I will send you around 4-6 quastion so please read the chapters from 11-15 also i will send u the chapters from the textbook latter. keep in mind that dome snswers will be from spesfice pages.
Health Administration Press
Chapter 11
Process Improvement and
Patient Flow
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Process Improvement (PI)
Measuring and improving systems
Systems
Processes
Subprocesses
Tasks
PI tools can be used at any level
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
PDCA
Plan: Define the entire process to be improved using
process mapping. Collect and analyze appropriate data for
each of element of the process.
Do: Use process improvement tool(s) to improve the
process.
Check: Measure the results of the process improvement.
Act to hold the gains: If the process improvement results
are satisfactory, hold the gains. If the results are not
satisfactory, repeat the PDCA cycle.
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
PDCA Graphically
4. Act to maintain it.
1. Plan your
corrective action.
3. Check to make sure
it is working properly.
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
2. Do it.
Health Administration Press
Flow
Theory of swift, even flow
Process is more productive as:
Speed of flow increases
Variability of process decreases
Example: advanced access
Decreased time from request to appointment (speed)
Decrease in no-shows (variability)
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Patient Flow
Hospital flow is negatively affected by variability in
scheduled demand
Surgical admissions (scheduled)
Medical admissions (emergency)
When surgical admissions have high variability, backlogs
and waiting occur
2013 study showed that IT investments were
positively related to smooth and even flow
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Actions to Improve Inpatient Flow
Establish uniform discharge time
Write discharge orders in advance
Centralize oversight of census and patient movements
(care traffic control)
Change physician rounding times
Coordinate with ancillary departments on critical
testing
Coordinate discharge with social services
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Why Use Process Mapping?
Provides a visual representation that offers an
opportunity for process improvement through
inspection
Allows for branching in a process
Provides the ability to assign and measure the
resources in each task in a process
Is the basis for process modeling via computer
simulation software
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Process-Mapping Basics
Assemble and train the team.
Determine the boundaries of the process (where it starts and ends)
and the level of detail desired.
Brainstorm the major process tasks and list them in order. (Sticky
notes are often helpful here.)
Once an initial process map (also called a flowchart) has been
generated, the chart can be formally drawn using standard symbols
for process mapping.
The formal flowchart should be checked for accuracy by all relevant
personnel.
Depending on the purpose of the flowchart, data may need to be
collected or more information may need to be added.
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Intensive
ED Care
High
Patient
Arrives
at the ED
Triage Clinical
Complexity
Vincent Valley
Hospital and
Health System
Emergency
Department (ED)
Patient Flow
Process Map
Low
Waiting
Admitting
Private
Insurance
Yes
Triage Financial
Private
Insurance
Waiting
No
Admitting
Medicaid
Nurse
History/
Complaint
Waiting
Exam/
Treatment
Waiting
Discharge
End
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Process Metrics
Capacity of a process: maximum possible amount of output (goods
or services) that a process or resource can produce or transform.
Capacity utilization: proportion of capacity actually being used.
Measured as actual output/maximum possible output
Throughput time: average time a unit spends in the process.
Includes both processing time and waiting time and is determined by
the critical (longest) path through the process
Throughput rate: average number of units that can be processed per
unit of time
Service time or cycle time: time to process one unit. Cycle time of a
process is equal to the longest task cycle time in that process
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Process Metrics (contd.)
Idle or wait time: time a unit spends waiting to be processed
Arrival rate: rate at which units arrive at the process
Work-in-process (WIP), things-in-process (TIP), patients-in-process
(PIP), or inventory: total number of units in the process
Setup time: amount of time spent getting ready to process the next
unit
Value-added time: time a unit spends in the process where value is
actually being added to the unit
Non-value-added time: time a unit spends in the process where no
value is being added. Wait time is non-value-added time
Number of defects or errors
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Basic Process Redesign Techniques
Eliminate non-value-added activities
Eliminate duplicate activities
Combine related activities
Process in parallel
Balance workloads
Develop alternative process flow paths and contingency plans
Establish the critical path
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Basic Process Redesign Techniques
(contd.)
