626
( 0.2 ) ( 0.2 + 0.29 )

4
Problem 4

Consider a simple two-stage supply chain where a manufacturer directly supplies a

retailer with a single product. Assume that the manufacturer is the only supplier for the
retailer and the retailer is the manufacturer’s only customer for the product. The retailer
observes customer demand for the product at a steady rate of D units per year and no
shortages are allowed. The manufacturer’s production rate is exactly matched to the
demand rate. Each time the retailer places an order to the manufacturer it costs $S in
administrative and processing costs. The retailer incurs a cost of $H for holding a unit in
inventory for one year. When the retailer places an order, the manufacturer makes a
shipment, which costs the manufacturer $β S in shipping costs (a fixed cost per shipment;
assume a constant or negligible lead time). The manufacturer incurs a cost of $ αH for
holding a unit in inventory for one year. Each time the retailer places an order it is for the
same quantity, Q.

(a) Draw the graph of the inventory level over time for both the retailer and
manufacturer below.

Retailer’s
Inventory
level

time
Manufacturer’s
Inventory
level

time

(b) In terms of the parameters given, what is the retailer’s optimal order quantity?

QR* = sqrt(2SD/H)

5
(c) In terms of the parameters given, what is the optimal quantity for the
manufacturer to ship?

QM* = sqrt(2β SD/α H)

(d) Write the sum of average annual retailer plus manufacturer order/shipment plus
holding costs.

ATC = SD/Q + HQ/2 + β SD/Q + α HQ/2 = (β +1)SD/Q + (α +1)HQ/2

(e) What is the optimal order/shipment quantity for minimizing average annual
retailer plus manufacturer order/shipment plus holding costs?

Q* = sqrt(2(β + 1)SD/((α + 1)H))

(f) In terms of the parameters α and β , what condition is required for the
manufacturer’s optimal shipment quantity to equal the retailer’s optimal order
quantity?

α=β

(g) If the condition required in your answer to part (f) is violated, should the retailer
and manufacturer work together to ensure that the order/shipment quantity in part
(e) is used?

Ideally yes, provided they can agree on an arrangement to share the savings.

6
Problem 5.

(a) Given the following data for demand at the XYZ Company, calculate the monthly
forecast for 2003 using a 3-month moving average. Calculate the MAD and the
tracking signal. Is this a good forecast?

Period Demand Forecast Error MAD Bias TS

Oct 02 850
Nov 02 950
Dec 02 900
Jan 03 1000
Feb 03 950
Mar 03 1050
Apr 03 850
May 03 1100
Jun 03 900
Jul 03 1150
Aug 03 1100
Sep 03 900
Oct 03 1000
Nov 03 800
Dec 03 1000

Answer:

Period Demand Forecast Error MAD Bias TS

Oct 02 850
Nov 02 950
Dec 02 900
Jan 03 1000 900 -100 100 -100 -1
Feb 03 950 950 0 50 -100 -2
Mar 03 1050 950 -100 66.7 -200 -3
Apr 03 850 1000 150 87.5 -50 -.57
May 03 1100 950 -150 100 -200 -2
Jun 03 900 1000 100 100 -100 -1
Jul 03 1150 950 -200 114.3 -300 -2.62
Aug 03 1100 1050 -50 106.3 -350 -3.29
Sep 03 900 1050 150 111.1 -200 -1.8
Oct 03 1000 1050 50 105 -150 -1.43
Nov 03 800 1000 200 113.6 50 .44
Dec 03 1000 900 -100 112.5 -50 -.44

The standard deviation of the random element of demand is approximately

140.63. The tracking signal indicates that this forecast method tends to under-
forecast demand, but not to the extreme.

7
(b) Using data from part (a), calculate the monthly forecast for 2003 using simple
exponential smoothing with an α = .2. Calculate the MAD and the tracking signal.
How does this forecast compare with the previous one?

Period Demand Level Forecast Error MAD Bias TS

Oct 02 850
Nov 02 950
Dec 02 900 983
Jan 03 1000
Feb 03 950
Mar 03 1050
Apr 03 850
May 03 1100
Jun 03 900
Jul 03 1150
Aug 03 1100
Sep 03 900
Oct 03 1000
Nov 03 800
Dec 03 1000

Answer:

