TRL3707 EXAM
PACK 2023
QUESTIONS
AND ANSWERS
,Question Bank from Self-Assessments, Exams & Learning Outcomes
[2015-S2-EXAM SCOPE: SHORTER QUESTIONS – STUDY TB, READ SG]
STUDY UNIT 1, PAGE 2
Additional
1) Define Forecasting Ref. Par. 1.2; [Was not taken from Bowersox 2013:133-134 as indicated by the scope]
Forecasting entails predicting the demand for each type of product (stock keeping unit or SKU) during a
particular period or at a specific place. Specific definition of what is projected to be sold, when & where.
Identifies requirements for which SC must schedule inventory & operational resources.
SAQ
2) Briefly explain main difference between dependent & independent demandRef. Par. 1.2
Independent demand is when demand for an item is not related to/dependent on the demand for any other
item. It cannot be calculated, only estimated by means of a forecast. E.g. finished goods & maintenance spares.
Dependent demand describes items whose demand is related to the demand for other items. Can be calculated
i.e. no need to forecast demand E.g. subassemblies, raw & packaging materials.
3) [PEx2] LO Name three reasons why logistics forecasts are necessary ()Ref. TB 2013:134 [Forecasting Requirements]
To Support Collaborative Planning
- A collaborative forecast jointly agreed to by supply chain partners provides a common goal that can be the
basis for developing effective operating plans.
- prevents never-ending cycle of inventory excesses/shortage that happens when independently forecasted;
To Drive Requirements Planning (a.k.a sales & operations planning)
- determines inventory projections & resulting replenishment/production requirements for planning horizon
- Integrates forecasts , open orders, available inventory & production plans into a definition of periodic
inventory availability and requirements;
- Ideally operates collaboratively & interactively both internally across the firms operations & externally with
supply chain partners to develop a common and consistent plan for each time period, location & item.
To Improve Resource Management
- Use completed plan to manage critical supp. chain processes e.g. production, inventory & transportation;
- Accurate forecast & consistent definition of SC resources & constraints → Enables timely identification and
effective evaluation of trade-offs → better match requirements to resources & better resources utilisation.
LO SAQ
4) Distinguish between centralised & decentralised approach to demand forecastingRef. Par. 1.2 p4
Approach followed depends on uncertainty of demand in individual distribution centres.
Top-down (centralised): when demand is fairly stable / changes in demand appear to be uniform at all centres.
Bottom-up (decentralised): when demand in individual markets fluctuates / changes in demand in individual
markets differ.
SAQ
5) Write a formula for a basic forecast model containing all components & explain the symbolsRef. Bowersox 2013:135
Forecast Ft = forecast quantity for period t
Model: Bt = base level demand for period t
(Bt x St x T x Ct x Pt) + I St = seasonality factor for period t
= Ft T = trend component index (↑ or ↓ per time period
Ct = cyclical factor for period t
Pt = promotional factor for period t
I = irregular / random quantity
,6) [PEx3] SAQ Explain the components of a basic forecast model (6 or 12)Ref. Par. 1.2; Bowersox (2013:136-137)
• = long-term average demand
Base Demand •Represents long term average demand after removing remaining components
•Forecast for items having no seasonality, trend, cyclic or promotional components
•Annually recurring upward & downward movement in demand
•e.g. annual toy demand : low demand ¾ year with increased demand just before Xmas =
Seasonal
consumer retail seasonality
•wholesale seasonality precedes consumer demand by ± ¼ of a year
• = long-range shift in periodic sales caused by change in population or consumption patterns
•May be positive, negative or neutral in direction.
•Positive = sales ↑ over time
•Trend component influences base demand in the successive time periods.
