BULLWHIP EFFECT

Uncertainty in supply chain

• One of the main business goals is to aim for the match between supply and
demand. In other words, the sourcing of raw materials, accessories, parts and
services must match production and distribution needs to meet customer demand,
which is inherently uncertain, without increasing inventory costs.
• Consequences of uncertainty and variability in the supply chain are
delayed/unfulfilled orders.
• What are reasons for uncertainty in
supply chain?
Uncertainty in supply chain

Reasons for uncertainty in supply chain: - Batch ordering;

- Inaccurate customer-demand forecast; - Price fluctuation;

- Long and changable “Lead time”; - Amplification of order variance

- Late delivery;

- Insufficient delivery;

- Product change;
The amplification of order variability, also known as bullwhip effect
 Uncertainty leads to variability in supply chain

Manufacturer Forecast
Volumes of Sales

Actual
Consumer
Retailer Warehouse Demand
Retailer Orders to Shop

Production Plan

Time
Tom Mc Guffry, Electronic Commerce and Value Chain Management, 1998
 What management gets…
Volumes

Consumer
Demand

Production Plan

Time
Tom Mc Guffry, Electronic Commerce and Value Chain Management, 1998
 What managment wants…
Volumes

Production Plan
Consumer
Demand

Time
Tom Mc Guffry, Electronic Commerce and Value Chain Management, 1998
4. Uncertainty in supply chain

• Business chooses inventory to cope with uncertainty in supply chain

• Buffer stock or extra inventory can exist in all supply chain stages

• However, inventory cost is too high. Therefore, business needs to minimize or

eliminate inventory
Bullwhip Effect

• The bullwhip effect refers to the amplification of variability in demand as you

move up the supply chain from retailers to manufacturers.
• When a retailer incorrectly forecasts demand, this mistake is often magnified as
orders are sent to distributors and manufacturers, eventually leading to massive
discrepancies between inventory produced and demand.
• Bullwhip effects can lead to excess inventory, lost revenue, and overinvestment in
production.
Bullwhip Effect

Stakeholders in supply
chain often have
overreaction when
customer demand increase
or decrease suddenly
without any reasons.

Source: Roberta S. Russel, Bernard W. Taylor III, Operations Management: Creating Value
Along the Supply Chain – 7th edition. John Wigley & Sons, Inc. (2011)
Bullwhip Effect

• Solutions to Bullwhip Effect

• Information sharing, especially information about
customer demand forecast.
• When the supply chain becomes "transparency",
supply chain stakeholder can access each other's
information, thereby reducing or eliminating
uncertainty.
• Vendor managed inventory
Beer Game
• This game simulates the supply and distribution chain of a beer company.

As one of the industrial stakeholders (manufacturer, distributor, wholesaler, retailer),

your objective is to order the good quantity of beer packs each week.

• Each stakeholder has to decide the outgoing order quantity on basis of incoming order

from its own customer

• Make sure you keep a reasonable level of stock and no backlog (= late) order as they

generate costs.
CHAPTER 5: FORECASTING
 5.1 Overview of forecast
• A forecast is a prediction of what will occur in the future. A forecast of product demand is the
basis for most important planning decisions. Planning decisions regarding scheduling,
inventory, production, facility layout and design, workforce, distribution, purchasing, and so on,
are functions of customer demand. Long-range, strategic plans by top management are based
on forecasts of the type of products consumers will demand in the future and the size and
location of product markets.

• Forecasting is an uncertain process.

• Companies sometimes use qualitative forecast methods based on judgment, opinion, past
experience, or best guesses, to make forecasts. A number of quantitative forecasting methods
are also available to aid management in making planning decisions.
5.2. The strategic role of forecasting in supply chain management
Accurate forecasting determines how much inventory a company must keep at various points along its
supply chain.

