Documente Academic
Documente Profesional
Documente Cultură
Chapter 13
Operations As a Competitive
Weapon
Operations Strategy
Project Management Process Strategy
Process Analysis
Process Performance and Quality
Constraint Management
Process Layout Supply Chain Strategy
Lean Systems Location
Inventory Management
Forecasting
Sales and Operations Planning
Resource Planning
Scheduling
Demand Forecasting:
A prediction of future events used for planning
purposes
Forecasting across the organization:
Internal customers throughout the organization
depend on forecasts to formulate and execute their
plans. Examples:
Finance: to project cash flow & capital requirements
HR: to anticipate hiring & training needs
Marketing: the primary source for sales forecasts
Operations: to plan output levels, inventory levels,
schedules, and long term capabilities.
© 2007 Pearson Education
Forecasting at Unilever
Customer demand planning (CDP), which is
critical to managing value chains, begins with
accurate forecasts.
Unilever has a state-of-the-art CDP system that
blends historical shipment data with promotional
data and current order data.
Statistical forecasts are adjusted with planned
promotion predictions.
Forecasts are frequently reviewed and adjusted
with point of sale data.
This has enabled Unilever to reduce its inventory
and improved its customer service.
Example:
Demand for lawn fertilizer predictably increases in the
spring and summer months. Also demand at weekends
may depend on uncontrollable factors such as the
weather
© 2007 Pearson Education
Demand Patterns
Time Series: The repeated observations of demand for a service
or product in their order of occurrence.
There are five basic patterns of most time series.
a. Horizontal. The fluctuation of data around a constant mean.
b. Trend. The systematic increase or decrease in the mean of the
series over time.
c. Seasonal. A repeatable pattern of increases or decreases in
demand, depending on the time of day, week, month, or season.
d. Cyclical. The less predictable gradual increases or decreases
over longer periods of time (years or decades).
e. Random. The unforecastable variation in demand.
Horizontal Trend
Seasonal Cyclical
Level of Aggregation:
For individual service or product (for manufacturing plan)
For total demand for groups and then divide to find
individual service or product forecast.
Units of measurement:
Forecasts of sales revenue are not helpful because
prices fluctuate.
regression Y = a + bX
equation
{ Actual
Y = dependent variable
X = independent variable
a = Y-intercept of the line
value b = slope of the line
of Y
Value of X used
to estimate Y
X
Independent variable
© 2007 Pearson Education
Linear Regression
Example 13.1
The following are sales and advertising data for the past 5
months for brass door hinges. The marketing manager says that
next month the company will spend $1,750 on advertising for the
product. Use linear regression to develop an equation and a
forecast for this product.
We use the computer to determine
Sales Advertising the best values of a, b, the correlation
Month (000 units) (000 $) coefficient (r), the coefficient of
1 264 2.5 determination (r2), and the standard
2 116 1.3 error of the estimate (syx).
3 165 1.4
a = – 8.135
4 101 1.0
5 209 2.0 b = 109.229X
r = 0.98
r2 = 0.96
© 2007 Pearson Education
syx= 15.603
Linear Regression Line for
Example 13.1
300 —
Y = a + bX
Sales (thousands of units)
250 —
Y = – 8.135 + 109.229X
200 —
150 — a = – 8.135
b = 109.229X
100 — r = 0.98
50 — r2 = 0.96
syx= 15.603
| | | |
1.0 1.5 2.0 2.5
Advertising (thousands of dollars)
Patient
Week Arrivals
1 400
2 380
3 411
b. If the actual number of patient arrivals in week
4 is 415, what is the forecast error for week 4?
410 —
Patient arrivals
390 —
370 —
Actual patient
arrivals
| | | | | |
0 5 10 15 20 25 30
Week
© 2007 Pearson Education
Comparison of
3- and 6-Week MA Forecasts
Week
© 2007 Pearson Education
Application 13.1
Example: if
80 —
Trend-adjusted
70 — forecast
60 —
Patient arrivals
50 —
40 — Actual blood
test requests
30 —
| | | | | | | | | | | | | | |
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15
Week
© 2007 Pearson Education
Forecast for Medanalysis Using the
Trend-Adjusted Exponential Smoothing Model
thousand
thousand
To make forecasts for periods beyond the next period, multiply the trend
estimate by the number of additional periods, and add the result to the
current
© 2007 average
Pearson Education
© 2007 Pearson
© 2007 Pearson Education
Seasonal Patterns
[ |Et | / Dt ](100)
MAPE =
n
– 15
Average forecast error (mean bias): E= = – 1.875
8
5275
Mean squared error: MSE = = 659.4
8
195
Mean absolute deviation: MAD = = 24.4
8
81.3%
Mean absolute percent error: MAPE = = 10.2%
8