0% found this document useful (0 votes)
5 views21 pages

9 Inventory Control - Part III

The document discusses various inventory management models, including fixed-order quantity models with both positive lead time and random demand, emphasizing the importance of safety stock. It also presents a case study of HP's DeskJet product, analyzing inventory challenges and exploring options like postponement to reduce safety stock and improve efficiency. The document concludes with a comparison of safety stock levels between traditional practices and postponement strategies.

Uploaded by

davidleekaleung
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
5 views21 pages

9 Inventory Control - Part III

The document discusses various inventory management models, including fixed-order quantity models with both positive lead time and random demand, emphasizing the importance of safety stock. It also presents a case study of HP's DeskJet product, analyzing inventory challenges and exploring options like postponement to reduce safety stock and improve efficiency. The document concludes with a comparison of safety stock levels between traditional practices and postponement strategies.

Uploaded by

davidleekaleung
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

DOTE 2030

Operations Management

Inventory Control
Part III

Page 1
Various Models for Inventory Management

Discussion continued

Page 2
Notation Revisited

 D = Annual demand
 d = D/365 = Daily demand
 L = Lead time

 C = Variable cost per unit


 S = Setup cost of a lot (i.e., a batch)
 H = Annual holding cost per unit of inventory
 H is often taken as a percentage of the cost of the item,
e.g., H = iC, where i is the percent carrying cost per year

Page 3
Fixed-Order Quantity Model with Positive Lead Time

 Demand for the product is constant and uniform throughout


the planning horizon

 Lead time (time from producing/ordering to receipt) is positive


Relaxed
 Variable cost per unit of product is constant
 for either production or order placement from supplier

 Inventory holding cost is based on average inventory

 All demands for the product will be satisfied

Page 4
Fixed-Order Quantity Model with Positive Lead Time
Now assume the lead time is positive

Lead time
Place an order Receive an order

1. When to order: R = demand in lead time = dL


2. How much to order: Q = Qopt, the same as before
Page 5
Example

 Consider inventory control of a product:


 Weekly demand is 20 units
 Lead time is zero (as assumed in last lecture)
 Variable cost per unit is $4
 Cost of placing an order is $20
 25% annual interest rate for money tied in inventory

 Now, what if Lead time L = 14 days?

Page 6
Fixed-Order Quantity Model with Random Demand and
Positive Lead Time

 Demand for the product is randomly distributed throughout


the planning horizon Relaxed

 Lead time (time from producing/ordering to receipt) is positive


Relaxed
 Variable cost per unit of product is constant
 for either production or order placement from supplier

 Inventory holding cost is based on average inventory

 All demands for the product will be satisfied

Page 7
Fixed-Order Quantity Model with Random Demand and
Positive Lead Time

Now assume the demand is random

1. When to order: R = Average demand in lead time + Safety stock


2. How much to order: Q = Qopt, the same EOQ as before
Page 8
Establishing Safety Stock Level

 Safety stock is the amount of inventory carried in


addition to average demand in lead time

 Safety stock can be determined based on many


different criteria
 A common approach is to simply keep a certain
number of days/weeks/months of supply
 A better approach is to use probability (such as a
service level of 80% for not running out of stock)

 Here we follow the approach of using probability and


assume demand is normally distributed

Page 9
Fixed–Order Quantity Model with Safety Stock

Reorder point R = Average demand in lead time + Safety stock


 R  dL  z L
where
R = Reorder point
d = Average daily demand
L = Lead time in days
z = Number of standard derivations for a specified service level
L = Standard deviation of demand in lead time

zL = Safety stock

Page 10
Another Example

 Find the economic order quantity and reorder point, given


(note that data is different from previous example)
 Daily demand being Normally distributed with mean
d = 60 and standard deviation d = 7
 Lead time = 6 days
 Cost of placing an order S = $10 per order
 Holding cost H = $0.5 per unit per year
 Service level being 95% of not stocking out during the
lead time

Page 11
Inventory Management at Item Level:
ABC Classification
ABC classification is a way to determine what items in
inventory are important and thus deserves close scrutiny

Type % of items % of value Inventory mgmt effort


A ~ 20% ~ 70% Close scrutiny
B ~ 30% ~ 25% Some effort
C ~ 50% ~ 5% Minimal effort
Page 12
HP DeskJet Case: Background

 DeskJet is HP’s most successful product

 The manufacturing of DeskJet was located in HP’s facility


in Vancouver, where different models were completed
before delivering to DCs (distribution centers)

Supplier IC Mfg US DC Customer

European
Supplier PCAT FAT Customer
DC

Print Far East Customer


Supplier DC
Mech
Supplier
Page 13
HP DeskJet Case: Analysis

 Problems in European DC
 High inventories
 Inventory imbalance

 Causes
 Uncertain market
 Long lead times
 Many geographical options (localization)
…

 What are HP’s options?

Page 14
HP DeskJet Case: Options

 Short-term options
Focus of
 Rationalize safety stock by postponement this case

 Long-term options
 Better forecasting
 More inventory
 Air shipment
 European factory
 Uniform product for all markets
…

Page 15
HP DeskJet Case: Postponement
Various Markets:
Vancouver European
Mfg Facility DC UK
France
Traditional Italy
Practice .
Complete Complete .
products products .
manufactured stocked

Various Markets:
Vancouver European
Mfg Facility DC UK
France
Italy
Postponement .
Generic Generic .
products products .
manufactured stocked Localization:
Power supply, manual
Page 16
Localization: Power Supply

UK

France

Italy

Page 17
HP DeskJet Case: Evaluating Postponement

 With postponement and thus localization at DCs,


inventories are pooled, which presumably can reduce
safety stocks while maintaining service levels

 To evaluate the impact, compare total safety stock if


complete products are held in inventory (as in traditional
practice) or if generic products are held (as with
postponement)

Page 18
HP DeskJet Case: Calculation of Safety Stock

 Recall the calculation of safety stock


Safety stock = zL

 Use printer A in the traditional practice as an example


 Service level of 98% gives z = NORMSINV(0.98) = 2.05
 Monthly sample data of the product yields
Monthly average = 42, monthly standard deviation = 32
 Lead time = 5 weeks = 35 days

 Safety stock of printer A in the traditional practice is


Safety stock = 2.05 × 32 × √ (35/30) = 72 units

Page 19
HP DeskJet Case: Comparison of Safety Stock

Printer Monthly Ave Monthly Std Safety stock


A 42 32 72
Traditional AB 15830 5625 12477 Safety stock
Practice AU 4208 2205 4890 in total
= 20713
AA 420 204 452
AQ 2301 1168 2592
AY 307 103 229

Printer Monthly Ave Monthly Std Safety stock


Postponement Generic 23108 6244 13851

Page 20
HP DeskJet Case: Impacts of Postponement

 Safety stock reduction


 Traditional practice: safety stock = 20,712 units
 Postponement: safety stock = 13,851 units

 Other benefits
 Less capital tied with inventory
 Ease of freight and logistics
 Local procurement of localization materials
 Local presence of “manufacturing”
 Customs implications

 But there are costs


 DC modifications to accommodate light manufacturing

Page 21

You might also like