Supply Chain Modeling: Analysis of Demand Variability and

Download Report

Transcript Supply Chain Modeling: Analysis of Demand Variability and

Supply Chain Modeling: Analysis of Demand Variability
and Volumetric Capacity Needs for Contraceptives and
MCH Products
James Gibney
Anabella Sánchez
Carlos Lamadrid
February 20, 2009
Agenda
Introduction: Overview of Unmet Need and Areas of Interest
Part A: Volumetrics of Guatemala’s Integrated Supply Chain
Part B: Effect of Demand Variability on Contraceptive Logistics
Introduction: Low Unmet Demand in 2007 For Contraceptives – But
Analysis Still Needed to Ensure Success in 2008/9
100%
Observation:
No significant stock-out
problems existed in 2007 for
contraceptives
1%
3%
2%
1%
3%
1%
1%
6%
3%
90%
80%
70%
97%
98%
97%
100%
100%
100%
99%
100%
100%
100%
100%
99%
94%
99%
97%
Or al
s
99%
40%
30%
20%
10%
Guatemala Total
Jutiapa
Consumption
Totonicapán
Unfulfilled Demand
Source: MSPAS National RH Program, “Real Demand for Contraceptive Methods (2007)”
Inj ec
t abl e
IUD
oms
Cond
Or al
s
s
Inj ec
t abl e
IUD
oms
Cond
Or al
s
s
Inj ec
t abl e
IUD
oms
Cond
Or al
s
s
Inj ec
t abl e
IUD
0%
oms
B. What is the effect that
variability has on the
system? Are stock-outs
being avoided due to high
level of emergency orders?
50%
Cond
Areas of Interest:
A. Given the integrated
supply system, what are the
volume requirements
needed to ensure full supply
for contraceptives and MCH
products?
s
60%
Sololá
Part A: Volumetrics of Guatemala’s Integrated Supply Chain
Background:
With little exception, the distribution network (warehousing and transport) of products in
Guatemala’s public health system is integrated. To better understand the volume capacity
needs of this system, data on all products from Jutiapa and Totonicapán was analyzed. A
special focus was placed on contraceptives and MCH products.
Questions this section answers:
1.
In terms of volumes, what does the average monthly demand for all products that flow
through a DAS warehouse look like? How does this differ for health centers, health posts,
NGO’s, and hospitals? What are the products that take up the most space?
2.
What are the space requirements needed for the family planning products? How do these
products compare to each other? What are the different requirements per SDP site type?
How to they compare to non family planning products?
3.
What are the space requirements needed for the MCH products? How do these products
compare in space needs to each other? How to they compare to non MCH products?
Volumes of Product Types in the Integrated System
Volume by % of Average Monthly Demand -- All Products
100%
•
Contraceptives represent less
than 1% of volume for
Totonicapán and 3% for
Jutiapa
0.5%
1.3%
3.1%
2.3%
32.0%
28.4%
30.1%
66.6%
68.5%
67.6%
Health Posts
Health Centers
0.8%
0.6%
90%
29.5%
38.4%
80%
33.6%
70%
60%
•
Difference is attributed to
higher demand of condoms in
Jutiapa
50%
40%
69.9%
60.8%
30%
•
MCH accounts for
approximately 1/3 and other
products take 2/3 of volume