Embed information feedback and real-time control
Ensure quality at the source
Match capacity to demand
Let the patient do the work
Use technology
Apply the theory of constraints
Identify best practices and replicate
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Queuing Theory
Queues (lines) form because of limited resources
Queuing theory is used to determine the best balance
between customer service and economic
considerations
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Queuing Notation
A/B/c/D/E
Where:
A = Interarrival time distribution
B = Service time distribution
c = Number of servers
D = Maximum queue size
E = Size of input population
When both queue and input population are assumed to be infinite, D and E are
typically omitted.
M/M/1 = exponential service time distribution, single server, infinite possible
queue length, infinite input population, assumes only one queue
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Queuing Formulas
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Queuing Formulas (contd.)
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Littles Law
Average throughput time =
People (or things) in the system/Arrival rate
Example
Clinic serves 200 patients in an 8-hour day (or 25 patients per hour).
Average number of patients in waiting room, exam rooms, etc., is 15.
15 patients/25 patients per hour = 0.6 hours in the clinic
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Discrete Event Simulation
Built on queuing theory
Basic simulation model
Entities (patients)
Queues (waiting lines)
Resources (people, equipment, space)
Based on states and events
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Discrete Event Simulation
Note: Created with Arena simulation software. M = exponential
distribution; MRI = magnetic resonance imaging
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Process Improvement Methods and
Tools
Six Sigma (chapter 9)
Seven basic tools
Benchmarking
Poka-yoke
Lean (chapter 10)
Value stream mapping
Takt time
Standardized work
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
End of Chapter 11
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Chapter 12
Scheduling and Capacity
Management
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Scheduling and Capacity Management
Hospital census and resource loading
Staff scheduling
Job/operation scheduling and sequencing rules
Patient appointment scheduling models
Advanced access scheduling
Operating and market advantages
Implementing advanced access
Metrics for advanced access
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Typical Daily Hospital Census
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Hourly Census
Patients in the System
60
50
40
30
20
10
0
1
2
3
4
5
6
7
8
9
10 11 12 13 14 15 16 17 18 19 20 21 22 23 24
Hour
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Staff Scheduling
Optimization/mathematical programming (chapter 6)
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Riverview Clinic Urgent Care Staffing
Using Linear Programming (LP)
Objective: Minimize salary and benefit expenses while satisfying
nurses
Five consecutive days, with two days off every seven days
Schedules chosen by seniority
Sun
Nurses Needed/Day
Salary and
Benefits/Nurse-Day
($/day)
5
Mon Tues
4
3
320 240 240
Wed
Thurs
3
3
240
240
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Fri
Sat
4
6
240 320
Health Administration Press
LP Problem
There are seven possible schedules (Sunday and
Monday off, Monday and Tuesday off, and so forth).
Objective is to minimize:
Salary and benefit expense = ($320 × Sun. # of nurses)
+ ($240 × Mon. # of nurses) + ($240 × Tues. # of
nurses) + ($240 × Wed. # of nurses) + ($240 × Thurs. #
of nurses) + ($240 × Fri. # of nurses) + ($320 × Sat. # of
nurses)
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
LP Problem (contd.)
Subject to:
The number of nurses scheduled each day must be
greater than the number of nurses needed each day.
Sun. # of nurses ? 5
Mon. # of nurses ? 4
The number of nurses assigned to each schedule must
be greater than 0 and an integer.
# A (B, C, D, E, F, or G) nurses ? 0
# A (B, C, D, E, F, or G) nurses = integer
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Excel? Solver Setup
Minimize Salary and Benefit Expense
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Excel? Solver Solution: Minimize Salary and Benefit
Expense (contd.)
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Excel? Solver Setup
Maximize Nurse Satisfaction
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Excel? Solver Solution: Maximize Nurse
Satisfaction (contd.)
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Job/Operation Scheduling and
Sequencing Rules
First come, first served (FCFS)
Shortest processing time (SPT)
Earliest due date (EDD)
Slack time remaining
Critical ratio (CR)
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Sequencing Rule Example
Job
A
B
C
D
E
Processing
Time
50
100
20
80
60
Due
Date
Slack
100
160
50
120
80
How many possible sequences for five jobs?
