+ All Categories
Transcript
Page 1: Newsvendor Model - The University of Texas at Dallasmetin/Or6366/Folios/Inventory/scnewsvendor.pdf · Newsvendor Model . utdallas.edu/~metin 2 ... 6 months: Seasonal Camera, ... production

utdallas.edu/~metin 1

Newsvendor Model

Page 2: Newsvendor Model - The University of Texas at Dallasmetin/Or6366/Folios/Inventory/scnewsvendor.pdf · Newsvendor Model . utdallas.edu/~metin 2 ... 6 months: Seasonal Camera, ... production

utdallas.edu/~metin 2

Motivation

Determining optimal orders

– Single order in a season

– Short lifecycle items

1 month: Printed Calendars, Rediform

6 months: Seasonal Camera, Panasonic

18 months, Cell phone, Nokia

Motivating Newspaper Article for toy manufacturer Mattel

Mattel [who introduced Barbie in 1959 and run a stock out for several years then on] was hurt last year by inventory cutbacks at Toys ―R‖ Us, and officials are also eager to avoid a repeat of the 1998 Thanksgiving weekend. Mattel had expected to ship a lot of merchandise after the weekend, but retailers, wary of excess inventory, stopped ordering from Mattel. That led the company to report a $500 million sales shortfall in the last weeks of the year ... For the crucial holiday selling season this year, Mattel said it will require retailers to place their full orders before Thanksgiving. And, for the first time, the company will no longer take reorders in December, Ms. Barad said. This will enable Mattel to tailor production more closely to demand and avoid building inventory for orders that don't come.

- Wall Street Journal, Feb. 18, 1999

Page 3: Newsvendor Model - The University of Texas at Dallasmetin/Or6366/Folios/Inventory/scnewsvendor.pdf · Newsvendor Model . utdallas.edu/~metin 2 ... 6 months: Seasonal Camera, ... production

utdallas.edu/~metin 3

Sales

Seasons

Supervalu – Post-New Year's. January/February

– Super Bowl . January/February

– Allergy Season. March/April

– Back to School/College. September/October

– Cough, Cold and Flu seasons.

September/October

– Baking Season. November/December

Sam’s Club – Health and Wellness. January, features exercise

equipment, supplements and vitamins, items tied to

shoppers' New Year's resolutions.

– Big Game. Late January to February, features Super

Bowl party products.

– Spring. March to May, includes Easter, Graduation

Day, Mother's Day, Spring Gardening—all at the same

time.

– Back to College/Back to School. July and August

– Pink/Women's Health. October, includes displays of

pink products and stores offer women's health

screenings.

– Fall Gatherings. Late September through November

– Date. Stands for 'the day after Thanksgiving event,' aka

Black Friday. Includes gifts and splurge items.

– Holiday Entertaining and Gifting. November, begins

the day after Date, see above.

WalMart – Superbowl or New Year's Resolutions or other themes. January

– Lawn and Garden or related theme. April

– Back to School/College. July through August

– Gifts for children; early entertaining décor. October, November

– Last-minute gifts, stocking stuffers, food/entertaining. December

Target – Organization and Storage.

January

– Back to School/College.

July/August

Bas

ed o

n: A

sea

son (

or

13)

for

Shoppin

g. B

y S

. N

assa

uer

in W

SJ

Ayg 1

7, 2011.

Page 4: Newsvendor Model - The University of Texas at Dallasmetin/Or6366/Folios/Inventory/scnewsvendor.pdf · Newsvendor Model . utdallas.edu/~metin 2 ... 6 months: Seasonal Camera, ... production

utdallas.edu/~metin 4

Example: Apparel Industry

How many L.L. Bean Parkas to order?

Demand

Di

Proba-

bility of

demand

being

this size

Cumulative

Probability

of demand

being this size

or less, F(.)

Probability

of demand

greater

than this

size, 1-F(.)

4 .01 .01 .99

5 .02 .03 .97

6 .04 .07 .93

7 .08 .15 .85

8 .09 .24 .76

9 .11 .35 .65

10 .16 .51 .49

11 .20 .71 .29

12 .11 .82 .18

13 .10 .92 .08

14 .04 .96 .04

15 .02 .98 .02

16 .01 .99 .01

17 .01 1.00 .00

Expected demand is 1,026 parkas,

order 1026 parkas regardless of costs?

Demand data / distribution

Cost per parka = c = $45

Sale price per parka = p = $100

Discount price per parka = $50

Holding and transportation cost = $10

Salvage value per parka = s = 50-10=$40

Profit from selling parka = p-c = 100-45 = $55

Cost of overstocking = c-s = 45-40 = $5

Cost/Profit data

Had the costs and demand been symmetric,

we would have ordered the average demand.

Cost of understocking=$55

Cost of overstocking=$5

Costs are almost always antisymmetric.