Period Demand Level Forecast Error MAD Bias TS

Oct 02 850
Nov 02 950
Dec 02 900 983
Jan 03 1000 986.4 983.0 -17.0 17.0 -17.0 -1.0
Feb 03 950 979.1 986.4 36.4 26.7 19.4 0.7
Mar 03 1050 993.3 979.1 -70.9 41.4 -51.5 -1.2
Apr 03 850 964.6 993.3 143.3 66.9 91.8 1.4
May 03 1100 991.7 964.6 -135.4 80.6 -43.5 -0.5
Jun 03 900 973.4 991.7 91.7 82.4 48.2 0.6
Jul 03 1150 1008.7 973.4 -176.6 95.9 -128.5 -1.3
Aug 03 1100 1027.0 1008.7 -91.3 95.3 -219.8 -2.3
Sep 03 900 1001.6 1027.0 127.0 98.8 -92.8 -0.9
Oct 03 1000 1001.3 1001.6 1.6 89.1 -91.3 -1.0
Nov 03 800 961.0 1001.3 201.3 99.3 110.0 1.1
Dec 03 1000 968.8 961.0 -39.0 94.3 71.0 0.8

The simple exponential smoothing has a smaller MAD and therefore a smaller
standard deviation than the moving average. The tracking signal indicates that
simple exponential smoothing over-forecasts a little, where the moving average
under-forecasts a little.

8
Problem 6.

The demand for Krispee Crunchies, a favorite breakfast cereal of people born in the
1960s, is experiencing a decline. The company wants to monitor demand for this product
closely as it nears the end of its lifecycle, and decides to use Holt’s method with α = 0.1
and β = 0.2. At the end of December, the January estimate for the average number of
cases sold per month was 900,000, and the trend was –50,000 per month. The following
table shows the actual sales history for January, February, and March. Generate forecasts
for February, March, and April.

Month Sales
January 890,000
February 800,000
March 825,000

LJ = 0.1(890,000) + 0.9(900,000 – 50,000) = 854,000

TJ = 0.2(854,000 – 900,000) + 0.8(–50,000) = –49,200

Forecast for Feb: 854,000 – 49,200 = 804,800

LF = 0.1(800,000) + 0.9(854,000 – 49,200) = 804,320

TF = 0.2(804,320 – 854,000) + 0.8(–49,200) = –49,296

Forecast for Mar: 804,320 – 49.296 = 755,024

LM = 0.1(825,000) + 0.9(804,320 – 49,296) = 762,021.6 ≈ 762,022

TM = 0.2(762,022 – 804,320) + 0.8(–49,296) = –47,896.4 ≈ –47,897

Forecast for Apr: 762,022 – 47,897 = 714,125

9
Problem 7

The Northville Post Office experiences a seasonal pattern of daily mail volume every
week. The following data for two representative weeks are expressed in thousands of
pieces of mail.

Day Week 1 Week 2

Sunday 5 8
Monday 20 15
Tuesday 30 32
Wednesday 35 30
Thursday 49 45
Friday 70 70
Saturday 15 10
Total 224 210

(a) Calculate a seasonal factor for each day of the week.

Day Seasonal factor

Sunday
Monday
Tuesday
Wednesday
Thursday
Friday
Saturday

(b) If the postmaster estimates that there will be 230,000 pieces of mail to sort next
week, forecast the volume for each day of the week.

Day Forecast
Sunday
Monday
Tuesday
Wednesday
Thursday
Friday
Saturday

10
 Solution option 1: Get seasonal factors by dividing by average demand in the week:

Day Week 1 SF1 Week 2 SF2 Avg SF

Sunday 5 5/32 = 0.15625 8 8/30 = 0.26667 0.21146
Monday 20 20/32 = 0.62500 15 15/30 = 0.50000 0.56250
Tuesday 30 30/32 = 0.93750 32 32/30 = 1.06667 1.00209
Wednesday 35 35/32 = 1.09375 30 30/30 = 1.00000 1.04688
Thursday 49 49/32 = 1.53125 45 45/30 = 1.50000 1.51563
Friday 70 70/32 = 2.18750 70 70/30 = 2.33333 2.26042
Saturday 15 15/32 = 0.46875 10 10/30 = 0.33333 0.40104
Total 224 210
Average 224/7 = 32 210/7 = 30

Average daily demand is 230,000/7 = 32,857

Day Calculation Forecast

Sunday 0.21146(32,857) 6,948
Monday 0.56250(32,857) 18,482
Tuesday 1.00209(32,857) 32,926
Wednesday 1.04688(32,857) 34,397
Thursday 1.51563(32,857) 49,799
Friday 2.26042(32,857) 74,271
Saturday 0.40104(32,857) 13,177

Solution option 2a: Using the moving average from the book, filling in first Wed. value back to Sunday,
and last Wed. value forward to Saturday.