Trends Relationship: Bt+1 = Bt x T with
Bt+1 = base demand in period t + 1;
Bt = base demand in period t;
T = periodic trend index
•If T > 1 periodic demand is ↑. If T < 1 ↓ trend
•Characterized by periodic shifts in demand lasting more than a year
Cyclic
•Cycles may be upward or downward e.g. business cycle (economic swings recession to growth)
•Demand swings initiated by a firm’s marketing activities e.g. ads, deals, promotions
Promotional •Sales increase, followed by sales decline
•Differ in that timing & magnitude are largely controlled by firm. Can be regular
•Random/unpredictable quantities that do not fit with other categories
Irregular •Random Nature = impossible to predict
•Minimise magnitude of random component by tracking & predicting other components
7) [PEx1] LO SAQ Explain the forecast process with the aid of a suitable sketch (5)Ref. Par. 1.3 & Bowersox (2013:137-139)
Selecting a Forecast Technique
Mathematical/statistical computation used to combine base, seasonal & cyclical components with elements of
promotion history into a forecast quantity. Techniques include:
- Time series modelling: sales history is a major factor
- Correlation modelling: relationships with other independent variables are the major forecast drivers.
Providing a Forecast Support System
- Includes supply chain [SC] intelligence to gather & analyse data, develop forecast and communicate it to
relevant personnel & planning systems.
- Allows consideration of external factors such as the impact of promotions, strikes, price changes.
- N.B that effective forecasting process includes a support system to facilitate maintenance, updates &
manipulation of the historical database & forecast.
Administration of the total Forecast Process
- Includes organisational, procedural, motivational & personnel aspects of forecasting and its integration int
other firm functions.
- Organisational aspect concerns individual roles & responsibilities.
- N.B. to specify these roles & responsibilities in detail when defining forecast administration function.
- If integrated forecast is desirable: specifically define each organisation's forecasting responsibility & hold it
accountable with specific metrics.
- Effective forecast administration requires:
o organizational responsibility & procedural guidelines are documented & measured;
o forecast analysts be trained in both process & input of forecasts on SC logistics operations.
, Fig 6.7: Forecast Management Process
SAQ
8) Why should the forecast process be supported by an effective administrative systemRef. Par.
?
9) [PEx4 Variations ] LO SAQ Distinguish between qualitative techniques, time series techniques and causal techniques as
[2015-S2-Scope: Read SG]
forecasting techniques for logistical purposes (3 or 4 or 7)Ref. Par. 1.4. Bowersox (2010:147-150) or (2013:139-143) Table 6.4
Qualitative
- Rely on expertise
- Costly and time consuming
- Ideal when little historical data and much managerial judgement are required
- E.g. using input from sales force as basis of forecast for new region/product
- Developed using panels, consensus meetings and surveys
- Not appropriate for supply chain forecasting because of time needed to generate SKU forecasts
Time series
- Statistical methods
- Focuses on historical patterns & pattern changes to generate forecast of future behaviour
- Historical sales data contain clear and stable relationships: identifies trends, seasonality & cyclical patterns
- Assumes that future will reflect the past
- Assumption OK for short term: most appropriate for short range
- Not sensitive to turning points: integrate with other approaches to determine when turning points will occ
- Various techniques: [see question 10 below]
Causal
- Estimates variables on the basis of the values of other independent variables (e.g. regression).
- Can effectively consider external factors and events. Therefore more appropriate for long term forecasting
aggregate demand forecasting & not individual outlets, e.g. annual or national sales
- Not suitable for instances where demand assumes irregular pattern or is extremely erratic
- Use refined & specific info regarding variables to develop relationship between lead event & forecasted ac
- Work well when a leading variable such as price can be identified
- Cause-effect relationship. If a good relationship can be identified, a prediction can be made
- Regression use the correlation between a leading/predictable event and SKU sales that depend on that eve
- A correlation assumes that forecasted sales are proceeded by some leading independent factor such as the
sale of related products
- Not particularly common for supply chain applications
- Techniques are more appropriate for long-term forecasting (thus not for individual outlets).
- Generally used to generate annual or national sales forecasts.