Demand is sometimes antisymmetric

Page 5: Newsvendor Model - The University of Texas at Dallasmetin/Or6366/Folios/Inventory/scnewsvendor.pdf · Newsvendor Model . utdallas.edu/~metin 2 ... 6 months: Seasonal Camera, ... production

utdallas.edu/~metin 5

Optimal Order Q*

p = sale price; s = outlet or salvage price; c = purchase price

CSL = Probability that demand will be at or below reorder point (ROP or Q)

Raising the order size if the order size is already optimal

Expected Marginal Benefit =

=P(Demand is above stock)*(Profit from sales)=(1-CSL)(p - c)

Expected Marginal Cost =

=P(Demand is below stock)*(Loss from discounting)=CSL(c - s)

Define Co= c-s=overstocking cost; Cu=p-c=understocking cost

(1-CSL)Cu = CSL Co

CSL= Cu / (Cu + Co)

917.0555

55)P(DemandCSL *

ou

u

CC

CQ

Page 6: Newsvendor Model - The University of Texas at Dallasmetin/Or6366/Folios/Inventory/scnewsvendor.pdf · Newsvendor Model . utdallas.edu/~metin 2 ... 6 months: Seasonal Camera, ... production

utdallas.edu/~metin 6

Optimal Order Quantity

0

0.2

0.4

0.6

0.8

1

1.2

4 5 6 7 8 9 10 11 12 13 14 15 16 87

Cumulative

Probability

Optimal Order Quantity = 13(‘00)

0.917

Page 7: Newsvendor Model - The University of Texas at Dallasmetin/Or6366/Folios/Inventory/scnewsvendor.pdf · Newsvendor Model . utdallas.edu/~metin 2 ... 6 months: Seasonal Camera, ... production

utdallas.edu/~metin 7

Marginal Profits at L.L. Bean Approximate additional (marginal) expected profit from ordering 1(‗00) extra parkas

if 10(‘00) are already ordered

=(55.P(D>1000) - 5.P(D≤1000)) 100=(55.(0.49) - 5.(0.51)) 100 =2440

Approximate additional (marginal) expected profit from ordering 1(‗00) extra parkas if 11(‘00) are already ordered

=(55.P(D>1100) - 5.P(D≤1100)) 100=(55.(0.29) - 5.(0.71)) 100 =1240

Additional

100s

Expected

Marginal Benefit

Expected

Marginal Cost

Expected Marginal

Contribution

1011 5500.49 = 2695 500.51 = 255 2695-255 = 2440

1112 5500.29 = 1595 500.71 = 355 1595-355 = 1240

1213 5500.18 = 990 500.82 = 410 990-410 = 580

1314 5500.08 = 440 500.92 = 460 440-460 = -20

1415 5500.04 = 220 500.96 = 480 220-480 = -260

1516 5500.02 = 110 500.98 = 490 110-490 = -380

1617 5500.01 = 55 500.99 = 495 55-495 = -440

<See expected_inventory_cost.xls to compute expected single-period costs. >

Page 8: Newsvendor Model - The University of Texas at Dallasmetin/Or6366/Folios/Inventory/scnewsvendor.pdf · Newsvendor Model . utdallas.edu/~metin 2 ... 6 months: Seasonal Camera, ... production

utdallas.edu/~metin

optimal. same theyield they ;equivalent are Q)]E[Cost(x,Min and Q)],E[Profit(xMax

Qin Constant )c)E(Demand-(pQ)]E[Cost(x,Q)],E[Profit(x

c)x(pQ xifc)x -(p

Q xifc)x -(pQ)Cost(x,Q)Profit(x,

Q xif Q)-c)(x-(p

Q xif x)-s)(Q-(cQ)Cost(x,

Q xif c)Q-(p

Q xif x)-s)(Q-(c-c)x-(pQ)Profit(x,

quantity.order :Q f(x); pdf with demand :x

unit.per cost):c value;salvage :s price;:(p

QQ

8

Cost or Profit; Does it matter?

Page 9: Newsvendor Model - The University of Texas at Dallasmetin/Or6366/Folios/Inventory/scnewsvendor.pdf · Newsvendor Model . utdallas.edu/~metin 2 ... 6 months: Seasonal Camera, ... production

utdallas.edu/~metin 9

Computing the Expected Profit with Normal Demands

σ,1))μ,,normdist(Q(1 Q c)(p σ,1)μ,,normdist(Q Q s)-(c -

σ,0)μ,,normdist(Q σ s)-(p - σ,1)μ,,normdist(Qμ s)-(p Profit Expected

σdeviation standard andμ mean with Normal is demand that theSuppose

dx f(x) Q)Profit(x,Profit Expected

Example: Follett Higher Education Group (FHEG) won the contract to operate the UTD bookstore. On

average, the bookstore buys textbooks at $100, sells them at $150 and unsold books are salvaged at

$50. Suppose that the annual demand for textbooks has mean 8000 and standard deviation 2000.