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Critical
Ratio
Health Administration Press
First Come, First Served
Sequence
A
B
C
D
E
Average
Start
Time
0
50
100
170
250
Processing
Time
50
100
20
80
60
Completion
Time
50
150
170
250
310
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Due
Date
100
160
50
120
80
Tardiness
Health Administration Press
First Come, First Served
(contd.)
Sequence
A
B
C
D
E
Average
Start
Time
0
50
100
170
250
Processing
Time
50
100
20
80
60
Completion
Time
50
150
170
250
310
186
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Due
Date
100
160
50
120
80
Tardiness
0
0
120
130
230
96
Health Administration Press
Shortest Processing Time
Sequence
C
A
E
D
B
Average
Start
Time
0
20
70
130
210
Processing
Time
20
50
60
80
100
Completion
Time
20
70
130
210
310
148
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Due
Date
50
100
80
120
160
Tardiness
0
0
50
90
150
58
Health Administration Press
Earliest Due Date
Sequence
C
E
A
D
B
Average
Start
Time
0
20
80
130
210
Processing
Time
20
60
50
80
100
Completion
Time
20
80
130
210
310
150
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Due
Date
50
80
100
120
160
Tardiness
0
0
30
90
150
54
Health Administration Press
Sequencing Rule Example
Job
Processing
Time
Due
Date
Slack
Critical
Ratio
A
50
100
100 ? 50 = 50
100/50 = 2.00
B
C
D
E
100
20
80
60
160
50
120
80
160 ? 100 = 60
50 ? 20 = 30
120 ? 80 = 40
80 ? 60 = 20
160/100 = 1.60
50/20 = 2.50
120/80 = 1.50
80/60 = 1.25
120 possible sequences for five jobs
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Slack Time Remaining
Slack for each job: A50, B60, C30, D40, E20
Sequence
E
C
D
A
B
Average
Start
Time
0
60
80
160
210
Processing
Time
60
20
80
50
100
Completion
Time
60
80
160
210
310
164
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Due
Date
80
50
120
100
160
Tardiness
0
30
40
110
150
66
Health Administration Press
Critical Ratio (CR)
CR for each job: A2.00, B1.60, C2.50, D1.50, E1.25
Sequence
E
D
B
A
C
Average
Start
Time
0
60
140
240
290
Processing
Time
60
80
100
50
20
Completion
Time
60
140
240
290
310
208
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Due
Date
80
120
160
100
50
Tardiness
0
20
80
190
260
110
Health Administration Press
Summary
Rule
Average
Completion Time
FCFS
SPT
EDD
SLACK
CR
*Best values
186
148*
150
164
208
Average
Tardiness
No. of
Jobs Tardy
96
58*
54*
66
110
3*
3*
3*
4
4
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Maximum
Tardiness
230
150*
150*
150*
260
Health Administration Press
Guidelines for Selecting a
Sequencing Rule
1.
SPT is most useful for a very busy resource.
2.
3.
Some jobs may never be completed.
SPT often is used with another rule.
Use EDD when only small tardiness values can be tolerated.
Use FCFS when there is excess capacity.
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Appointment Scheduling Models
Purpose is to balance the competing goals of:
Maximizing resource utilization
Minimizing waiting time
Four types:
Block appointment
Individual appointment
Mixed block-individual appointment
Combinations
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Bailey-Welch Schedule
Bailey-Welch
Individual Appointment
Time
# Scheduled
Time
# Scheduled
0:00
1
0:00
2
0:20
1
0:20
1
0:40
1
0:40
1
1:00
1
1:00
1
1:20
1
1:20
0
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Advanced Access
Traditional scheduling systems
Long times until next appointment
High no-show rates
Double/triple bookingqueues form
Advanced access
Patients seen same day as request
Reduces no-show rate
Better continuity of care
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Implementing Advanced Access
Obtain buy-in
Predict demand
Predict capacity
Littles law (chapter 11)
Standardize and minimize types of visit times
Assess operations
Work down backlog
Go live
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Advanced Access Metrics
PCP match: percentage of same-day patients who see their
PCP
PCP coverage: percentage of same-day patients seen by any
physician
Wait time for next appointment (or third next available
appointment)
Good backlog: appointments scheduled in advance because
of patient preference
Bad backlog: appointments waiting because of lack of slots
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
End of Chapter 12
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Chapter 13
Supply Chain Management
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Supply Chain Management (SCM)
What is supply chain management (SCM)?
