CH 13 Supply Chain Management

Healthcare Operations
Management
An Integrated Approach to
Improving Quality and Efficiency
Chapter 13. Supply Chain

Management
Daniel B. McLaughlin
Julie M. Hays
Chapter 13
Supply Chain Management
• What is Supply Chain Management (SCM)?
• Why is SCM Important for Healthcare
Organizations?
• Tracking and Managing Inventory
• Forecasting
• Order Amount and Timing
• Inventory Systems
• Procurement and Vendor Relationship
Management
• Strategic SCM
Copyright 2008 Health Administration Press. All rights reserved. 13-2

Supply Chain Management (SCM)
• The management of all activities and
processes related to both upstream vendors
and downstream customers in the value
chain
• Tracking and managing demand, inventory,
and delivery
• Procurement and vendor relationship
management
• Technology enabled
SCM in Healthcare
• In 2006, the United States will spend over
$2 trillion on healthcare.
• Annual cost/family for health insurance is
forecasted to be $22,000 by 2010.
• By 2016, it is predicted that one dollar of
every five dollars of the U.S. economy will
be devoted to healthcare.

SCM in Healthcare
• Supply costs in hospitals account for 15–25
percent of operating costs (HFMA 2002;
HFMA 2005).
• Transaction costs are estimated at $150 per
order for buyer and seller (HFMA 2001).
• There is 35 percent inconsistency between
hospital and supplier data, and it costs $15
to $50 to research and correct a single order
discrepancy.
Inventory
• Inventory is the stock of items held to meet
future demand.
• Inventory management answers three
questions:
- How much to hold
- How much to order
- When to order

Functions of Inventory
• To meet anticipated demand
• To level process flow
• To protect against stockouts
• To take advantage of order cycles
• To help hedge against price increases or to
take advantage of quantity discounts
• To decouple process steps
Effective Inventory Management
• Classification system
• Inventory tracking system
• Reliable forecast of demand
• Knowledge of lead times
• Reasonable estimates of:
- Holding or carrying costs
- Ordering or setup costs
- Shortage or stockout costs

ABC Classification System
Classifying inventory according to some
measure of importance and allocating control
efforts accordingly
Pareto Principle High (80%)
-A very important Annual

-B moderately $ volume A
of items
important B
-C least important
Low (5%) C
Few Many
(20%) (50%)
Number of Items
Inventory Tracking
• Track additions and removals
- Bar-coding
- Point of use or point of sale (POS)
- RFID
• Physical count of items
- Periodic intervals
- Cycle count
- Find and correct errors

Forecasting
• Exercise
• Averaging methods
• Trend, seasonal, and cyclical models
• Model development and evaluation
• VVH example

Forecasting
Exercise I
• Identify the pattern and construct a formula
that will “predict” successive numbers in
the series.
• What is the next number in the series?
(a) 3.7, 3.7, 3.7, 3.7, 3.7, 3.7, 3.7, 3.7
(b) 2.5, 4.5, 6.5, 8.5, 10.5, 12.5, 14.5, 16.5
(c) 5.0, 7.5, 6.0, 4.5, 7.0, 9.5, 8.0, 6.5
• What is the formula for the next number in
the series?
Exercise I—Graphs
18
Series b
16
Series a
14
12
4.4
10
Series1
4.2 8
4.0 6
4
3.8
2
3.6 Series1
0
1 2 3 4 5 6 7 8
3.4
3.2
3.0
10.0
Series c
9.0
2.8
1 2 3 4 5 6 7 8 8.0
7.0
6.0
5.0 Series1
4.0
3.0
2.0
1.0
0.0
1 2 3 4 5 6 7 8

Exercise I Solution
a) 3.7, 3.7, 3.7, 3.7, 3.7, 3.7, 3.7, 3.7
- Constant
- Next number is 3.7
b) 2.5, 4.5, 6.5, 8.5, 10.5, 12.5, 14.5, 16.5

- 0.5 + 2x, where x specifies the position (index) of the number in the
series
- Next number is 18.5
c) 5.0, 7.5, 6.0, 4.5, 7.0, 9.5, 8.0, 6.5

- 4.5 + 0.5x + Cs, where x specifies the position (index) of the number in
the series
- Cs represents the seasonality factor
- C1 = 0, C2 = 2, C3 = 0, C4 = −2
- Next numbers: 9, 11.5, 10, 8.5

