Forecast uncertain demand products
Choose the suitable place for production
Better forecast
Use both quantitative + qualitative method
Analyze historical data => forecast future trend
Use weighted factors (weighted average independent forecasts)
Shorten forecasting duration
9. SUPPLY CHAIN FLOW
TIMELINE 1993-1994
Design
• February
1992
Design
Finalized
• September
1992
• Sample
Production
Production
• November
1992
• 1st
production
order
Las Vegas
Show
• March
1993
Additional
orders
Apr-Jul
1993
Ship to
Seattle
June-
Jul
1993
Delivery to
Retailer
Sep 1993
Replenish
ment
Orders
Dec 1993-
Feb 1994
10. SUPPLY CHAIN FLOW
Total products:
200K unit/year
How to determine an
appropriate production
commitment with no
clear indications?
11. QUESTION 1&3
Using the given data, make recommendation for how many units for each style Wally
should make during the initial phase of production.
- Firstly, assume all styles are made in Hong Kong with minimum total is 10K units
- Secondly, apply the same forecast method if all product were made in China
12. FORECASTING : SPORT OBERMEYER
06
Style Laura Carolyn Greg Wendy Tom Wally Average Std. Dev
2X Std
Dev
Assault
Seduced
Entice
Electra
Gail
Daphne
Isis
Anita
Teri
Stephanie
Total
Employee’ names
Productname
X1 + X2 + ⋯ . +Xn
n
Standard Deviation
2 x Standard Deviation
13. FORECASTING : SPORT OBERMEYER
06
Style Laura Carolyn Greg Wendy Tom Wally Average Std. Dev
2X Std
Dev
Assault 2,500 1,900 2,700 2,450 2,800 2,800 2,525 340 680
Seduced
Entice
Electra
Gail
Daphne
Isis
Anita
Teri
Stephanie
Total
15. X1 + X2 + ⋯ . +Xn
n
AVERAGE FORECASTMETHOD
A normal random variable
with the mean equal to
average of member’s
forecast and 2 x standard
deviation
FORECASTING : SPORT OBERMEYER
2 X STANDARD DEVIATION
16. METHOD
Coefficient of variation
(CV) is popular index of
statistic to measure risk.
FORECASTING : OUR RECOMMENDATION
Standard deviation
Average (mean)
COEFFIECIENT OF VARIATION
0.2RULE OF THUMB:
Benchmark point is
CV <= 0.2: Low risk
0.2 < CV < 1: Higher risk
18. FORECASTING : OUR RECOMMENDATION
• COEFFICIENT OF VARIATION IS RISK
• % TO BE ORDERED = 1 – RISK = 1 –CV
• QUANTITY ORDER = EXPECTED QUANTITY x (1- CV) = AVG x (1- CV)
Example:
For style Gail ~ Avg: 1017, Standard Deviation: 194
• CV = 194/1017 = 0.19 ~ 19 % (Low risk)
• 1- CV = 1- 0.19 = 0.81
• Quantity order = 1017 x 0.81 = 823 Units
FORMULA
19. SOLUTIONS FOR QUESTION 1 &3
• INITIAL PRODUCTION
COMMITMENT : 10.000 UNITS
• 10 STYLES REPRESENTED 10 %
DEMAND
• NOW IS NOVEMBER. TIME FOR LAS
VEGAS SHOW : MARCH
• MINIMUM ORDER:
• HONGKONG: 60 UNITS / MONTH
• CHINA: 120 UNITS / MONTH
ASSUMPTIONS
• MONTHS FOR PRODUCTION: 4
• MINIMUM ORDER:
• HONG KONG: 60 X 4 = 240 UNITS
• CHINA: 120 X 4 = 480 UNITS
20. SOLUTIONS FOR QUESTION 1 &3
• CALCULATE CV
• EVALUATE RISK FOR EACH STYLE
• CHOOSE TACTICS FOR PRODUCTION
STEPS
24. QUESTION 2
Can you come up with a measure associated
with your ordering policy?
25. •Measuring the dispersion of
the quantities of order items
and relative variability of order
items
•Compare the risk between
different average forecast item
Coefficient
Variation
Higher CV
Higher deviation
compare to average
Higher risk
STANDARD DEVIATION
AVERAGE FORECAST
Coefficient
Variation
=
27. 80%
Obermeyer usually
received orders
representing 80% of
its annual volume by
the week following Las
Vegas show
20%
20% of annual
volume might be at
risk
0.2
CV < 0.2 – low risk
CV > 0.2 – high risk
items with 80% items
Produce LOWER RISK
items with 80% items
for 1st production
HIGHER RISKProduce HIGHER RISK
items with minimum
for 1st production
29. Better forecast
•Use both quantitative +
qualitative method
•Analyze historical data =>
forecast future trend
•Use weighted factors
(weighted average
independent forecasts)
•Shorten forecasting
duration
Better production
proceed
•Improve labor quality in
China
•Diversify sub-contractors
=> more flexible + increase
capacity
•Decrease lead time: raw
material located near
produce factories
•Decrease inventory cost:
prepare raw material as
latest as possible
Better sales & mkt
strategy
•Promotion strategy to
encourage pre-order
•Better mkt plan to collect
feedback/demand from
customer
•Enter new market: Northern
Hemisphere
30. QUESTION 5
How should Wally think (short-term & long-term) about
sourcing in Hong Kong versus China? What kind of policy
do you recommend?
31. - Lower labor cost (
0.78%)
- Large lot size orders (
1.200 units in same style)
- Slower production
- Quality output
concerns
- Less skilled workforce
- Less flexible production
- US trade relations risks
China
- Faster production
- High/ Reliable quality out
put
- Skilled workforce more
efficient
- More flexible production
- Better US trade relations
- Higher labor costs (10$)
- Small lot size orders (
600 units in same style)
Hong Kong
Pros
Cons
Pros
Cons
Short term
Hong Kong
Overcome
limitations of
manufacturing
in China
Long term
China
Hong Kong Vs China