• A company’s supply chain encompasses all of the facilities, functions, and activities involved in
producing a product or service from suppliers (and their suppliers) to customers (and their customers)

• Supply chain functions include purchasing, inventory, production, scheduling, facility location,
transportation, and distribution.

• Forecasts of product demand determine how much inventory is needed, how much product to make,
and how much material to purchase from suppliers to meet forecasted customer needs. This in turn
determines the kind of transportation that will be needed and where plants, warehouses, and
distribution centers will be located so that products and services can be delivered on time.
5.2. The strategic role of forecasting in supply chain management
5.2. The strategic role of forecasting in supply chain management
a) In supply chain management

• If forecasting customer demand is inaccurate:

(1) To much inventory -> high inventory cost

OR

(2) Insufficient inventory -> service quality is affected negatively (late deliveries and stockouts)

• While accurate forecasts are necessary, completely accurate forecasts are never possible.
Hopefully, the forecast will reduce uncertainty about the future as much as possible, but it
will never eliminate uncertainty. As such, all of the supply chain processes need to be flexible
to respond to some degree of uncertainty .
5.2. The strategic role of forecasting in supply chain management

a) In supply chain management

• One trend in supply chain design is continuous replenishment, wherein continuous updating
of data is shared between suppliers and customers.

- In this system, customers are continuously being replenished, daily or even more often, by
their suppliers based on actual sales.

- Continuous replenishment, typically managed by the supplier, reduces inventory for the
company and speeds customer delivery. Variations of continuous replenishment include
quick response, just-in-time (JIT), VMI (vendor-managed inventory), and stockless inventory.
5.2. The strategic role of forecasting in supply chain management
a) In supply chain management

• One trend in supply chain design is continuous replenishment, wherein continuous updating of data is
shared between suppliers and customers.

- Such systems rely heavily on accurate short-term forecasts, usually on a weekly basis, of end-use sales
to the ultimate customer. The supplier at one end of a company’s supply chain must forecast the
company’s customer demand at the other end of the supply chain in order to maintain continuous
replenishment.

- The forecast also has to be able to respond to sudden, quick changes in demand. Longer forecasts
based on historical sales data for 6 to 12 months into the future are also generally required to help
make weekly forecasts and suggest trend changes.
5.2. The strategic role of forecasting in supply chain management

b) In quality management

• Forecasting customer demand is a key to providing good-quality

service.

- Ex: When customers walk into a McDonald’s to order a meal, they do

not expect to wait long to place orders. They expect McDonald’s to
have the item they want, and they expect to receive their orders
within a short period of time. A good forecast of customer traffic
flow and product demand enables McDonald’s to schedule enough
servers, to stock enough food, and to schedule food production to
provide high-quality service.
 5.2. The strategic role of
forecasting in supply chain
management

c) In strategic planning

• The ultimate objective of strategic planning is to determine

what the company should be in the future—what markets to
compete in, with what products, to be successful and grow.

• Successful strategic planning requires accurate forecasts of

future products and markets.

• The type of forecasting method depends on time frame,

demand behavior, and causes of behavior
5.3. Components of forecasting demand
a) Time frame:

• Indicates how far into the future is forecast.

• Short-range)/ mid-range: typically for daily, weekly, or monthly sales demand for up to approximately
two years into the future, depending on the company and the type of industry. They are primarily used
to determine production and delivery schedules and to establish inventory levels

• Long-range: usually encompasses a period of time longer than two years. A long-range forecast is
normally used for strategic planning—to establish long-term goals, plan new products for changing
markets, enter new markets, develop new facilities, develop technology, design the supply chain, and
implement strategic programs.
 5.3. Components of forecasting demand
a) Time frame:

• Indicates how far into the future is forecast.