capacity
65.8%
20%
10%
0%
Health Posts
and Health
Centers
Health Posts
Jutiapa
Other Products
Source: MSPAS Logistics Module (January 2007 to April 2008)
Health Centers
Totonicapán
Maternal & Child Health
Contraceptives
Health Posts
and Health
Centers
Volumetric Composition of Average Monthly Demand By Products
(Combined Average Monthly Real Demand for Totonicpán and Jutiapa)
•
•
80%
volume
•
•
Table provides 105
products
Contraceptives are
highlighted blue,
Condoms take 1.3%
of total
Volume composition
follows 80/20 rule
Acetaminofen syrup
and Bromexina take
approx. 20% of
volume
Volume (cubic cm) Requirements of Average Monthly Demand
•
•
Jutiapa has double the
capacity need at SDP’s
than Totonicapán
In Totonicapán Health
Posts require more volume
capacity than health
centers for MCH and Other
Drugs, but not
Contraceptives
253,399
8,000,000
7,000,000
89,559
2,309,746
6,000,000
2,182,770
5,000,000
21,660
4,000,000
1,187,586
•
In Jutiapa, the Health
Centers require more
volume capacity for all
product types, with
Contraceptives requiring
significantly more
volumetric capacity
(approx. 3x)
3,000,000
5,565,604
25,633
1,286,735
4,540,039
2,000,000
2,814,298
2,035,063
1,000,000
Health Posts
Health Centers
27,321
14,177
NGOs
Hospital
Health Posts
Jutiapa
Other Products
Source: MSPAS Logistics Module (January 2007 to April 2008)
Health Centers
15,984
3,287
NGOs
Hospital
Totonicapán
Maternal & Child Health
Contraceptives
Contraceptive Volumes – Monthly Demand Averages For SDP Types
•Percentages show relative capacity needs
for storage and transport
•Commonalities between Jutiapa and
Totonicapan:
•Same order of magnitude
(Hospitals, then NGO’s, then Health
Posts, then Health Centers)
•Hospital percentages most similar
•Health Post percentages are 2nd
most similar
•Differences between Jutiapa and
Totonicapan
100%
90%
80%
3.7%
7.1%
4.9%
24.0%
23.3%
70%
60%
32.5%
50%
40%
30%
65.9%
20%
•Totonicapan higher relative flow to
NGO’s
•Jutiapa higher relative flow to Health
Centers (almost 2 times that of
Totonicipan)
38.5%
10%
0%
Jutiapa
Health Centers
Totonicapán
Health Posts
Source: MSPAS Logistics Module (January 2007 to April 2008)
NGOs
Hospital
For the contraceptives part of the integrated order (1%-3% of total volume, what
is the volumetric composition by product?
Volume by % of Average Monthly Demand -- Contraceptives
100%
0.0%
0.6%
11.6%
7.7%
90%
0.4%
8.8%
3.5%
0.0%
17.3%
0.0%
9.2%
5.0%
0.0%
3.9%
10.0%
19.6%
25.7%
36.9%
31.4%
22.1%
36.3%
70%
6.9%
25.6%
12.4%
80%
0.5%
33.2%
42.1%
39.7%
19.0%
60%
27.4%
51.1%
21.0%
50%
30.6%
24.1%
22.7%
79.3%
40%
71.3%
67.4%
30%
37.6%
55.4%
48.7%
43.4%
20%
42.6%
31.6%
32.5%
31.8%
10%
38.7%
13.4%
0%
Health
Posts
Health
Centers
Health Hospital
Posts
and
Health
Centers
NGOs
All
Sites
Jutiapa
CONDOM
Health
Posts
Health
Centers
Health Hospital
Posts
and
Health
Centers
NGOs
Totonicapán
LOFEMENAL
Source: MSPAS Logistics Module (January 2007 to April 2008)
DEPO PROVERA
IUD
All
Sites
What is the Average Monthly MCH Demand By Volume?
