BUSS211

OM

Lecture 3
Matching Supply With
Demand
Professor Kihoon Kim

Supply Chain Management

Slide 1

Contents
Littles Law (Flow Time vs. Inventory Amount)

Matching Supply with Known Demand

Steamed Oyster-Rice Game
Matching Supply with Unknown Demand

Postponement & Bullwhip Effect

Supply Chain Management

Slide 3

Costs of not Matching Supply and Demand

Cost of overstocking
liquidation, obsolescence, holding

Cost of under-stocking
lost sales and resulting lost margin

What are the causes (challenges) driving this mismatch?

Uncertainty
Timing
Number of SKUs

Supply Chain Management

Slide 4

The Grocery Industry 1985-1992

Number of products in average supermarket
1985
11,036
1990
16,486
1992
20,000
2004
??
2008
??

20,000

2,000
1975
Supply Chain Management

1992
Slide 5

2008

Even Toothpaste

Supply Chain Management

Slide 6

A Key to Matching Supply and Demand

When would you rather place your bet?

A
A:
B:
C:
D:
Supply Chain Management

C D

A month before start of Derby (horse racing)

The Monday before start of Derby
The morning of start of Derby
The winner is an inch from the finish line
Slide 7

Timing: Flow Time

Long Flow Time makes it hard to match supply with demand.

Supply Chain Management

Buffer

Operation

Waiting

Processing
Slide 8

Operational Flows (Littles Law)

Throughput R
Inventory I
FLOW TIME T

I=RT
Flow time T = Inventory I / Throughput R

Supply Chain Management

Slide 9

Why do Buffers Build?

Why hold Inventory?
Economies of scale
Fixed costs associated with batches
Quantity discounts

Cycle/Batch stock

Trade Promotions

Uncertainty
Information Uncertainty

Safety stock

Supply/demand uncertainty

Seasonal Variability

Seasonal stock

Strategic

Strategic stock

Flooding, availability
Supply Chain Management

Slide 10

Matching Supply with Known Demand

How to find the right Cycle/Batch stock quantity?

Supply Chain Management

Slide 11

What Characterizes Inventory Systems?

1.

Demand

2.

3.
4.

Known Vs. Uncertain (Random)

Constant Vs. Variable (Aggregate planning, MRP): the rate of demand

Lead Time: time that elapses from placement of order until its arrival.
During the lead time, a shortage can happen.
Review Cycle: Continuous vs. Periodic
Treatment of Excess Demand.

5.

Backorder all Excess Demand

Lose all excess demand
Backorder some and lose some

Inventory that changes over time

Supply Chain Management

perishability
obsolescence

Slide 12

Cost of Holding Inventory

Ex:

Physical holding cost

a) Physical Cost of Space (ex:3%)

b) Taxes and Insurance (ex: 2 %)
c) Breakage Spoilage and
Deterioration (ex: 1%)
d) Opportunity Cost of
alternative investment: cost of
capital (ex: 18%)
Total: 24% (of unit price)

(out-of-pocket)
Financial holding cost

: Max{Cost of capital r (%/yr),

Annual opportunity cost }

Low responsiveness
(hard to measure)
to demand/market changes
to supply/quality changes

Supply Chain Management

Slide 13

Order Cost
Fixed order cost (regardless

of order size)
Setup cost
Transportation cost

Order Cost

Temporary workers wages

Fixed
Cost

Variable order cost

: price of a good # of units

# of units

ordered

Supply Chain Management

Slide 14

Penalty (or Shortage) Cost

All costs that accrue when insufficient stock is available to meet

demand. These include:
Loss of revenue for lost demand
Costs of bookkeeping for backordered demands
Loss of goodwill for being unable to satisfy demands when they
occur.
Generally assume cost is proportional to number of units of excess
demand.
If the demand is known, shortage cost is generally irrelevant; Only
when the demand is random, we incorporate shortage cost in
calculating an optimal number of units to order.

Supply Chain Management

Slide 15

Batch Size (order size) vs. Inventory Level

Production with large batches

Cycle
Inventory

Production with small batches

Cycle
Inventory

Produce Sedan
Produce Station wagon

Beginning of
Month

Supply Chain Management

End of
Month

Slide 16

Beginning of
Month

End of
Month

Economic Order Quantity (EOQ) Model

Assumptions:

1. Demand is fixed at l units per unit time.

2. Shortages are not allowed.
3. Orders are received instantaneously. (this will be relaxed later).

4. Order quantity is fixed at Q per cycle. (can be proven optimal.)

5. Cost structure:
a) Fixed and Variable order costs (K + cx)

b) Holding cost: h per unit per unit time.