VD: Hewlett-Packard monthly forecasts for printers are constructed from 12 to 18 months into
the future, while at Levi Strauss weekly forecasts for jeans are prepared for five years into the
future.
5.3. Components of forecasting demand
b) Demand behavior

Cycle

Trend Seasonality

Three
types of
demand
behavior
5.3. Components of forecasting
demand
b) Demand behavior

Trends is a gradual, long-term up or down movement of

demand. For example, the demand for houses has followed an
upward trend during the past few decades, without any
sustained down- ward movement in the market.
 b) Demand behavior
5.3. Components of Seasonal patterns: is an oscillating movement in demand
forecasting demand that occurs periodically (in the short run) and is
repetitive.
5.3. Components of
forecasting demand
b) Demand behavior

Cycle is an up-and-down movement in

demand that repeats itself over a lengthy time
span (i.e., more than a year). For example,
new housing starts and, thus, construction-
related products tend to follow cycles in the
economy. Automobile sales also tend to
follow cycles.
 5.4
Forecasting
Process
5.5. Forecasting methods (Quantitative)
Time series methods

• use historical demand data over a period of time to predict future demand.

• Assumption: identifiable historical patterns or trends for demand over time will repeat
themselves

- Including: Moving average; exponential smoothing; and linear trend line

 Moving average
 Moving average is good for stable demand with no pronounced behavioral patterns such
as a trend or seasonal pattern.

 Moving averages are computed for specific periods, such as three months or five months,
de- pending on how much the forecaster desires to “smooth” the demand data.

 The formula for computing the simple moving average is

Moving average
Ex 1: The Heartland Produce Company sells and delivers food produce to restaurants and
catering services within a 100-mile radius of its warehouse. To enhance its competitiveness,
the manager wants to forecast customer demand.

From records of delivery orders, management has accumulated the following data for the past
10 months, from which it wants to compute three- and five-month moving averages.
 Moving average
• Solution: Let us assume that it is the end of October. The forecast resulting from either the
three- or five-month moving average is typically for the next month in the sequence,
which in this case is November. The moving average is computed from the demand for
orders for the prior three months in the sequence according to the following formula:
 Moving average

• Solution: The five-month moving average is computed from the prior five months of
demand data as follows:
 Moving average
the earlier forecasts for prior months allow us to compare the forecast with actual
demand to see how accurate the forecasting method is—that is, how well it does.
 Moving average
Both moving average forecasts in the preceding table tend to smooth out the variability
occurring in the actual data
 Weighted Moving average

• The moving average method can be adjusted to more closely reflect fluctuations in the data. In
the weighted moving average method, weights are assigned to the most recent data according
to the following formula:
 Weighted Moving average
• EX 2: The Heartland Produce Company in Example 1 wants to compute a three-month
weighted moving average with a weight of 50% for the October data, a weight of 33% for
the September data, and a weight of 17% for the August data. These weights reflect the
company’s desire to have the most recent data influence the forecast most strongly.
 Weighted Moving average

• Ex 2: The weighted moving average is computed as

• This forecast is slightly lower than our previously computed three-month average
forecast of 110 orders, reflecting the lower number of orders in October (the most
recent month in the sequence).
 Exponential smoothing
• an averaging method that reacts more strongly to recent changes in demand.

• Exponential smoothing requires minimal data. Only the forecast for the current period,
the actual demand for the current period, and a weighting factor called a smoothing
constant are necessary
 Exponential smoothing
Ex 3: HiTek Computer Services repairs and services personal computers at its store, and it makes local service
calls. they need a good forecast of demand for repairs so that they will know how many computer component
parts to purchase and stock, and how many technicians to hire.