Totonicapan - MCH Average Monthly Demand by Volume
Jutiapa - MCH Average Monthly Demand by Volume
100%
LIDOCAINA CLORHIDRATO (SIMPLE) 2
% Vial
90%
80%
70%
60%
50%
40%
LIDOCAINA CLORHIDRATO (SIMPLE) 2 %
Vial
100%
ALBENDAZOL 200 mg Tableta
ACIDO FOLICO 5 mg Tableta
90%
ALBENDAZOL 200 mg Tableta
7.7%
4.5%
9.1%
11.1%
11.6%
13.5%
5.2%
3.8%
8.7%
13.3%
11.5%
12.5%
9.5%
9.1%
11.8%
ACIDO FOLICO 5 mg Tableta
PRENATALES 0 Sin Concentración
Gragea
ALBENDAZOL 200/5 mg/ml Frasco
suspensión
HARTMAN (RINGER LACTATO) 1000 ml
Bolsa/Frasco
6.1%
6.6%
70%
AMOXICILINA 500 mg Tableta
ACETAMINOFEN 500 mg Tableta
14.4%
80%
60%
50%
40%
6.7%
7.5%
11.0%
11.6%
7.6%
4.6%
5.6%
ALBENDAZOL 200/5 mg/ml Frasco
suspensión
8.0%
AMOXICILINA 500 mg Tableta
5.1%
AMOXICILINA + ACIDO CLAVULANICO
125/5 mg/ml Frasco suspensión
9.6%
ACETAMINOFEN 500 mg Tableta
10.3%
AMOXICILINA + ACIDO CLAVULANICO
250+62.5/5 mg/ml Frasco suspensión
5.2%
10.1%
12.6%
13.0%
PRENATALES 0 Sin Concentración
Gragea
ERITROMICINA 250/5 mg/ml Frasco
suspensión
30%
TRIMETROPRIMA SULFAMETOXAZOL
40-200/5 mg/ml Frasco suspensión
20%
30.5%
30.8%
30.2%
10%
SALBUTAMOL 2/5 mg/ml Frasco Jarabe
AMOXICILINA 250/5 mg/ml Frasco
suspensión
SALES DE REHIDRATACION ORAL 55.8
g Sobre
0%
Health Posts Health Posts
and Health
Centers
Health
Centers
ACETAMINOFEN 120/5 mg/ml Frasco
Jarabe
SALBUTAMOL 2/5 mg/ml Frasco Jarabe
30%
20%
HARTMAN (RINGER LACTATO) 1000 ml
Bolsa/Frasco
34.3%
36.8%
31.5%
10%
TRIMETROPRIMA SULFAMETOXAZOL 40200/5 mg/ml Frasco suspensión
ACETAMINOFEN 125/5 mg/ml Frasco
Jarabe
ERITROMICINA 250/5 mg/ml Frasco
suspensión
AMOXICILINA 250/5 mg/ml Frasco
suspensión
0%
Health Posts Health Posts
and Health
Centers
Health
Centers
Acetaminofen takes up over 30% in both Departments
SALES DE REHIDRATACION ORAL 55.8 g
Sobre
ACETAMINOFEN 120/5 mg/ml Frasco
Jarabe
Effect of Demand Variability on Contraceptive Logistics
Background:
Given the complexity of large logistical networks, such as the MSPAS
system, the effect of high variability on a supply chains can be both significant
and hard to measure. Accordingly, it is necessary to measure the variability
levels and use advanced analysis methods, such as modeling, to determine
the effect.
Questions this section answers:
1. Does the use of 3 months historical averages in making orders reduce the
variation to insignificant amounts?
2. How does variability effect service levels? Can there still be a stock-out even
if 1/3 min/max levels are followed? Does an increase in variability increase
the propensity for emergency orders? Does it increase the propensity to be
overstocked?
Variability in Contraceptive Monthly Real Demand
Observations:
•
Real demand changes
significantly on a monthly basis
•
Jutiapa shows lowest variability in
all products
•
High IUD variability partially
attributed to low order amounts
(ie. change from 4 to 8 represents
100%)
Average Change in Monthly Real Demand for 2007
Jutiapa
Totonicapan
Solola
31%
37%
orales
30%
18%
inyectable
39%
14%
Ramifications:
•
Variability levels can
significantly effect service
rates of logistics system, even
if system follows the min/max
ordering system
•
Causes of Variability should be
addressed
– Are reporting procedures
being followed
– or are there really high
swings in real demand?
101%
99%
tdecobre
91%
28%
29%
condones
15%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
110%
Does Using 3-Month Historical Average Smooth-away Variability?