Why is SCM important for healthcare organizations?
Tracking and managing inventory
Forecasting
Inventory models
Inventory systems
Procurement and vendor relationship management
Strategic SCM
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Supply Chain Management
(contd.)
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
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
SCM in Healthcare
Potential savings of 28% of overall operating
costs with effective supply chain management
of tangible goods
Procurement costs can be reduced >10%
Quantity of items purchased can be reduced
>20%
With $500M+ hospital budgets, savings
potential is enormous
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Inventory
Inventory is the stock of items held by the
organization either for sale or to support the
delivery of a service
Inventory management answers three
questions:
How much to hold
How much to order
When to order
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
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
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
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
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
ABC Classification System
Classifying inventory
according to some
measure of importance
and allocating control
efforts accordingly
Pareto Principle
– A very important
– B moderately
important
High (80%)
Annual
$ volume
of items
A
B
C
Low (5%)
Few
(20%)
– C least important
Many
(50%)
Number of Items
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
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
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Forecasting
Exercises
Averaging methods
Trend, seasonal, and cyclical models
Model development and evaluation
VVH example
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Forecasting: Exercise 1
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?
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Exercise 1: Graphs
Series b
18
16
14
12
10
Series1
8
6
Series a
Series c
4
4.4
2
10.0
4.2
0
9.0
1
2
3
4
5
6
7
8
4.0
8.0
7.0
3.8
6.0
Series1
3.6
Series1
5.0
3.4
4.0
3.2
3.0
3.0
2.0
1.0
2.8
1
2
3
4
5
6
7
8
0.0
1
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
2
3
4
5
6
7
8
Health Administration Press
Exercise 1: 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
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Forecasting: Exercise 2
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
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Exercise 2: 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 + ?
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
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
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
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
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Demand
Demand
Forms of Forecast Movement
Trend
Cycle
Random
movement
Time
Seasonal
pattern
Demand
Demand
Time
Trend with
seasonal pattern
Time
Time
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Forecasting: Averaging Methods
Simple moving average
Weighted moving average
Exponential smoothing
Averaging methods all assume that the variable of
interest is relatively constant over time; no trends or
cycles
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
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
F
t
=D
t ?1
+ Dt ? 2 ? + Dt ? n
n
Ft = Forecasted demand for the period
Dt-1 = Actual demand in period t ? 1
n = Number of periods in the moving average
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
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 D t ? 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
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Exponential Smoothing
Times series forecasting technique that does not
require large amounts of historical data
F = (1? ? )F
t
Ft =
Ft-1 =
Dt-1 =
? =
+
?
Dt ?1
t ?1
Exponentially smoothed forecast for period t
Exponentially smoothed forecast for prior period
Actual demand in the prior period
Desired response rate, or smoothing constant
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Forecasting: Trend, Seasonal, and
Cyclical Models
Holts trend-adjusted exponential smoothing
technique
Winters triple exponential smoothed model
ARIMA models
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Holts Trend-Adjusted Exponential Smoothing
Exponentially smoothed forecast that accounts for a trend in the data
FITt = Ft + Tt
and
Ft = ?Dt ?1 + ( 1 ? ?)FITt ?1
Tt = Tt ?1 + ?(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
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Forecast Accuracy
Error = Actual ? Forecast
Find a method that minimizes error