Day Value Deseasonalized SF Averaged SFs Forecast

Sunday 5 32.00 0.156250 0.205979 6767.8
Monday 20 32.00 0.625000 0.556686 18291.0
Tuesday 30 32.00 0.937500 0.989680 32517.9
Wednesday 35 32.00 1.093750 1.046875 34397.2
Thursday 49 32.43 1.511013 1.505507 49466.4
Friday 70 31.71 2.207207 2.270270 74594.3
Saturday 15 32.00 0.468750 0.401042 13177.0
Sunday 8 31.29 0.255708
Monday 15 30.71 0.488372
Tuesday 32 30.71 1.041860
Wednesday 30 30.00 1.000000
Thursday 45 30.00 1.500000
Friday 70 30.00 2.333333
Saturday 10 30.00 0.333333

11
Solution option 2b: Using the moving average from the book, filling in avg of first Wed. & Thurs. values
back to Sunday, and last Tues. & Wed. values forward to Saturday.

Day Value Deseasonalized SF Averaged SFs Forecast

Sunday 5 32.21 0.155211 0.205459 6750.8
Monday 20 32.21 0.620843 0.554607 18222.7
Tuesday 30 32.21 0.931264 0.986562 32415.5
Wednesday 35 32.00 1.093750 1.046875 34397.2
Thursday 49 32.43 1.511013 1.496683 49176.5
Friday 70 31.71 2.207207 2.256545 74143.3
Saturday 15 32.00 0.468750 0.399081 13112.6
Sunday 8 31.29 0.255708
Monday 15 30.71 0.488372
Tuesday 32 30.71 1.041860
Wednesday 30 30.00 1.000000
Thursday 45 30.36 1.482353
Friday 70 30.36 2.305882
Saturday 10 30.36 0.329412

12
Problem 8

The XYZ Company has an assembly plant in Cincinnati and its parts plant in
Indianapolis. Parts are transported from Indianapolis to Cincinnati using trucks. Each
shipment costs $100. The Cincinnati plant assembles and sells 300 finished products each
day and operates 50 weeks a year (working 5 days per week). Part #456 costs $50 and
XYZ Company incurs a holding cost of 20 percent per year. How many of part #456
should XYZ Company put in each shipment? What is the cycle inventory of part #456 at
XYZ Company?

Answer: Q* = √2DS/hC
= √(2(300 x 50 x 5)$100)/(.2 x $50)
= 1224.745 ≈ 1225
Average cycle inventory = Q*/2
= 1225/2
= 612.5

13
Problem 9

Using the information from the previous problem, suppose XYZ gets price breaks on part
#456 such that for shipments between 2500 and 3499, the cost is $49.98 per unit, and for
shipments of 3500 and greater, the cost is $49.90 per unit (for all units). How many of
part #456 should XYZ Company put in each shipment?

Answer: At C0, Q* = √2DS/hC

= √(2(300 x 50 x 5)$100)/(.2 x $50)
= 1224.745 ≈ 1225
Avg. annual cost = (300*50*5)*50 +
sqrt(2*100*(300*50*5)*(0.2*50))
= $3,762,247.45

At C1, Q* = sqrt(2(300 x 50 x 5)$100)/(.2 x $49.98))

≈ 1225
Therefore, we need to check the cost at Q = 2500:
Avg. annual cost = (300*50*5)*49.98 + 100(300*50*5)/2500 +
(0.2*49.98)2500/2
= $3,763,995.00

At C2, Q* = sqrt(2(300 x 50 x 5)$100)/(.2 x $49.95))

≈ 1226
Avg. annual cost = (300*50*5)*49.90 + 100(300*50*5)/3500 +
(0.2*49.90)3500/2
= $3,762,107.86

So the optimal lot size is 3500.

14
Problem 10

Formulate a linear programming model for the following aggregate planning problem:

Number of time periods: 3

Demand schedule:
Current regular time capacity:
Overtime capacity:
Unit variable cost:
Inventory carrying cost:
Cost to add one unit of capacity:
Cost to reduce capacity by one unit:
D1 = 45, D2 = 52, D3 = 60
45 units per period
25% of regular time capacity
$800 on regular time, $1,000 on overtime
$12 per unit per period
$240
$180

No shortages are allowed

No initial inventory exists

Xt = production in period t during regular time,

Ot = production in period t during overtime,
It = Inventory at end of period t,
Ht = capacity increase in period t,
Ft = capacity decrease in period t.