Exercise II
• Identify the pattern and construct a formula
that will “predict” successive numbers in the
series.
• What is the next number in the series?
(a) 4.1, 3.3, 4.0, 3.8, 3.9, 3.4, 3.5, 3.7
(b) 2.9, 4.7, 6.8, 8.2, 10.3, 12.7, 14.2, 16.3
(c) 5.3, 7.2, 6.4, 4.5, 6.8, 9.7, 8.2, 6.3

Exercise II Solution
• Same as series above, but with a random

component generated from normal random
number generator with mean 0
(a) 4.1, 3.3, 4.0, 3.8, 3.9, 3.4, 3.5, 3.7
• 3.7 + 
(b) 2.9, 4.7, 6.8, 8.2, 10.3, 12.7, 14.2, 16.3
• 0.5 + 2x + 
(c) 5.3, 7.2, 6.4, 4.5, 6.8, 9.7, 8.2, 6.3
• 4.5 + 0.5x + Cs + 

Forecasting Methods
• Qualitative methods
- Based on expert opinion
and intuition; often used
when there are no data available
• Quantitative methods
- Time series methods, causal methods

Demand Behavior
• Trend
- Gradual, long-term up or down movement
• Cycle
- Up and down movement repeating over long
time frame
• Seasonal pattern
- Periodic, repeating oscillation in demand
• Random movements follow no pattern

Forms of Forecast Movement
Trend Cycle
Demand
Demand
Random
movement
Time Time
Seasonal
pattern
Demand
Demand
Trend with
seasonal pattern
Time Time

Forecasting
Averaging Methods
• Simple moving average
• Weighted moving average
• Exponential smoothing
• Averaging methods all assume that the
dependent variable is relatively constant
over time; no trends or cycles

Simple Moving Average
Average over a given number of periods that

is updated by replacing the data in the oldest
period with that in the most recent period
D  D   Dt  n
F t
 t 1 t 2
n
Ft = Forecasted demand for the period
Dt-1 = Actual demand in period t − 1
n = Number of periods in the moving average

Weighted Moving Average
Simple moving average where weights are

assigned to each period in the average.
The sum of all the weights must equal one.
Ft w D t 1 t 1
 w t  2 Dt  2    w t  n Dt  n
Ft = Forecasted demand for the period

Dt-1 = Actual demand in period t − 1
wt-1 = Weight assigned to period t − 1

Exponential Smoothing
Times series forecasting technique that does not require large
amounts of historical data
F  1    F
t t 1
  Dt 1
Ft = Exponentially smoothed forecast for period t

Ft-1 = Exponentially smoothed forecast for prior period
Dt-1 = Actual demand in the prior period
 = Desired response rate, or smoothing constant

Forecasting
Trend, Seasonal, and Cyclical Models
• Holt’s trend-adjusted exponential smoothing

technique
• Winter’s triple exponential smoothed model
• ARIMA models

Holt’s Trend Adjusted
Exponential Smoothing
Exponentially smoothed forecast that accounts for a
trend in the data
FITt  Ft  Tt
and
Ft  αDt 1  ( 1  α)FITt 1
Tt  Tt 1  δ(Ft 1 -FITt 1 )
FITt = Forecast for period t including the trend

Ft = Smoothed forecast for period t
Tt = Smoothed trend for period t
Dt−1 = Value in the previous period
0  = smoothing constant 1; 0  = smoothing constant 1
Forecast Accuracy
• Error = Actual − Forecast
• Find a method that minimizes error
• Mean absolute deviation (MAD)
• Mean squared error

Forecasting
Model Development and Evaluation
• Identify purpose of forecast
• Determine time horizon of forecast
• Collect relevant data
• Plot data and identify pattern
• Select forecasting model(s)
• Make forecast
• Evaluate quality of forecast
• Adjust forecast and monitor results
VVH Diaper Example
Week of Period Actual
Weekly Demand
1-Jan 1 70
8-Jan 2 42 80
15-Jan 3 63 70
22-Jan 4 52 60
29-Jan 5 56 50
5-Feb 6 53 40
12-Feb 7 66 30
19-Feb 8 61 20
26-Feb 9 45 10
5-Mar 10 54 0
12-Mar 11 53 1 2 3 4 5 6 7 8 9 10 11 12 13
Period
19-Mar 12 43
26-Mar 13 60

VVH Simple Moving Average
D
 Dt  2   Dt  n
F t
t 1
D
 D12  D11  D10  D9
F 14
13
5
60  43  53  54  45
F 14  5
 51

VVH Weighted Moving Average
F t  w D  w D  w D
t 1 t 1 t 2 t 2 t n t n
F 14  w D  w D  w D
13 13 12 12 11 11
F 14  0.5  60  0.3  43  0.2  53  53.5