Mean absolute deviation (MAD)
Mean squared error
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
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
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
VVH Diaper Example
Week of Period Actual
1-Jan
8-Jan
15-Jan
22-Jan
29-Jan
5-Feb
12-Feb
19-Feb
26-Feb
5-Mar
12-Mar
19-Mar
26-Mar
1
2
3
4
5
6
7
8
9
10
11
12
13
70
42
63
52
56
53
66
61
45
54
53
43
60
Weekly Demand
80
70
60
50
40
30
20
10
0
1
2
3
4
5
6
7
Period
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
8
9
10
11
12
13
Health Administration Press
VVH Simple Moving Average
F
F
=D
t ?1
t
+ Dt ?2? + Dt ?n
n
=D
13
14
+ D12 + D11 + D10 + D9
5
60 + 43 + 53 + 54 + 45
=
= 51
F 14
5
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
VVH Weighted Moving Average
F t = w D + w D +?+ w D
F 14 = w D + w D + w D
F 14 = 0.5 ? 60 + 0.3 ? 43 + 0.2 ? 53 = 53.5
t ?1
13
t ?1
13
t ?2
12
t ?2
12
t ?n
11
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
11
t ?n
Health Administration Press
VVH Exponential Smoothing
F
F
F
t
= ? Dt ?1 + (1 ? ? ) F t ?1
14
= ? D13 + (1 ? ? ) F 13
14
= (0.25 ? 60) + (0.75 ? 52) = 54
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
VVH Comparison
(from the Excel? template)
Weight 3
Weight 2
0.3
Periods
5
Least
Recent
0.2
MAD
MSE
7
86
MAD
MSE
6
75
Period Actual Forecast Error
1
70
2
42
3
63
4
52
5
56
6
53
57
4
7
66
53
13
8
61
58
3
9
45
58
13
10
54
56
2
11
53
56
3
12
43
56
13
13
60
51
9
14
51
Weight 1
Most
0.5 Recent
Period Actual Forecast Error
1
70
2
42
3
63
4
52
58
6
5
56
53
3
6
53
56
3
7
66
54
12
8
61
60
1
9
45
61
16
10
54
54
0
11
53
53
0
12
43
52
9
13
60
48
12
14
53.5
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
?
MAD
MSE
0.25
8
135
Period Actual Forecast Error
1
70
2
42
70
28
3
63
63
0
4
52
63
11
5
56
60
4
6
53
59
6
7
66
58
8
8
61
60
1
9
45
60
15
10
54
56
2
11
53
56
3
12
43
55
12
13
60
52
8
14
54
Health Administration Press
Realities of Forecasting
Forecasts are seldom perfect.
Most forecasting methods assume
that there is some underlying
stability in the system.
Service family and aggregated
service forecasts are more accurate
than individual service
forecasts.
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
I see that you will
get an A this
semester.
Health Administration Press
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
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Definitions
Lead timetime between placing an order and receiving the
order
Holding (or carrying) costscosts associated with keeping goods
in storage
Ordering (or setup) costscosts of ordering and receiving goods
Shortage costscosts of not having something in inventory
when it is needed
Back ordersunfilled orders
Stockoutsoccur when the desired good is not available
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
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).
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
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.
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
Inventory Order Cycle
Demand
rate
Order
quantity, Q
Average
amount of
inventory
held = Q/2
Inventory
Level
Reorder
point, R
0
Time
Lead
Lead
time
time
Order Order
Order Order
Placed Received Placed Received
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
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)
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
EOQ Model Cost Curves
Minimum
Total Cost
Annual
cost ($)
Total Cost
Holding Cost = h*Q/2
Ordering Cost = o*D/Q
Optimal
Order Quantity
Q OPT =
2Do
2(Annual Demand)(Or der or Setup Cost)
=
h
Annual Holding Cost
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Order
Quantity, Q
Health Administration Press
EOQ Model Insights
As holding costs increase, the optimal order quantity
decreases. (Order smaller amounts more often
because inventory is expensive to hold.)
As ordering costs increase, the optimal order quantity
increases. (Order larger amounts less often because it
is expensive to order.)
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
EOQ Model Implications
Total Cost
Annual
Cost ($)
Holding Cost
Ordering Cost
Q*
Q*
Order Quantity
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
EOQ Model Implications
Total Cost
Annual
Cost ($)
Holding Cost
Ordering Cost
Q*
Q*
Order Quantity
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
Health Administration Press
VVH Diaper Example
Cost $5/case
Holding costs 33% or $1.67/case-year
Ordering costs $100
Lead time 1 week
Annual demand calculated as:
D = d ? period
= 53.5 cases of diapers
= 2,782 cases
week
? 52 weeks
year
Copyright © 2017 Foundation of the American
College of Healthcare Executives.
year
Health Administration Press
VVH Diaper E
Recent Comments