1/1/25

Inventory Management - 2
Continuous review, periodic review and single period
models

EOQ Models with Constant Lead Time

and Uncertain Demand
Inventory

Safety stock

time
• There are chances of stocking-out. ROP should be adjusted upwards to account for this variability
• When to order?
• At ROP
• ROP = average demand during Lead Time + safety stock
• Safety stock = Std dev of demand during lead time * z-value = sigma during LT * Norm.S. Inv (Service level)

• How much to order? Ans = EOQ

1
 1/1/25

EOQ Models with Constant LT and

uncertain demand
𝑠𝑒𝑟𝑣𝑖𝑐𝑒 𝑙𝑒𝑣𝑒𝑙(𝛼) = 95%

Suppose Std deviation of lead time

demand = 21.3 units

1.65 ∗ 𝜎!

𝑀𝑒𝑎𝑛 𝑑𝑒𝑚𝑎𝑛𝑑 𝑖𝑛 𝑙𝑒𝑎𝑑 𝑡𝑖𝑚𝑒

𝜇! = 150 186 𝑢𝑛𝑖𝑡𝑠

ROP = Mean demand during Lead Time + safety stock

𝑅𝑂𝑃 = 𝜇7 + 𝑍 ∗ 𝜎7 = 150 + 1.65 ∗ 21.3 = 186 𝑢𝑛𝑖𝑡𝑠

EOQ Models with Constant Lead Time

and uncertain demand
• When Demand is uncertain, and Lead Time is constant
• Possibility of stock-outs
• Need safety stock – to prevent stock-outs

• Choosing safety stock

• Quantity of safety stock
• depends on service level
• Service level
• the desired probability of not running out of stock in any cycle

2
 1/1/25

EOQ Models with Constant LT and

uncertain demand
• Computing safety stock when demand follows normal distribution

• Step 1: Determine the mean and std deviation of demand during Lead time (L)

𝜎! = 2 ∗ 15 = 21.3
𝜎 = 15 𝜎 = 15

𝜇 = 75 𝜇 = 75 𝜇! = 2 ∗ 75 = 150
Demand for day 1 Demand for day 2 Demand for Lead time of
two days
𝑀𝑒𝑎𝑛 𝑑𝑒𝑚𝑎𝑛𝑑 𝑑𝑢𝑟𝑖𝑛𝑔 𝐿 = 𝜇! = 𝐿 ∗ 𝜇

𝑆𝑡𝑑. 𝐷𝑒𝑣𝑖𝑎𝑡𝑖𝑜𝑛 𝑜𝑓 𝑑𝑒𝑚𝑎𝑛𝑑 𝑑𝑢𝑟𝑖𝑛𝑔 𝐿 = 𝜎! = √(𝐿 ∗ 𝜎 " )

EOQ Models with Constant LT and

uncertain demand
• Step 2 : Choose the service level, say 𝛼 = 95% and compute 𝑍#

𝑍# = 𝑛𝑜𝑟𝑚. 𝑠. 𝑖𝑛𝑣 𝛼

• Step 3 : Compute SS
𝑆𝑎𝑓𝑒𝑡𝑦 𝑠𝑡𝑜𝑐𝑘 = 𝜎! 𝑍#

• Step 4 : Compute ROP 𝛼 = 95%

𝜎! 𝑍#
𝑅𝑂𝑃 = 𝜇8 + 𝜎8 𝑍9
𝜇!

3
 1/1/25

Inventory control system -

Continuous review (Q) system
• A system designed to track the remaining inventory each time a withdrawal is made

• Upon reaching ROP, place an order for Q units

• Level of inventory is measured in terms of inventory position (IP)

• IP = On-hand Inventory + Scheduled Receipts – Back orders

• Order Quantity is same in each cycle =EOQ

• Time between orders (TBO) varies

• Assumption : The daily demand distribution are identical and independent of each other

Designing Continuous Review System (Q-

System) – inventory position vs. on-hand
inventory
Inventory Position
Q Physical Inventory
Inventory Level

ROP

Mean Demand during LT

SS
Safety Stock

L Time

4
 1/1/25

Designing Continuous Review

System (Q-System)
Criterion Q-system (constant demand, constant lead time)
Mean demand during lead time (L) 𝜇! = 𝐿 ∗ 𝑎𝑣𝑔. 𝑑𝑎𝑖𝑙𝑦 𝑑𝑒𝑚𝑎𝑛𝑑

How much to order? Fixed Quantity Q, every cycle

Determined using EOQ model

When to order ? ROP = 𝜇!