Supply Chain Management

Slide 17

Inventory Cycle
Inventory, units

downward
slope

T
Supply Chain Management

Time, t
Slide 18

Total Inventory Cost

TC (Q)

Kl
Q

hQ
2

Fixed Order Cost

cl
Variable Order Cost

Holding Cos t

where
K: setup cost per order
c: order cost per unit

l: demand rate per unit time (# of units)

h: holding cost per unit per unit time

Supply Chain Management

Slide 19

Economic Order Quantity

K
h
Q
c
i

Demand per year,

Setup or Order Cost ($/setup; $/order),

Marginal annual holding cost ($/per unit per year),

Order quantity.

Cost per unit ($/unit),

Cost of capital (%/yr),

h = i c.

Total annual
costs

2K l h
h Q/2:
Annual

The order quantity that minimizes total

supply chain cost is:

holding cost

2K l
Q*
h

K /Q:Annual
setup cost

EOQ

Order Size

Q
Supply Chain Management

Slide 20

Balancing between order cost and holding cost

Hyundai Sonata has been sold at a rate of 1000 cars/month

during the past year. Hyundai Sonata has released a new

version, and they expect it to be sold at the same rate. The new
Sonata costs $12,000. A dealer needs to determine how many
new Sonatas they are going to buy regularly to balance
between order-processing costs and inventory-holding
costs? You can use the information below:
Order (setup) cost: $1000 per order

Holding cost: 10% of the variable cost (annually)

Supply Chain Management

Slide 21

Calculating EOQ
Which information they need to calculate EOQ?
Order cost

Fixed order cost: $1,000

Variable cost per unit: $12,000

Holding cost: 10% of the variable cost (annually)

EOQ?
What will happen to EOQ if order cost quadruples?

Supply Chain Management

Slide 22

Role of Lead Time: Reorder Point (ROP)

An order can generally takes some time to be fulfilled by a 3rd

party.
Need to order in advance
When to order?

Reorder point is the inventory amount that indicates the time to order

If it takes 3 days for the order to arrive at the dealer shop, when the

dealer should order?

Supply Chain Management

3 days:= 0.1 month

0.1 month < 0.1414 (cycle time)
0.1x1000=100 (Reorder Point)

Slide 23

Pal Gear Case

Annual jacket revenues at a Pal Gear retail store are roughly $1M. Pal
jackets sell at an average retail price of $325, which represents a mark-up
of 30% above what Pal Gear paid its manufacturer. Being a profit center,
each store made its own inventory decisions and was supplied directly from
the manufacturer by truck. A shipment up to a full truck load, which was
about 1500 jackets, was charged a flat fee of $2,200. To exploit economies
of scale, stores typically ordered full truck loads. (Pals cost of capital is
approximately 20%.)
What order size would you recommend for a Pal store in current supply
network?

manufacturer

retailer

Supply Chain Management

Slide 24

Pal Gear: evaluation of current policy of ordering Q =

1500 units each time

1.

What is average inventory I?

2.

How often do we order?

3.

Annual throughput =
# of orders per year = Throughput / Batch size
Annual order cost = Order cost # of orders

What is total cost? (excluding annual variable order cost)

4.

I = Q/2
Annual cost to hold one unit H =
Annual cost to hold I = Holding cost Inventory

TC = Annual holding cost + Annual order cost =

What happens if order size changes?

Supply Chain Management

Slide 25

Find most economical order quantity:

Spreadsheet for a Pal Gear retailer
Number of
units
per
Number of
order/batch Batches per
Annual
Annual
Q
Year: R/Q Setup Cost Holding Cost
50
62
$135,385
$1,250
100
31
$67,692
$2,500
150
21
$45,128
$3,750
200
15
$33,846
$5,000
250
12
$27,077
$6,250
300
10
$22,564
$7,500
350
9
$19,341
$8,750
400
8
$16,923
$10,000
450
7
$15,043
$11,250
500
6
$13,538
$12,500
510
6
$13,273
$12,750
520
6
$13,018
$13,000
530
6
$12,772
$13,250
540
6
$12,536
$13,500
550
6
$12,308
$13,750
600
5
$11,282
$15,000
650
5
$10,414
$16,250
700
4
$9,670
$17,500
750
4
$9,026
$18,750
800
4
$8,462
$20,000
850
4
$7,964
$21,250
900
3
$7,521
$22,500
1000
3
$6,769
$25,000