10) [PEx3+2 ] Name four time series techniques in forecasting. Could also ask to fully discuss (4 or 8 or 9)Ref. Par. 1.4
[2015-S2-Scope: Read SG]
Moving Average
- Uses average of the most recent period's sales
- May use any number of previous time periods
PACK 2023
QUESTIONS
AND ANSWERS
,Question Bank from Self-Assessments, Exams & Learning Outcomes
[2015-S2-EXAM SCOPE: SHORTER QUESTIONS – STUDY TB, READ SG]
STUDY UNIT 1, PAGE 2
Additional
1) Define Forecasting Ref. Par. 1.2; [Was not taken from Bowersox 2013:133-134 as indicated by the scope]
Forecasting entails predicting the demand for each type of product (stock keeping unit or SKU) during a
particular period or at a specific place. Specific definition of what is projected to be sold, when & where.
Identifies requirements for which SC must schedule inventory & operational resources.
SAQ
2) Briefly explain main difference between dependent & independent demandRef. Par. 1.2
Independent demand is when demand for an item is not related to/dependent on the demand for any other
item. It cannot be calculated, only estimated by means of a forecast. E.g. finished goods & maintenance spares.
Dependent demand describes items whose demand is related to the demand for other items. Can be calculated
i.e. no need to forecast demand E.g. subassemblies, raw & packaging materials.
3) [PEx2] LO Name three reasons why logistics forecasts are necessary ()Ref. TB 2013:134 [Forecasting Requirements]
To Support Collaborative Planning
- A collaborative forecast jointly agreed to by supply chain partners provides a common goal that can be the
basis for developing effective operating plans.
- prevents never-ending cycle of inventory excesses/shortage that happens when independently forecasted;
To Drive Requirements Planning (a.k.a sales & operations planning)
- determines inventory projections & resulting replenishment/production requirements for planning horizon
- Integrates forecasts , open orders, available inventory & production plans into a definition of periodic
inventory availability and requirements;
- Ideally operates collaboratively & interactively both internally across the firms operations & externally with
supply chain partners to develop a common and consistent plan for each time period, location & item.
To Improve Resource Management
- Use completed plan to manage critical supp. chain processes e.g. production, inventory & transportation;
- Accurate forecast & consistent definition of SC resources & constraints → Enables timely identification and
effective evaluation of trade-offs → better match requirements to resources & better resources utilisation.
LO SAQ
4) Distinguish between centralised & decentralised approach to demand forecastingRef. Par. 1.2 p4
Approach followed depends on uncertainty of demand in individual distribution centres.
Top-down (centralised): when demand is fairly stable / changes in demand appear to be uniform at all centres.
Bottom-up (decentralised): when demand in individual markets fluctuates / changes in demand in individual
markets differ.
SAQ
5) Write a formula for a basic forecast model containing all components & explain the symbolsRef. Bowersox 2013:135
Forecast Ft = forecast quantity for period t
Model: Bt = base level demand for period t
(Bt x St x T x Ct x Pt) + I St = seasonality factor for period t
= Ft T = trend component index (↑ or ↓ per time period
Ct = cyclical factor for period t
Pt = promotional factor for period t
I = irregular / random quantity
,6) [PEx3] SAQ Explain the components of a basic forecast model (6 or 12)Ref. Par. 1.2; Bowersox (2013:136-137)
• = long-term average demand
Base Demand •Represents long term average demand after removing remaining components
•Forecast for items having no seasonality, trend, cyclic or promotional components
•Annually recurring upward & downward movement in demand
•e.g. annual toy demand : low demand ¾ year with increased demand just before Xmas =
Seasonal
consumer retail seasonality
•wholesale seasonality precedes consumer demand by ± ¼ of a year
• = long-range shift in periodic sales caused by change in population or consumption patterns
•May be positive, negative or neutral in direction.
•Positive = sales ↑ over time
•Trend component influences base demand in the successive time periods.