Safety stock? No

Time between Orders? No. of orders annually = Annual Demand/EOQ

Time between orders = 1/No. of orders

Annual ordering cost (D/Q) * Co

Annual Carrying cost (Q/2 )* Cc

Annual Material cost D * Cu

Total cost Annual ordering cost + Annual carrying cost + Annual material cost

Designing Continuous Review

System (Q-System)
Criterion Q-system (uncertain demand, constant lead time)
Mean and Std. deviation of demand distribution during lead time (L) Mean demand during LT = 𝜇! = 𝐿 ∗ 𝑎𝑣𝑔. 𝑑𝑎𝑖𝑙𝑦 𝑑𝑒𝑚𝑎𝑛𝑑 ,
SD of demand during LT = 𝜎! = 𝑆𝑡𝑑. 𝑑𝑒𝑣 𝑜𝑓 𝑑𝑎𝑖𝑙𝑦 𝑑𝑒𝑚𝑎𝑛𝑑 ∗ √𝐿

How much to order? Fixed Quantity Q, every cycle

Determined using EOQ model

When to order ? ROP = 𝜇! + 𝑆𝑎𝑓𝑒𝑡𝑦 𝑠𝑡𝑜𝑐𝑘

Safety stock? Service level = 𝛼

𝑆𝑎𝑓𝑒𝑡𝑦 𝑠𝑡𝑜𝑐𝑘 = 𝑧 − 𝑣𝑎𝑙𝑢𝑒 ∗ 𝜎! = 𝑁𝑜𝑟𝑚. 𝑠. 𝑖𝑛𝑣 𝛼 ∗ 𝜎!

Time between Orders? No. of orders annually = Annual Demand/EOQ

Time between orders = 1/No. of orders

Annual ordering cost (D/Q) * Co

Annual Carrying cost (Q/2 + safety stock) Cc

Annual Material cost D * Cu

Total cost Annual ordering cost + Annual carrying cost + Annual material cost

10

5
 1/1/25

Practice problem - 1
Daily demand for a certain product is normally distributed, with a mean of 60 and a standard deviation of
7. The source of supply is reliable and maintains a constant lead time of six days. The cost of placing the
order is $10 and annual holding costs are $0.50 per unit. It costs $20 to purchase the item. There are no
stock-out costs, and unfilled orders are filled as soon as the order arrives. Assume sales occur over the
entire 365 days of the year. Find the order quantity and reorder point to satisfy a 95 percent probability
of not stocking out during the lead time.

11

Practice problem - 1
• Annual Demand (D) = 60 * 365 = 21, 900

"$! % " ∗()∗"(,+))

• Optimal Q = = = 935. 95 units
$" ).-

• Mean demand during lead time = 𝜎 ∗ 𝐿 = 60 ∗ 6 = 360

• SD of demand during lead time = 𝜎 " ∗ 𝐿 = 7" ∗ 6 = 17.146 𝑢𝑛𝑖𝑡𝑠

• Z-value for 95% service level = 1.644

• Safety stock = 𝑧 − 𝑣𝑎𝑙𝑢𝑒 ∗ 𝜎! = 1.644 ∗ 17. 146 = 28.2 𝑢𝑛𝑖𝑡𝑠

• ROP = mean demand during LT + safety stock = 388.2 units

12

6
 1/1/25

Practice problem - 1
;! < =
• Total cost = =
+ >
+ 𝑠𝑎𝑓𝑒𝑡𝑦 𝑠𝑡𝑜𝑐𝑘 𝐶? + 𝐶@ 𝐷

AB∗>ACBB [Link]
• TC = [Link]
+ >
+ 28.203 0.5 + 20 ∗ 21900 = $4,38,483

13

Practice problem -2
Gentle Ben’s Bar and Restaurant uses 5,000 quart bottles of an imported wine each year. The
effervescent wine costs $3 per bottle and is served only in whole bottles because it loses its bubbles
quickly. Ben figures that it costs $10 each time an order is placed, and holding costs are 20 percent of the
purchase price. It takes three weeks for an order to arrive. Weekly demand is 100 bottles (closed two
weeks per year) with a standard deviation of 30 bottles. Ben would like to use an inventory system that
minimizes inventory cost and will provide a 95 percent service probability. a. What is the economic
quantity for Ben to order? b. At what inventory level should he place an order?