Supply Chain Management

Annual
Total Cost
$136,635
$70,192
$48,878
$38,846
$33,327
$30,064
$28,091
$26,923
$26,293
$26,038
$26,023
$26,018
$26,022
$26,036
$26,058
$26,282
$26,664
$27,170
$27,776
$28,462
$29,214
$30,021
$31,769

$160,000

Setup Cost

$140,000

Holding Cost

$120,000

Total Cost

$100,000
$80,000
$60,000
$40,000
$20,000
$0

100 200 300 400 500 600 700 800 900 1000
Order (batch) size Q

Slide 26

Optimal Economies of Scale:

For a Pal Gear retailer
= 3077 units/ year
i = 0.20/year

c = $ 250 / unit
K = $ 2,200 / order

Unit annual holding cost = h = 0.20/yr x $250 = $50/yr

Optimal order quantity = Q = sqrt(2 x 3077 x 2200/50) = 520
Number of orders per year = /Q = 5.9
Time between orders = Q/ = 0.17yr = 8.8weeks
Annual order cost = ( /Q)K = $13,008.87/yr
Average inventory I = Q/2 = 260
Annual holding cost = (Q/2)h =$13,008.87/yr
Average flow time T = I/R = 0.084 yr = 4.4weeks
If the lead time is 2 weeks, when they should order?
Supply Chain Management

Slide 27

Optimal Economies of Scale:

Managerial Insights

QEOQ

2K l

TCEOQ 2 K l h cl

How cut inventories (economically smart)? Reduce Fixed Order Cost

Budgeting for growth

Last FY:
Next year:

Sales = $100M
Sales = $400M

Centralized inventory management

Supply Chain Management

Slide 28

Inventories = $20M
Inventories = ?

Summary of Matching Supply with Known Demand

Increasing batch size Q of order (or production) increases average

inventories (and thus flow times).

Average inventory for a batch size of Q is Q/2.

The optimal batch size minimizes supply chain costs by trading off setup
cost and holding cost and is given by the EOQ formula.
To reduce batch size, one must reduce setup cost (time).
Economies of scale are manifested by the square-root relationship between
QEOQ and (, K):

If demand increases by a factor of 4, it is optimal to increase batch size by a factor of 2

and produce (order) twice as often.
To reduce batch size by a factor of 2, setup cost has to be reduced by a factor of 4.

Supply Chain Management

Slide 29

Matching Supply with Unknown Demand

How to find the right safety stock quantity?

Supply Chain Management

Slide 30

Oyster-Rice in a Hot stone pot

For it to be delicious, Oysters should be fresh

enough
Supply Chain Management

Slide 31

Time to make the Oyster-Rice

The demand for Oyster-Rice is

random but the distribution does
not vary from day to day.
For simplicity, you tell your staff

Distribution of Daily Demand

to make the same number of the

oyster-rice each day that you are
out of town.

14%

12%

10%

How many oyster rices should

they prepare?

Probability

To maximize the profits

8%

6%

4%

We will simulate the demand for

the oyster-rice by

2%

0%
3

10 11 12 13 14 15 16 17 18
Demand

The team(s) that earns the

maximum profit is the winner(s)!

Supply Chain Management

Slide 32

Summary data
Distribution of Daily Demand
14%
12%

Probability

10%
8%
6%
4%
2%
0%
3

10

11

12

13

14

15

16

17

18

Demand

Retail price = $7

Cost = $3

Leftover price = $1

Mean demand = 10.5

Supply Chain Management

Slide 33

Demand Distribution Information

Supply Chain Management

P(Dem = D)

P(Dem D)

0.5%

0.5%

1.4%

1.9%

2.8%

4.6%

4.6%

9.3%

6.9%

16.2%

9.7%

25.9%

11.6%

37.5%

10

12.5%

50.0%

11

12.5%

62.5%

12

11.6%

74.1%

13

9.7%

83.8%

14

6.9%

90.7%

15

4.6%

95.4%

16

2.8%

98.1%

17

1.4%

99.5%

18

0.5%

100%

Slide 34

Suppose we prepare 11 pots of oyster-rice

We spend 11 $3 = 33 to prepare them.

Overstock: If demand is 8, we sell 8 at $7 for $56 and sell (11 8) at $1 for

$3.
Total profit is $26.

Understock: If demand is 13, we sell 11 at $7 for $77 and we have no

leftovers.
Total profit is $44.
Same profit as when demand is exactly 11.