VVH Exponential Smoothing
F t
  Dt 1  1    F t 1
F 14
  D13  1    F 13 
F 14
 (0.25  60)  (0.75  52)  54

VVH Comparison
(from the Excel template)
Weight 3 Weight 2 Weight 1
Least Most
Periods 5 Recent 0.2 0.3 0.5 Recent α 0.25
MAD 7 MAD 6 MAD 8

MSE 86 MSE 75 MSE 135
Period Actual Forecast Error Period Actual Forecast Error Period Actual Forecast Error
1 70 1 70 1 70
2 42 2 42 2 42 70 28
3 63 3 63 3 63 63 0
4 52 4 52 58 6 4 52 63 11
5 56 5 56 53 3 5 56 60 4
6 53 57 4 6 53 56 3 6 53 59 6
7 66 53 13 7 66 54 12 7 66 58 8
8 61 58 3 8 61 60 1 8 61 60 1
9 45 58 13 9 45 61 16 9 45 60 15
10 54 56 2 10 54 54 0 10 54 56 2
11 53 56 3 11 53 53 0 11 53 56 3
12 43 56 13 12 43 52 9 12 43 55 12
13 60 51 9 13 60 48 12 13 60 52 8
14 51 14 53.5 14 54

Realities of Forecasting
• Forecasts are seldom perfect.
• Most forecasting methods assume that
there is some underlying stability in the
system.
• Service family and aggregatedI see
service
that you will
forecasts are more accurate get an A this semester.
than individual service
forecasts.

Order Amount and Timing
How much to hold
How much to order
When to order
• Basic economic order quantity (EOQ)

• Fixed order quantity with safety stock
• More models

Definitions
Lead time—time between placing an order and
receiving the order
Holding (or carrying) costs—costs associated with
keeping goods in storage
Ordering (or setup) costs—costs of ordering and
receiving goods
Shortage costs—costs of not having something in
inventory when it is needed
Back orders—unfilled orders
Stockouts—occur when the desired good is not
available
Definitions
Independent demand is demand that
is generated by the customer and is
not a result of demand for another
good or service.
Dependent demand is demand that

results from another demand.
Demand for tires and steering
wheels (dependent) is related to
the demand for cars (independent).
Assumptions of the
Basic EOQ Model
• Demand for the item in question is
independent.
• Demand is known and constant.
• Lead time is known and constant.
• Ordering costs are known and constant.
• Back orders, stockouts, and quantity
discounts are not allowed.

Inventory Order Cycle
Demand
Order rate
quantity, Q
Inventory Average
Level amount of
inventory
Reorder held = Q/2
point, R
0 Lead Lead
Time time time
Order Order Order Order
Placed Received Placed Received
Reorder Point
The point in time by which stock must be ordered
to replenish inventory before a stockout occurs
R  dL
R = Reorder point
d = average demand per period
L = lead time (in the same units as above)

EOQ Model Cost Curves
Minimum
Annual Total Cost
Total Cost
cost ($)
Holding Cost = h*Q/2
Ordering Cost = o*D/Q

Optimal Order
Quantity, Q
Order Quantity
2Do 2(Annual Demand)(Order or Setup Cost)
Q OPT = =
h Annual Holding Cost
EOQ Model Insights
• As holding costs increase, the optimal order
quantity decreases. (Order smaller amounts
more often because inventory is expensive
to hold.)
• As ordering costs increase, the optimal
order quantity increases. (Order larger
amounts less often because it is expensive
to order.)

EOQ Model Implications
Total Cost
Annual
Cost ($) Holding Cost
Ordering Cost
Q* Q*
Order Quantity
EOQ Model Implications
Total Cost
Annual
Cost ($)
Holding Cost
Ordering Cost
Q* Q*
Order Quantity
VVH Diaper Example
• Cost $5/case
• Holding costs 33% or $1.67/case-year
• Ordering costs $100
• Lead time 1 week
• She calculates annual demand as:
D  d  period
 53.5 cases of diapers  52 weeks
week year
 2,782 cases
year
VVH Diaper Example
She calculates the reorder point as
Reorder point  R  d L
 53.5 cases  1 week  53.5 cases
week
She calculates the EOQ as:
2 o D
Economic order quantity  Q* 
h
2  $100  2,782 cases

$1.67 case
 333,174 cases 2  577 cases

VVH Diaper Example
Annual demand D= 2,782 units/year

Ordering cost per order (setup) S= 100 $/order
Annual carrying cost per unit H= 1.67 $/unit-year
Working days per year = 365 days/year
Economic order quantity EOQ = 577.21 units
Actual order quantity Q= 577