14

7
 1/1/25

Practice problem-3
A local service station is open 7 days per week, 365 days per year. Sales of 10W40 grade premium oil
average 20 cans per day. Inventory holding costs are $0.50 per can per year. Ordering costs are $10 per
order. Lead time is two weeks. Backorders are not practical— the motorist drives away.

a. Based on these data, choose the appropriate inventory model and calculate the economic order
quantity and reorder point. Describe in a sentence how the plan would work. Hint: Assume demand is
deterministic.

b. The boss is concerned about this model because demand really varies. The standard deviation of
demand was determined from a data sample to be 6.15 cans per day. The manager wants a 99.5 percent
service probability. Determine a new inventory plan based on this information and the data in part (a).
Use Qopt from part (a).

15

Practice problem - 4
Retailers Warehouse (RW) is an independent supplier of household items to department stores. RW
attempts to stock enough items for a 98 percent service probability. A stainless steel knife set is one item
it stocks. Demand (2,400 sets per year) is relatively stable over the entire year. Whenever new stock is
ordered, a buyer must ensure that numbers are correct for stock on-hand and then phone in a new order.
The total cost involved to place an order is about $5. RW figures that holding inventory in stock and
paying for interest on borrowed capital, insurance, and so on, add up to about $4 holding cost per unit
per year. Analysis of the past data shows that the standard deviation of demand from retailers is about
four units per day for a 365-day year. Lead time to get the order is seven days.

a. What is the economic order quantity?

b. What is the reorder point?

16

8
 1/1/25

Questions of interest for managers

Which items should be How much to order ? When to order? What is the right level of safety
monitored periodically? stock?
(i.e, controlled with “slightly
less” intensity)

17

Periodic Review System (P-system)

• An alternate inventory control system in which an item’s
inventory position is reviewed periodically rather then
continuously

• At the time of review, place an order to replenish to a

predetermined level

18

9
 1/1/25

Designing Periodic Review System (P-

System)
Inventory Position
Physical Inventory

Order Up to Level
S
Inventory Level

P 2P 3P
L

Time

19

Designing Periodic Review System (P-

System)
Inventory Position
Physical Inventory

QP Q2P Q3P
Order Up to Level
S
Inventory Level

SS

Safety Stock

P 2P 3P
L

Time

20

10
 1/1/25

Designing Periodic Review System

(P-System)
Criterion P-system (uncertain demand, constant lead time)

Demand distribution during lead time (L) and review period (P) 𝜇!"# = 𝐴𝑣𝑔. 𝑑𝑎𝑖𝑙𝑦 𝑑𝑒𝑚𝑎𝑛𝑑 ∗ (𝐿 + 𝑃) ,
𝜎!"# = 𝑆𝑡𝑑 𝐷𝑒𝑣 𝑜𝑓 𝑑𝑎𝑖𝑙𝑦 𝑑𝑒𝑚𝑎𝑛𝑑 ∗ √(𝐿 + 𝑃)

How much to order? Order Quantity (Q) varies every cycle

Q = S – on-hand Inventory (OHI)

When to order ? Every P period

Order upto level? Order upto level, 𝑆 = 𝜇!"# + 𝑠𝑎𝑓𝑒𝑡𝑦 𝑠𝑡𝑜𝑐𝑘

Safety stock? Service level = 𝛼

𝑆𝑎𝑓𝑒𝑡𝑦 𝑠𝑡𝑜𝑐𝑘 = 𝑍 − 𝑣𝑎𝑙𝑢𝑒 ∗ 𝜎!"# = 𝑁𝑜𝑟𝑚. 𝑠. 𝑖𝑛𝑣 𝛼 ∗ 𝜎!"#

Annual Ordering cost Co * Annual Demand / (Avg daily demand* P)

Annual Carrying cost Cc * (Avg. daily demand*P/2 + safety stock)

Annual materials cost Cu * Annual demand

Total cost Annual ordering cost + Annual carrying cost + Annual Material cost

21

Periodic Review System (P-system)

• Average demand during (P+L) periods
• 𝜇TU7 = 𝐴𝑣𝑒𝑟𝑎𝑔𝑒 𝑑𝑎𝑖𝑙𝑦 𝑑𝑒𝑚𝑎𝑛𝑑 ∗ 𝑃 + 𝐿

• Safety stock for (P+L) periods

• SS = 𝑍V 𝜎TU7 = 𝜇TU7 + 𝜎TU7 ΦWX(𝛼)

• Safety stock for P-system > Safety stock for Q-system for
a given service level

22

11
 1/1/25

Practice problems- 1
• Daily demand for a product is 10 units, with a standard deviation of 3 units. The review period is 30 days,
and the lead time is 14 days. Management has set a policy of satisfying 98 percent of demand from items in
stock. At the beginning of this review period, there are 150 units in inventory. The cost of placing the order is
$10 and annual holding costs are $0.50 per unit. It costs $20 to purchase the item. How many units should
be ordered?