Trends Relationship: Bt+1 = Bt x T with
Bt+1 = base demand in period t + 1;
Bt = base demand in period t;
T = periodic trend index
•If T > 1 periodic demand is ↑. If T < 1 ↓ trend
•Characterized by periodic shifts in demand lasting more than a year
Cyclic
•Cycles may be upward or downward e.g. business cycle (economic swings recession to growth)
•Demand swings initiated by a firm’s marketing activities e.g. ads, deals, promotions
Promotional •Sales increase, followed by sales decline
•Differ in that timing & magnitude are largely controlled by firm. Can be regular
•Random/unpredictable quantities that do not fit with other categories
Irregular •Random Nature = impossible to predict
•Minimise magnitude of random component by tracking & predicting other components
7) [PEx1] LO SAQ Explain the forecast process with the aid of a suitable sketch (5)Ref. Par. 1.3 & Bowersox (2013:137-139)
Selecting a Forecast Technique
Mathematical/statistical computation used to combine base, seasonal & cyclical components with elements of
promotion history into a forecast quantity. Techniques include:
- Time series modelling: sales history is a major factor
- Correlation modelling: relationships with other independent variables are the major forecast drivers.
Providing a Forecast Support System
- Includes supply chain [SC] intelligence to gather & analyse data, develop forecast and communicate it to
relevant personnel & planning systems.
- Allows consideration of external factors such as the impact of promotions, strikes, price changes.
- N.B that effective forecasting process includes a support system to facilitate maintenance, updates &
manipulation of the historical database & forecast.
Administration of the total Forecast Process
- Includes organisational, procedural, motivational & personnel aspects of forecasting and its integration int
other firm functions.
- Organisational aspect concerns individual roles & responsibilities.
- N.B. to specify these roles & responsibilities in detail when defining forecast administration function.
- If integrated forecast is desirable: specifically define each organisation's forecasting responsibility & hold it
accountable with specific metrics.
- Effective forecast administration requires:
o organizational responsibility & procedural guidelines are documented & measured;
o forecast analysts be trained in both process & input of forecasts on SC logistics operations.
, Fig 6.7: Forecast Management Process
SAQ
8) Why should the forecast process be supported by an effective administrative systemRef. Par.
?
9) [PEx4 Variations ] LO SAQ Distinguish between qualitative techniques, time series techniques and causal techniques as
[2015-S2-Scope: Read SG]
forecasting techniques for logistical purposes (3 or 4 or 7)Ref. Par. 1.4. Bowersox (2010:147-150) or (2013:139-143) Table 6.4
Qualitative
- Rely on expertise
- Costly and time consuming
- Ideal when little historical data and much managerial judgement are required
- E.g. using input from sales force as basis of forecast for new region/product
- Developed using panels, consensus meetings and surveys
- Not appropriate for supply chain forecasting because of time needed to generate SKU forecasts
Time series
- Statistical methods
- Focuses on historical patterns & pattern changes to generate forecast of future behaviour
- Historical sales data contain clear and stable relationships: identifies trends, seasonality & cyclical patterns
- Assumes that future will reflect the past
- Assumption OK for short term: most appropriate for short range
- Not sensitive to turning points: integrate with other approaches to determine when turning points will occ
- Various techniques: [see question 10 below]
Causal
- Estimates variables on the basis of the values of other independent variables (e.g. regression).
- Can effectively consider external factors and events. Therefore more appropriate for long term forecasting
aggregate demand forecasting & not individual outlets, e.g. annual or national sales
- Not suitable for instances where demand assumes irregular pattern or is extremely erratic
- Use refined & specific info regarding variables to develop relationship between lead event & forecasted ac
- Work well when a leading variable such as price can be identified
- Cause-effect relationship. If a good relationship can be identified, a prediction can be made
- Regression use the correlation between a leading/predictable event and SKU sales that depend on that eve
- A correlation assumes that forecasted sales are proceeded by some leading independent factor such as the
sale of related products
- Not particularly common for supply chain applications
- Techniques are more appropriate for long-term forecasting (thus not for individual outlets).
- Generally used to generate annual or national sales forecasts.
10) [PEx3+2 ] Name four time series techniques in forecasting. Could also ask to fully discuss (4 or 8 or 9)Ref. Par. 1.4
[2015-S2-Scope: Read SG]
Moving Average
- Uses average of the most recent period's sales
- May use any number of previous time periods