•
•
•
•
Table below compares projected demand based on 3 month historical averages to the actual
demand for 2007
Red highlights represent projected under-estimating by more than 20%
Yellow highlights represent projected demand over-estimating by more than 20%
Results: 50% of the time the projected was either over or under by more than 20%. Using
projections based on 3-months historical data does not necessarily remove variability.
Complexity of system requires the use of advanced supply chain modeling software.
BR
E
E
IEM
BR
VIE
M
RE
TU
B
SE
PT
IEM
B
-23%
22%
3%
-8%
-5%
-63%
-10%
-22%
5%
44%
2%
8%
-21%
-64%
-8%
-13%
13%
0%
1%
13%
46%
700%
64%
104%
Totonicapan
condones
tdecobre
inyectable
orales
-17%
-47%
-24%
-50%
-12%
11%
12%
27%
7%
33%
-17%
-4%
4%
22%
-10%
-5%
35%
-70%
28%
10%
-18%
-41%
-23%
-19%
-26%
267%
19%
-6%
-47%
-22%
-13%
-8%
171%
67%
32%
29%
88%
67%
19%
32%
Solola
condones
tdecobre
inyectable
orales
52%
-70%
-11%
-8%
-3%
33%
31%
44%
9%
-33%
24%
31%
18%
0%
-9%
-10%
7%
-10%
-14%
-31%
50%
-39%
-61%
-5%
-12%
-20%
-37%
27%
-17%
8%
-77%
6%
4%
10%
157%
-23%
4%
22%
97%
AB
MA
DIC
-12%
183%
7%
13%
NO
5%
33%
-7%
2%
OC
AG
OS
TO
JU
3%
0%
5%
51%
LIO
NIO
JU
-13%
-38%
-11%
-40%
MA
condones
tdecobre
inyectable
orales
RIL
Jutiapa
RZ
YO
O
RE
Projected Demand as Percentage of Actual Demand For Aggregated 2007 Quantities
*Projected amounts calculated using average of previous 3 months
Modeling The Department of Sololá
Structure
•
1 National Warehouse
•
1 Department
Warehouses
•
10 District
Warehouses
•
33 Health Centers
•
10 NGOs
•
1 Hospital
Rules
•
Products
•
Inventory Policies
•
Sourcing Policies
•
Transportation Policies
Effect of Variability on Sololá Health Center Condom Stock Outs and Emergency Orders
Scenario Outputs
25% Variability in Monthly Demand for
Condoms
500
Inventory
Stock Out Periods
– no variability scenario – 0
– 25% variability scenario – 1
– 50% variability scenario – 5
Instances Crossing Above Max (overstocked)
– No variability scenario – 0
– 25% variability scenario – 5
– 50% variability scenario – 2
Instances Crossing Below Min (emergency order)
– No variability scenario – 0
– 25% variability scenario – 8
– 50% variability scenario – 6
400
300
200
100
0
Aug-07 Nov-07 Feb-08 May-08 Aug-08 Nov-08 Feb-09 May-09 Aug-09
(Settings: 106 condoms/month; 1/3 month min/max; 2 year period, reorder
quantity based on previous 3 month demand)
50% Variability in Monthly Demand for
Condoms
500
500