23

Practice problem - 1
• Annual demand (D) = 10*365 = 3650 units

• Mean demand during (P+L) time = 𝜇 𝑃 + 𝐿 = 10 ∗ 44 = 440

• SD of demand during lead time = 𝜎 Y(𝑃 + 𝐿) = 3Y ∗ 44 = 19.89 𝑢𝑛𝑖𝑡𝑠

• Z-value for 98% service level = 2.053

• Safety stock = 𝑧 − 𝑣𝑎𝑙𝑢𝑒 ∗ 𝜎7 = 2.053 ∗ 19.89 = 40.86 𝑢𝑛𝑖𝑡𝑠

• Order upto level = 440+40.86= 480.86 units

24

12
 1/1/25

Practice problem -1
;! < G∗HIJ KHLMN KOPHQK
• Total cost = G∗HIJ KHLMN KOPHQK
+; >
+
𝑠𝑎𝑓𝑒𝑡𝑦 𝑠𝑡𝑜𝑐𝑘 < 𝐶? + 𝐶@ 𝐷

AB∗DREB DB∗AB
• TC = + + 40.869 0.5 + 20 ∗ 3650 = $73,218
DB∗AB >

25

Practice problems -2
• Daily demand for a product is 60 units with a standard deviation of 10 units. The review period is 10 days,
and lead time is 2 days. At the time of review, there are 100 units in stock. . The cost of placing the order is
$1.5 and annual holding costs are $5 per unit. It costs $10 to purchase the item. If 98 percent service
probability is desired, how many units should be ordered?

26

13
 1/1/25

Practice problems - 3
• University Drug Pharmaceuticals orders its antibiotics every two weeks (14 days) when a salesperson visits
from one of the pharmaceutical companies. Tetracycline is one of its most prescribed antibiotics, with an
average daily demand of 2,000 capsules. The standard deviation of daily demand was derived from
examining prescriptions filled over the past three months and was found to be 800 capsules. It takes five
days for the order to arrive. University Drug would like to satisfy 99 percent of the prescriptions. The
salesperson just arrived, and there are currently 25,000 capsules in stock. How many capsules should be
ordered?

27

Practice problem - 4
• UA Hamburger Hamlet (UAHH) places a daily order for its high-volume items (hamburger patties, buns, milk,
and so on). UAHH counts its current inventory on-hand once per day and phones in its order for delivery 24
hours later. Determine the number of hamburgers UAHH should order for the following conditions:

28

14
 1/1/25

Comparison of Q and P systems

• Continuous review system • Periodic review system
• Requires perpetual • Periodic monitoring
monitoring • Responsive to changes to
• Less responsive to changes demand
to demand • Order for multiple items can
• Not suitable for ordering be combined
multiple items • High level of safety stock
• Low level of safety stock

29

Stock-up vs. Stock-out?

Source: [Link]

30

15
 1/1/25

Special case

Morning 5 am: Evening 9 pm:

Stocks newspaper early in the morning Disposes unsold newspaper

Railway platform

Question faced by newsvendor : How many units of newspaper to stock?

31

Single-period Inventory model

• Number of situations in which what is ordered in one period can’t be used for the next
period

• Example :
• Perishable items (Mysuru Pak, Ras Malai, Gulab Jamoon)
• T-shirts promoting events
• Daily Newspapers
• 2024 new year greeting cards

• Question for the retailer

• How many units to order?

• Intuitive answer = Exactly equal to demand

• Challenge = Exact demand is unknown. It can vary.

32

16
 1/1/25

Single-period/ Single-order inventory

models
Inventory

Period = 1 Situation 1 : under-stock Situation 2 : over-stock

Vs. - High unused inventory
- Low service level
- Cost of understocking - Cost of overstocking
Q units

Overstock = Q-D

Understock= D-Q
time

Min Total cost = cost of overstocking + cost of understocking

33

Single-period/ Single-order inventory

models

Understocking scenario
Loss for every unit demand missed = Rs 5 = Cu
(opportunity cost)

Overstocking scenario
Loss for every unit unsold = Rs 3 = Co

Question faced by newsvendor : How many units of newspaper to stock?