The company has accumulated the demand data shown in the accompanying table for repair and service calls
for the past 12 months, from which it wants to consider exponential smoothing forecasts using smoothing
constants (α) equal to 0.30 and 0.50.
 Exponential smoothing (San bằng mũ)
 Solution:

- Using α = 0.3, forecast value for February and March are

 Exponential smoothing (San bằng mũ)
• Solution:

The remainder of the monthly forecasts are shown in the following table. The final forecast is
for period 13, January, and is the forecast of interest to HiTek:
 Adjusted Exponential smoothing

The adjusted exponential smoothing forecast consists of the exponential smoothing forecast
with a trend adjustment factor added to it:

The trend factor is computed much the same as the exponentially smoothed forecast. It is, in
effect, a forecast model for trend:
Adjusted Exponential smoothing

EX 4: HiTek Computer Services now wants

to develop an adjusted exponentially
smoothed forecast using the same 12
months of demand shown in the table for
Example 3. It will use the exponentially
smoothed forecast with α = 0.5
computed in Example 3 with a smoothing
constant for trend, β of 0.30.
Adjusted Exponential smoothing

• The adjusted forecast for February, AF2, is the same as the exponentially smoothed fore-
cast, since the trend computing factor will be zero (i.e., F1 and F2 are the same and T2
0). Thus, we compute the adjusted forecast for March, AF3, as follows, starting with the
determination of the trend factor, T3:
Adjusted Exponential smoothing

• This adjusted forecast value for period 3 is shown in the accompanying table, with all
other adjusted forecast values for the 12-month period plus the forecast for period 13,
computed as follows:
 Linear Trend Line

• Linear regression is a method of forecasting in which a mathematical relationship is

developed between demand and some other factor that causes demand behavior.

• The form of a linear equation:

 Linear Trend Line

• These parameters of the linear trend line can be calculated using the least squares formulas
for linear regression:
 • Ex 5: The data for HiTek Computer Services (shown in the
table for Example 12.3) appears to follow an increasing
linear trend. The company wants to compute a linear
Linear Trend trend line to see if it is more accurate than the
Line exponential smoothing and adjusted exponential
smoothing forecasts developed in Examples 3 and 4.
 Linear Trend Line
• The values required for the least squares calculations are as follows:
 Linear Trend Line
• Using these values, we can compute the parameters for the linear trend line as follows:
 Linear Trend Line
• Therefore, the linear trend line equation is

- To calculate a forecast for period 13, let x 13 in the linear trend line:

y =35.2 + 1.72 (13) =57.56 service calls

Linear Trend Line
SEASONAL ADJUSTMENTS
• A seasonal factor is a numerical value that is multiplied by the normal forecast to get a seasonally
adjusted forecast.

• One method for developing a demand for seasonal factors is to divide the demand for each
seasonal period by total annual demand, according to the following formula:

• The resulting seasonal factors between 0 and 1.0 are, in effect, the portion of total annual
demand assigned to each season. These seasonal factors are multiplied by the annual forecasted
demand to yield adjusted forecasts for each season.
SEASONAL ADJUSTMENTS
EX 6: Wishbone Farms grows turkeys to sell to a meat-processing company throughout the
year. However, its peak season is obviously during the fourth quarter of the year, from
October to December. Wishbone Farms has experienced the demand for turkeys for the
past three years shown in the following table:

2020
2021
2022
SEASONAL
ADJUSTMENTS

Solution:

• Compute the seasonal factors by

dividing total quarterly demand
for three years by total demand
across three years:
SEASONAL ADJUSTMENTS
• Identify a demand forecast for 2023. In this case, since the demand data in the table seem to
exhibit a generally increasing trend, we compute a linear trend line for the three years of data in
the table to get a rough forecast estimate:

Thus, the forecast for 2023 is 58.17, or 58,170 turkeys.

SEASONAL ADJUSTMENTS
• Using this annual forecast of demand, we find that the seasonally adjusted forecasts, SFi, for 2023
are:
FORECAST ACCURACY

• Forecast error: the difference between the forecast

and actual demand.

• Different measures of forecast error

• Mean absolute deviation (MAD)

• Mean absolute percent deviation (MAPD)

• Cumulative error,

• Average error or bias (E).