400
400
300
200
100
0
Aug-07 Nov-07 Feb-08 May-08 Aug-08 Nov-08 Feb-09 May-09 Aug-09
Inventory
Inventory
No Variability in Monthly Demand for
Condoms
300
200
100
0
Aug-07 Nov-07 Feb-08 May-08 Aug-08 Nov-08 Feb-09 May-09 Aug-09
How variable is the data? Examples of condom real demand - Jutiapa
Agua Blanca - Health Center
450
Mean: 214
Standard Deviation: 103 (48% of
mean)
Asunción Mita - Health Center
4500
3500
300
3000
250
2500
200
2000
150
1500
100
1000
50
500
0
0
Ja
n0
Fe 7
b0
M 7
ar
-0
Ap 7
r0
M 7
ay
-0
Ju 7
n07
Ju
l-0
Au 7
g0
Se 7
p0
O 7
ct
-0
N 7
ov
-0
D 7
ec
-0
Ja 7
n0
Fe 8
b0
M 8
ar
-0
Ap 8
r08
4000
350
Ja
n0
Fe 7
b0
M 7
ar
-0
Ap 7
r0
M 7
ay
-0
Ju 7
n07
Ju
l-0
Au 7
g0
Se 7
p0
O 7
ct
-0
N 7
ov
-0
D 7
ec
-0
Ja 7
n0
Fe 8
b0
M 8
ar
-0
Ap 8
r08
400
Mean: 2,564
Standard Deviation: 826 (32% of
mean)
Source: MSPAS Logistics Module data
Atescatempa - Health Center
1800
Mean: 1,059
Standard Deviation: 311 (29% of
mean)
1600
1400
1200
Jutiapa - Health Center
4000
Mean: 2,245
Standard Deviation: 709 (32% of
mean)
3500
3000
2500
1000
800
600
2000
1500
200
500
0
0
Ja
n0
Fe 7
b0
M 7
ar
-0
Ap 7
r0
M 7
ay
-0
Ju 7
n07
Ju
l-0
Au 7
g0
Se 7
p0
O 7
ct
-0
N 7
ov
-0
D 7
ec
-0
Ja 7
n0
Fe 8
b0
M 8
ar
-0
Ap 8
r08
1000
Ja
n0
Fe 7
b0
M 7
ar
-0
Ap 7
r0
M 7
ay
-0
Ju 7
n07
Ju
lAu 07
g0
Se 7
p0
O 7
ct
-0
N 7
ov
-0
D 7
ec
-0
Ja 7
n0
Fe 8
b0
M 8
ar
-0
Ap 8
r08
400
How variable is the data? Examples of depo real demand - Jutiapa
Moyuta - Health Center
300
Mean: 170
Standard Deviation: 37 (22% of
mean)
Yupiltepeque - Health Center
100
Mean: 73
Standard Deviation: 11 (15% of
mean)
90
250
80
70
200
60
150
50
40
100
30
20
50
10
0
Ja
n0
Fe 7
b0
M 7
ar
-0
Ap 7
r0
M 7
ay
-0
Ju 7
n07
Ju
l-0
Au 7
g0
Se 7
p0
O 7
ct
-0
N 7
ov
-0
D 7
ec
-0
Ja 7
n0
Fe 8
b0
M 8
ar
-0
Ap 8
r08
Ja
n0
Fe 7
b0
M 7
ar
-0
Ap 7
r0
M 7
ay
-0
Ju 7
n07
Ju
l-0
Au 7
g0
Se 7
p0
O 7
ct
-0
N 7
ov
-0
D 7
ec
-0
Ja 7
n0
Fe 8
b0
M 8
ar
-0
Ap 8
r08
0
Source: MSPAS Logistics Module data
Comapa - Health Center
200
Mean: 156
Standard Deviation: 14 (9% of
mean)
180
160
140
120
El Adelanto - Health Center
140
Mean: 79
Standard Deviation:21 (26% of
mean)
120
100
80
100
80
60
60
40
40
Ja
n0
Fe 7
b0
M 7
ar
-0
Ap 7
r0
M 7
ay
-0
Ju 7
n07
Ju
l-0
Au 7
g0
Se 7
p0
O 7
ct
-0
N 7
ov
-0
D 7
ec
-0
Ja 7
n0
Fe 8
b0
M 8
ar
-0
Ap 8
r08
0
20
0
Ja
n0
Fe 7
b0
M 7
ar
-0
Ap 7
r0
M 7
ay
-0
Ju 7
n07
Ju
l-0
Au 7
g0
Se 7
p0
O 7
ct
-0
N 7
ov
-0
D 7
ec
-0
Ja 7
n0
Fe 8
b0
M 8
ar
-0
Ap 8
r08
20
Questions?