34

17
 1/1/25

Single period models – Optimal order

quantity

Chance of overstocking
Chances of OS/ US *

Chance of understocking

Order quantity, Q
*Assuming Cu= Co

35

Single period models – Optimal order

quantity
Marginal analysis of increasing order from Q to Q +1:

Q*
Marginal gain(cost of US)from adding one more unit = Cu * P(D>Q)
Expected gain/loss
from an extra unit

Marginal loss(cost of OS) from adding one more unit = Co * P(D<= Q)

understock

Optimal Newsvendor Quantity:

overstock
𝐶.
𝐹 𝑄∗ =
(𝐶.+𝐶/)

Order quantity, Q

36

18
 1/1/25

Single-period inventory models

• Retailer can estimate the average demand, variability and demand distribution.
• Demand is normally distributed
• Average demand = 100, SD = 20

• He also has the cost information

• Selling price (v) = 10 Rs/unit
• Cost price (c) = 5 Rs/ unit
• Salvage (s) = Rs 2/ unit

• Tradeoff
• Cost of stocking one excess unit = c – s = Rs. 3/unit [cost of overstocking, Co]
• Cost of stocking one less unit = v – c = Rs 5 / unit [cost of understocking, Cu]

• Question : should he understock, overstock or stock equal to average demand?

• Hint : Observe the costs

37

Single period inventory models

• Question : Should retailer understock, overstock or stock
equal to average demand?

• Solution : Choose a stocking quantity that satisfies critical ratio:

$G
𝐹 𝑄∗ = ($
G1$H)

i.e., Order quantity (Q*)= mean demand + sd of demand * z-value associated with the critical ratio

• In the example, when Cu= 5, Co=3, we obtain a critical ratio of 5/

(3+5) = . Therefore, Q* = 106. 4 = 107 units

38

19
 1/1/25

Single period inventory models

• Thumb rule for Managers:
• High cost of lost sale, overstock
• High cost of having excess, understock

• Insight on the effect of variability

39

Practice problems -1
You are a newsvendor selling New York Times every morning. Before you get to work, you go to the distributor
and buy the day’s paper for $0.25 a copy. You sell a copy of the Times for $1.00. Daily demand is distributed
normally with mean = 250 and standard deviation = 50. At the end of each day, any leftover copies are
worthless, and they go to a recycle bin. a. How many copies of the New York Times should you buy each
morning? b. Based on part (a), what is the probability that you will run out of stock?

40

20
 1/1/25

Practice problems -1
• Cost of understocking, Cu = 1-0.25= $0.75

• Cost of overstocking, Co = $0.25

• Demand : mean = 250, sd = 50

• Critical ratio, F(Q)= 1/1.25 =0.8

• Optimal Q = mean demand + sd * z-value = mean demand + sd* [Link](0.8) = 284 units

41

Practice problems - 2
The local supermarket buys lettuce each day to ensure really fresh produce. Each morning, any lettuce that is
left from the previous day is sold to a dealer that resells it to farmers who use it to feed their animals. This
week, the supermarket can buy fresh lettuce for $4.00 a box. The lettuce is sold for $10.00 a box and the dealer
that sells old lettuce is willing to pay $1.50 a box. Past history says that tomorrow’s demand for lettuce
averages 250 boxes with a standard deviation of 34 boxes. How many boxes of lettuce should the supermarket
purchase tomorrow?

42

21
 1/1/25

Practice problems - 3
Wholemark is an Internet order business that sells one popular New Year’s greeting card once a year. The cost
of the paper on which the card is printed is $0.05 per card, and the cost of printing is $0.15 per card. The
company receives $2.15 per card sold. Because the cards have the current year printed on them, unsold cards
have no salvage value. Its customers are from the four areas: Los Angeles, Santa Monica, Hollywood, and
Pasadena. Based on past data, the number of customers from each of the four regions is normally distributed
with a mean of 2,000 and a standard deviation of 500. (Assume these four are independent.) What is the
optimal production quantity for the card?

43

Practice problem -4
Famous Albert prides himself on being the Cookie King of the West. Small, freshly
baked cookies are the specialty of his shop. Famous Albert has asked for help to
determine the number of cookies he should make each day. From an analysis of
past demand, he estimates demand for cookies as provided in the table:

Each dozen sells for $0.69 and costs $0.49, which includes handling and
transportation. Cookies that are not sold at the end of the day are reduced to
$0.29 and sold the following day as day-old merchandise.

a. Construct a table showing the profits or losses for each possible quantity.

b. What is the optimal number of cookies to make?

44

22
 1/1/25

References
• Textbook readings
• Continuous and Periodic review systems: Chapter 12 (Pages 530-541, 553-554)
• Single period models : Chapter 12 (Pages 551-552)

• Additional readings on continuous and periodic review systems

• Refer Chapter 9, Pages 393-405 in ’Operations Management : Process and Supply Chains’ by
Krajeweski et al. (12th edition) (see LMS)

• Cases
• Stock-up Vs Stock-out: The Inventory Management Dilemma At A Mobile Clinic

45

23