What is the annual expected profit of FHEG from ordering 10000 books? What happens to the profit

when standard deviation drops to 20 and order drops to 8000?

Expected Profit is $331,706 with order of 10,000 and standard deviation of 2000:

= (150-50) *8000*normdist(10000,8000,2000,1)-(150-50)*2000*normdist(10000,8000,2000,0)

-(100-50)*10000*normdist(10000,8000,2000,1)+(150-100)*10000*(1-normdist(10000,8000,2000,1))

Expected Profit is $399,960 with order of 8000 and standard deviation of 20:

= (150-50)*8000*normdist(8000,8000,20,1)-(150-50)*20* normdist(8000,8000,20,0)

-(100-50)*8000*normdist(8000,8000,20,1)+(150-100)*8000*(1-normdist(8000,8000,20,1))

Page 10: Newsvendor Model - The University of Texas at Dallasmetin/Or6366/Folios/Inventory/scnewsvendor.pdf · Newsvendor Model . utdallas.edu/~metin 2 ... 6 months: Seasonal Camera, ... production

utdallas.edu/~metin 10

Ordering Women’s Designer Boots

Under Capacity Constraints

Autumn Leaves Ruffle

Retail price $150 $200 $250

Purchase price $75 $90 $110

Salvage price $40 $50 $90

Mean Demand 1000 500 250

Stand. deviation of demand 250 175 125

Available Store Capacity is 1,500.

Ignoring this capacity constraint yields:

Autumn Leaves Ruffle

pi-ci 150-75=$75 200-90=$110 250-110=$140

ci-si 75-40=$35 90 - 50 = $40 110-90 = $20

Critical Fractile 75/110 = 0.68 110/150= 0.73 140/160=0.875

zi 0.47 0.61 1.15

Qi 1118 607 394

Page 11: Newsvendor Model - The University of Texas at Dallasmetin/Or6366/Folios/Inventory/scnewsvendor.pdf · Newsvendor Model . utdallas.edu/~metin 2 ... 6 months: Seasonal Camera, ... production

utdallas.edu/~metin

11

Algorithm for Ordering

Under Capacity Constraints {Initialization}

ForAll products, Qi := 0. Remaining_capacity:=Total_capacity.

{Iterative step}

While Remaining_capacity > 0 do

ForAll products,

Compute the marginal contribution of increasing Qi by 1

If all marginal contributions <=0, STOP

{Order sizes are already sufficiently large for all products}

else Find the product with the largest marginal contribution, call it j

{Priority given to the most profitable product}

Qj := Qj+1 and Remaining_capacity=Remaining_capacity-1

{Order more of the most profitable product}

Page 12: Newsvendor Model - The University of Texas at Dallasmetin/Or6366/Folios/Inventory/scnewsvendor.pdf · Newsvendor Model . utdallas.edu/~metin 2 ... 6 months: Seasonal Camera, ... production

utdallas.edu/~metin 12

Marginal Contribution=(p-c)P(D>Q)-(c-s)P(D<Q)

Order Quantity Marginal Contribution

Remaining_Capacity Autumn Leaves Ruffle Autumn Leaves Ruffle

1500 0 0 0 74.997 109.679 136.360

1490 0 0 10 74.997 109.679 135.611

1360 0 0 140 74.997 109.679 109.691

1350 0 0 150 74.997 109.679 106.103

1340 0 10 150 74.997 109.617 106.103

1330 0 20 150 74.997 109.543 106.103

1320 0 30 150 74.997 109.457 106.103

1310 0 40 150 74.997 109.357 106.103

890 0 380 230 74.997 73.033 70.170

880 10 380 230 74.996 73.033 70.170

870 20 380 230 74.995 73.033 70.170

290 580 400 230 69.887 67.422 70.170

280 580 400 240 69.887 67.422 65.101

1 788 446 265 53.196 53.176 52.359

0 789 446 265 53.073 53.176 52.359

Page 13: Newsvendor Model - The University of Texas at Dallasmetin/Or6366/Folios/Inventory/scnewsvendor.pdf · Newsvendor Model . utdallas.edu/~metin 2 ... 6 months: Seasonal Camera, ... production

utdallas.edu/~metin 13

Summary

Newsvendor Model

Cost ≡ Profit

Space Constraint

Page 14: Newsvendor Model - The University of Texas at Dallasmetin/Or6366/Folios/Inventory/scnewsvendor.pdf · Newsvendor Model . utdallas.edu/~metin 2 ... 6 months: Seasonal Camera, ... production

utdallas.edu/~metin 14

Revisit Newsvendor Problem with Calculus

Total cost by ordering Q units:

C(Q) = overstocking cost + understocking cost

Q

u

Q

o dxxfQxCdxxfxQCQC )()()()()(0

0))(())(1()()(

uuouo CCCQFQFCQFCdQ

QdC

Marginal cost of raising Q* - Marginal cost of decreasing Q* = 0

uo

u

CC

CQDPQF

)()( **


Top Related