Supply Chain Management

Slide 35

Safety Stocks

Inventory on hand
I(t)

Q
Q
order

order

order

ROP
R

mean demand during

supply lead time:

mL = R L
Is

safety stock s

Time t

L
Supply Chain Management

L
Slide 42

Safety Stocks Service Levels

Cycle Service
Level (CSL)

Stock-out
probability
F(z)

Is = z s
mean
0

ROP
z

demand during
supply lead time

Raise ROP until we reach appropriate SL

To do numbers, we need:
Mean m and stdev s of demand during lead time
Either Excel or tables with z-value such that CSL = F(z)

Supply Chain Management

Slide 43

1. How to find service level (given ROP)?

2. How to find re-order point (given SL)?
L

Lead time

Demand per unit time

Mean: R
Std. Dev.: sR

DL

Demand during lead time

Mean: mL
Std. Dev.: sL

mL = RL
sL = sR L

1.

Given ROP, find SL

= P(no stock out)

= P(demand during lead time < ROP)
= F(z*= (ROP- mL)/sL)
[use table]
= NORMDIST(ROP, mL, sL, True)
[or Excel]

2.

Given SL, find ROP

= mL + Is
= mL + z*sL
[use table to get z* ]
= NORMINV (SL, mL, sL) [or Excel]

Safety stock Is = z*sL ; Reorder point ROP = mL + Is

Supply Chain Management

Slide 44

Palu Gear:
Determining the required Safety Stock for 95% service
DATA:

sR = 30 jackets/ week

R = 59 jackets/ week
h = $50 / jacket-year
K = $ 2,200 / order

L = 2 weeks

QUESTION: What should safety stock be to insure a desired cycle service level of
95%?
ANSWER:
1. Required # of standard deviations z* for SL of 95% = 1.65
2. Determine std. dev. of demand during lead time: s L = sR L = 30 2 = 42
3. Answer: Safety stock
Supply Chain Management

Is = z* sL = 1.65 42 = 70
Slide 46

Total Inventory Costs of Pal Gear

1. Cycle Stock (Economies of Scale)
1.1 Optimal order quantity
1.2 # of orders/year
1.3 Annual ordering cost per store
1.4 Annual cycle stock holding cost.

= 520
= 5.9
= $13,009
= $13,009

2. Safety Stock (Uncertainty hedge)

2.1 Safety stock per store
2.2 Annual safety stock holding cost

= 70
= $3,500.

3. Total Costs

Supply Chain Management

= (13,009 + 13,009 + 3,500)

= $29,500
Slide 47

Summary of safety stocks

Safety stock is a hedge against uncertainty
Which factors drive safety stock ?
level of service z

Impact of increased service level on required safety stock

demand variability or forecast error sR,

delivery lead time L for the same level of service,
delivery lead time variability for the same level of service.

I s z sR L
*

Supply Chain Management

Slide 48

Matching supply with demand:

Summary
Implications:

Goal of a Supply Chain

Match Demand with Supply
It is hard Why?
Hard to anticipate demand,
Forecasts are wrong Why?

Economies of Scale

As long as you are in the flat

Manage the trade-off

between
ordering cost and inventory
holding cost. Use:

order quantity: QEOQ

2K l
h

Lead time (flow time) = Activity

time+ Waiting Time
Because there is waiting time...
Why there is waiting time?

scale, hence should reflect that

into ordering decisions.

To reduce the order size n times

Rules of Forecasting

Because there is inventory in the SC

(Littles Law)

Uncertainty
To hedge against forecast error,
increase inventories so that we keep

safety stock Is = zs R
Supply Chain Management

Growth brings economies of

one has to cut the fixed cost per

order by n2 times (the square root
formula!)

There is lead time. Why?

Why do we hold inventory?

region of total cost you are fine.

Slide 49

1. Forecasts are probability

distributions: Use at
least two numbers (mean
and s)
2. Shorter forecasts are
less uncertain
3. Aggregate forecasts are
less uncertain

Supply Chain Management Summary Contd:

How deal with Uncertainty?
Implications:

1. Is service level SL
(or z) appropriate?
2. Reduce lead time L

Balance overstocking and understocking

Newsvendor formula gives optimal service level:

SL* = Cu / (Co+Cu)
How do we deal with it?

3. Reduce uncertainty
per period sR
Customer Demand Uncertainty

Normal Variations

Where does sR come

from?

Supply Chain Management

Slide 50

1. Better forecasting
2. Pooling:
- physical centralization
- information
- specialization
- substitution
- commonality
3. Postponement (& Pooling)
4. Quick response:
- reduce leadtime and its
variability