Increment DQ = 500
Number of orders per year D/Q = 4.8 orders/year
Length of order cycle (days) Q/D = 75.7 days
Average inventory Q/2 = 288.5 units
Annual carrying cost (Q/2) * H = $ 481.80
Annual ordering cost (D/Q) * S = $ 482.15
Total annual cost TC = $ 963.94

Reorder Point with Safety Stock
Order
quantity (Q)
Inventory
level
Reorder
point (R)
Safety
stock (SS)
0
Lead Lead
time Time time

Reorder Point with Safety Stock
Reorder point
R  d L  SS
Safety stock
SS  z   L
where
z is the z-score associated with the desired service
level (number of standard deviations above the
mean)
L= standard deviation of demand during lead time
Normal(100, 20)
2.5
2.0
Safety Stock Normal(100, 20)
2.5
1.5
Probability of 2.0
BestFit Student Version
For Academic Use Only Reorder
meeting demand during
1.0
point
lead time = service level = 84%
1.5
0.5 BestFit Student Version

For Academic Use Only
1.0
0.0
Probability of
a stockout = 16%
10 0
120
140
16 0
40
60
80
0.5
< 84.1% 15.9% >
-Infinity 120.0
0.0
10 0
120
140
160
40
80
60
Example < 84.1%

100 120 15.9% >
units -Infinity 120.0
Average demand during

Lead time = dL
Z 0 1

Model Insights
• As the desired service level increases, the
amount of safety stock increases. (If fewer
stockouts are desired, more inventory must
be carried.)
• As the variation in demand during lead time
increases, the amount of safety stock
increases. (If demand variation or lead time
can be decreased, less safety stock is
needed.)

VVH Diaper Example
• Desired service level = 95 percent
- With five orders/year, this means that the hospital would
experience one stockout every four years
• Standard deviation of demand during lead = σL =
11.5 cases of diapers
• Amount of safety stock needed:
SS  z   L  1.64 11 .5  18.9 cases
• New reorder point:

R  d L  SS  53.5 cases
week

1 week  18.9 cases  72.4 cases

VVH Diaper Example
Average daily
demand d= 7.64 units
Average lead time L= 7 days
Reorder Point
Std dev demand during lead
time L = 11.5 units
Probability
Service level SL = 0.95
Increment SL =
Stock out risk 0.05
0.0 20.0 40.0 60.0 80.0 100.0
z associated with service level 1.64 Daliy Demand
Average demand during lead daily demand ROP

time dL = 53.48 units
Safety
stock SS = 18.9 units
Reorder point ROP = 72.4 units

VVH Diaper Example
Average demand
Order
= 53.5
quantity (577) cases/week
Inventory
level Reorder
point (72)
Safety
stock (19)
0
Lead Lead
time = Time time
1 week
More Inventory Models
• Fixed period with safety stock
- Orders are bundled and/or vendors deliver
according to a set schedule
• Quantity discounts
• Price breaks
• Etc.

Inventory Systems
• Simple
• JIT
• MRP
• ERP

Two-Bin System
When the first bin is empty,
stock is taken from the
second bin and an order is
placed. There should be
enough stock in the second
bin to last until more stock
is delivered.

JIT—Kanbans
Empty Empty
Kanban Kanban
Full Full
Kanban Kanban
Task 1 Task 2
Customer
Workstation Workstation
Order
1 2
Microsoft Visio® screen shots reprinted with permission from Microsoft Corporation.

Flow and Pull
• Continuous or single piece flow—move items
(jobs, patients, products) through the steps of the
process one at a time without interuptions or
waiting.
• Pull or just-in-time (JIT)—products or services are
not produced until the downstream customer
demands them.
• Heijunka (i.e., “make flat and level”)—eliminate
variation in volume and variety of production.
- Level patient demand

Enterprise Information
Technology Trends
E-Business
E-Commerce
Automation
Collaborative
Data Processing
Engineering
Concurrent
Computer Engineering
Integrated Business Webs
Manufacturing
SCM
ERP
MRP II Appliances
MRP I Handheld
CAD/CAM Microcomputer
Minicomputer Mobile Networks
Mainframe Networks TCP/IP
1960 1970 1980 1990 2000 2010

MRP Product Structure
Table
(end item)
Lead time = 1 week
Table top Leg

(1) (4)
Lead time = 2 weeks Lead time = 3 weeks

MRP Logic
Order
table tops
Order
table legs
Week 1 2 3 4 5

ERP Systems Link Functional
Areas

Procurement and Vendor
Relationship Management
• E-procurement
• Value-based standardization
• Outsourcing
• Vendor managed inventory (VMI)
• Automated supply carts
• Group purchasing organizations (GPO)
• Disintermediation