Minimize 800(X1 + X2 + X3) + 1000(O1 + O2 + O3) + 12(I1 + I2 + I3)

+ 240(H1 + H2 + H3) + 180(F1 + F2 + F3)

Subject to: X1 ≤ 45 + H1 – F1
O1 ≤ 0.25(45 + H1 – F1)
X2 ≤ 45 + H1 – F1 + H2 – F2
O2 ≤ 0.25(45 + H1 – F1 + H2 – F2)
X3 ≤ 45 + H1 – F1 + H2 – F2 + H3 – F3
O3 ≤ 0.25(H1 – F1 + H2 – F2 + H3 – F3)
X1 + O1 = 45 + I1
I1 + X2 + O2 = 52 + I2
I2 + X3 + O3 = 60 + I3
Xt, It, Ot, Ht, Ft ≥ 0, t = 1, 2, …, 3.

15
Problem 11

L-Print manufacturing makes laser printers. One plant assembles the L-4000 model.
Standards indicate that one worker can assemble five printers per day. This model costs
about $350 to make (including labor), and the company estimates it costs $7 to hold one
printer in inventory for one month. Workers earn $1600 per month and can be hired for
$2000 each. Laying off a worker costs $2600. Currently, there are 12 workers in the
assembly department. If a printer is backordered, the cost is $35 per unit per month. The
inventory on hand at the end of June was 300 units (Assume zero inventory at the end of
December, no overtime is permitted).

Month Working days Forecasted Demand

July 21 1060
August 22 990
September 21 840
October 23 1040
November 19 1290
December 20 690

(a) Develop an aggregate plan for L-Print based on a chase (zero inventory) strategy
(with no backorders).

We determine the capacity (in terms of number of workers) required to meeting demand
in each month, and hire/fire accordingly.
Working Net Required Hiring/ Workforce Holding Backorder Production
Month days Demand Production workers Layoffs Cost Cost Cost Cost
July 21 760 760 8 Layoff 4 $ 12,800 $ - $ - $ 266,000
August 22 990 990 9 Hire 1 $ 14,400 $ - $ - $ 346,500
September 21 840 840 8 Layoff 1 $ 12,800 $ - $ - $ 294,000
October 23 1040 1040 10 Hire 2 $ 16,000 $ - $ - $ 364,000
November 19 1290 1290 14 Hire 4 $ 22,400 $ - $ - $ 451,500
December 20 690 690 7 Layoff 7 $ 11,200 $ - $ - $ 241,500 Grand Total
Total Cost $ 45,200 $ 89,600 $ - $ - $ 1,963,500 $ 2,098,300

16
(b) Develop an aggregate plan for L-Print using a level strategy, where the production
output is the same in each month, the number of workers is the same in each month,
backorders are allowed, and all demand is met during the six months.

Working Net Required Hiring/ Workforce Holding Backorder Production

Month days Demand Production workers Layoffs Cost Cost Cost Cost
July 21 760 935 10 Layoff 2 $ 16,000 $ 1,225 $ - $ 327,250
August 22 990 935 10 $ 16,000 $ 840 $ - $ 327,250
September 21 840 935 10 $ 16,000 $ 1,505 $ - $ 327,250
October 23 1040 935 10 $ 16,000 $ 770 $ - $ 327,250
November 19 1290 935 10 $ 16,000 $ - $ 8,575 $ 327,250
December 20 690 935 10 $ 16,000 $ - $ - $ 327,250 Grand Total
5610 Total Cost $ 5,200 $ 96,000 $ 4,340 $ 8,575 $ 1,963,500 $ 2,077,615

(c) Which of the two solutions above would you recommend to L-Print Management?
Justify your answer.

The level strategy is cheaper, and has far less layoffs, although we do have a shortage of
245 units for one month (November). If the $35 is a true reflection of the shortage cost,
then we should go with the level strategy.

17
(d) Formulate the aggregate planning problem for L-Print as a linear programming model
assuming no backordering is allowed. Be sure to define all variables and include all of
the required constraints.

Define:

Ht = # hires, beginning of pd. t

Lt = # layoffs, beginning of pd. t
WFt = # workers in pd. t
Xt = production, pd. t
It = inventory, end pd. t
Dt = net demand pd. t
WDt = working days, pd. t

2000�t =1 H t + 2600�t =1 Lt + 1600�t =1WFt + 7�t =1 I t

6 6 6 6
Min
It-1 + Xt = Dt + It, t = 1, …, 6,
WFt = WFt-1 + Ht – Lt, t = 1, …, 6,
Xt ≤ 5WDt*WFt, t = 1, …, 6
Xt Ht, It, Lt, WFt ≥ 0, t = 1, …, 6.

18