Mean absolute deviation (MAD)
The mean absolute deviation, or MAD, is
one of the most popular and simplest to
use measures of forecast error. MAD is an
average of the difference between the
forecast and actual demand, as computed
by the following formula:
Mean absolute deviation (MAD)
Ex 7: In Examples 3, 4, and 5, forecasts were developed using exponential smoothing, (alpha =
0.30 and alpha = 0.50), adjusted exponential smoothing ( alpha = 0.50, beta = 0.30), and a
linear trend line, respectively, for the demand data for HiTek Computer Services. The company
wants to compare the accuracy of these different forecasts using MAD.
Mean absolute
deviation (MAD)
• The following table shows
the values necessary to
compute MAD for the
exponential smoothing
forecast with alpha 0.30:
Mean absolute deviation (MAD)
• Using the data in the table, MAD is computed as

• The MAD values for the remaining forecasts are as follows:

Mean absolute percent deviation (MAPD)
• The mean absolute percent deviation (MAPD) measures the absolute error as a
percentage of demand rather than per period. As a result, it eliminates the problem of
interpreting the measure of accuracy relative to the magnitude of the demand and
forecast values, as MAD does.

• The mean absolute percent deviation is computed according to the following formula:
Mean absolute percent deviation (MAPD)
• Using the data from the table in Example 7 for the exponential smoothing forecast (alpha
= 0.30) for HiTek Computer Services:

• A lower percent deviation implies a more accurate forecast. The MAPD values for other
three forecasts are:
Cumulative error

• Cumulative error is computed simply by summing the forecast errors, as shown in the
following formula.

• Large +E indicates forecast is biased low; large - E, forecast is biased high.

Cumulative error
• The cumulative error for the exponential smoothing forecast ( alpha = 0.30) for HiTek
Computer Services can be read directly from the table in Example 7; it is simply the sum
of the values in the “Error” column:

• The cumulative error for the other forecasts are

Average Error

• A measure closely related to cumulative error is the average error, or bias. It is

computed by averaging the cumulative error over the number of time periods:
Comparison of Forecasts for HiTek Computer Services
FORECAST CONTROL
• Tracking signal indicates if the forecast is consistently biased high or low. It is computed
by dividing the cumulative error by MAD, according to the formula

• The tracking signal is recomputed each period, with updated, “running” values of
cumulative error and MAD. The movement of the tracking signal is compared to control
limits; as long as the tracking signal is within these limits, the forecast is in control.
FORECAST CONTROL
In Example 7, the mean absolute deviation was computed for the exponential smoothing forecast
(alpha 0.30) for HiTek Computer Services. Using a tracking signal, monitor the forecast accuracy
using control limits of 3 MADs.
Solution
• To use the tracking signal, we must recompute MAD each period as the cumulative error is
computed.
• Using MAD 3.00, we find that the tracking signal for period 2 is

• The tracking signal for period 3 is

FORECAST CONTROL
• The remaining tracking signal values are shown in the following table:
EXERCISE 1
• A manufacturing company has monthly demand for one of its products as follows:
MONTH DEMAND MONTH DEMAND
Feb 520 Jul 420
Mar 490 Aug 510
Apr 550 Sep 610
May 580 Oct
Jun 600

• Develop a three-period moving average forecast and a three- period weighted moving
average forecast with weights of 0.50, 0.30, and 0.20 for the most recent demand values,
in that order. Calculate MAD for each forecast, and indicate which would seem to be
most accurate.
EXERCISE 2
• A computer software firm has experienced the following

demand for its “Personal Finance” software package

• Develop an exponential smoothing forecast using 0.40

and an adjusted exponential smoothing forecast using

0.40 and 0.20. Compare the accuracy of the two forecasts

using MAD and cumulative error.

EXERCISE 3
• A local building products store has
accumulated sales data for 2X4 lumber (in
board feet) and the number of building
permits in its area for the past 10 quarters:

• Develop a linear regression model for these

data and determine the strength of the
linear relationship using correlation. If the
model appears to be relatively strong,
determine the forecast for lumber given ten
building permits in the next quarter.