Strategic Supply Chain
Management
Many are the same as any other
improvement/change initiative:
• Top management support
• Employee buy-in
• Structure and staffing need to support the desired
improvements
• Process analysis and improvement
• Need relevant, accurate data and metrics
• Training

Management
• Need to evaluate cost and benefits of
technology-enabled solutions
• Need to highlight the necessity and benefits
of strategic supply chain management
• Improved inventory management through
better understanding of the systems
- Consequences of unofficial inventory
- Just-in-time systems
- Improved inventory tracking systems
Management
• Vendor partnerships
- Information sharing
- Investigation and determination of mutually
beneficial solutions
- Performance tracking
• Continually educate and support a system-
wide view of the supply chain and seek
improvement for the system rather than for
individual departments or organizations in that
system.

CH 13 Supply Chain Management

Uploaded by

Copyright:

Available Formats

CH 13 Supply Chain Management

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

CH 13 Supply Chain Management

Uploaded by

Copyright:

Available Formats

Healthcare Operations

Chapter 13. Supply Chain

Copyright 2008 Health Administration Press. All rights reserved. 13-2

Copyright 2008 Health Administration Press. All rights reserved. 13-4

Copyright 2008 Health Administration Press. All rights reserved. 13-6

Copyright 2008 Health Administration Press. All rights reserved. 13-8

-A very important Annual

Copyright 2008 Health Administration Press. All rights reserved. 13-10

Copyright 2008 Health Administration Press. All rights reserved. 13-11

Copyright 2008 Health Administration Press. All rights reserved. 13-13

b) 2.5, 4.5, 6.5, 8.5, 10.5, 12.5, 14.5, 16.5

c) 5.0, 7.5, 6.0, 4.5, 7.0, 9.5, 8.0, 6.5

Copyright 2008 Health Administration Press. All rights reserved. 13-14

Copyright 2008 Health Administration Press. All rights reserved. 13-15

• Same as series above, but with a random

Copyright 2008 Health Administration Press. All rights reserved. 13-16

Copyright 2008 Health Administration Press. All rights reserved. 13-17

Copyright 2008 Health Administration Press. All rights reserved. 13-18

Copyright 2008 Health Administration Press. All rights reserved. 13-19

Copyright 2008 Health Administration Press. All rights reserved. 13-20

Average over a given number of periods that

Copyright 2008 Health Administration Press. All rights reserved. 13-21

Simple moving average where weights are

Ft = Forecasted demand for the period

Copyright 2008 Health Administration Press. All rights reserved. 13-22

Ft = Exponentially smoothed forecast for period t

Copyright 2008 Health Administration Press. All rights reserved. 13-23

• Holt’s trend-adjusted exponential smoothing

Copyright 2008 Health Administration Press. All rights reserved. 13-24

FITt = Forecast for period t including the trend

Copyright 2008 Health Administration Press. All rights reserved. 13-26

Copyright 2008 Health Administration Press. All rights reserved. 13-28

Copyright 2008 Health Administration Press. All rights reserved. 13-29

F 14  0.5  60  0.3  43  0.2  53  53.5

Copyright 2008 Health Administration Press. All rights reserved. 13-30

Copyright 2008 Health Administration Press. All rights reserved. 13-31

MAD 7 MAD 6 MAD 8

Copyright 2008 Health Administration Press. All rights reserved. 13-32

Copyright 2008 Health Administration Press. All rights reserved. 13-33

• Basic economic order quantity (EOQ)

Copyright 2008 Health Administration Press. All rights reserved. 13-34

Dependent demand is demand that

Copyright 2008 Health Administration Press. All rights reserved. 13-37

Copyright 2008 Health Administration Press. All rights reserved. 13-39

Ordering Cost = o*D/Q

Copyright 2008 Health Administration Press. All rights reserved. 13-41

Copyright 2008 Health Administration Press. All rights reserved. 13-45

Annual demand D= 2,782 units/year

Actual order quantity Q= 577

Copyright 2008 Health Administration Press. All rights reserved. 13-46

Copyright 2008 Health Administration Press. All rights reserved. 13-47

0.5 BestFit Student Version

Example < 84.1%

units -Infinity 120.0

Average demand during

Copyright 2008 Health Administration Press. All rights reserved. 13-49

Copyright 2008 Health Administration Press. All rights reserved. 13-50

Copyright 2008 Health Administration Press. All rights reserved. 13-51

0.0 20.0 40.0 60.0 80.0 100.0