Basics of Supply Chain Managment (Lesson 2)

UUnit nit 11 SSupply upply CChain hain M anagement BBasics asics Management Lesson 2 Forecasting Introduction

Unit 1

Basics of Supply Chain Management

Preface............................................................................................................3 Course Description................................................................................................................. 3

Lesson 2 – Forecasting Introduction ...............................................................4 Introduction and Objectives.................................................................................................. 4 Factors that Influence Demand............................................................................................. 4 Patterns of Demand................................................................................................................ 5 What to Forecast .................................................................................................................... 7 Forecasting Principles............................................................................................................ 7 Data Collection ....................................................................................................................... 8 Forecasting Techniques ......................................................................................................... 9 Moving Averages.................................................................................................................. 11 Exponential Smoothing ........................................................................................................ 11 Seasonality............................................................................................................................. 12 Forecast Accuracy ................................................................................................................ 14 Gathering Forecast Information......................................................................................... 17 Summary ............................................................................................................................... 18 Further Reading ................................................................................................................... 18 Review ................................................................................................................................... 19 What’s Next? ........................................................................................................................ 21

Appendix.......................................................................................................22 Answers to Review Questions .............................................................................................. 23

Glossary ........................................................................................................25

© Copyright Leading Edge Training Institute Limited

ii

Unit 1


Unit 1

Basics of Supply Chain Management Preface Course Description This document contains the second lesson in the Basics of Supply Chain Management unit, which is one of five units designed to prepare students to take the APICS CPIM examination. The Basics of Supply Chain Management unit provides the foundation upon which the other four units build. It is necessary to complete this unit, or gain equivalent knowledge, before progressing to the other units. The five units, which together cover the CPIM syllabus, are: Basics of Supply Chain Management Master Planning of Resources Detailed Scheduling and Planning Execution and Control of Operations Strategic Management of Resources Please refer to the preface of Lesson 1 for further details about the support available to you during this course of study. This publication has been prepared by E-SCP under the guidance of Yvonne Delaney MBA, CFPIM, CPIM. It has not been reviewed nor endorsed by APICS nor the APICS Curricula and Certification Council for use as study material for the APICS CPIM certification examination.


3

Unit 1

Basics of Supply Chain Management Lesson 2 – Forecasting Introduction Introduction and Objectives Before planning production, it is necessary to estimate what conditions will exist in the near future. Most firms cannot wait until orders are received before they start planning production: they must anticipate future demand. This lesson looks at the factors influencing demand and the principles and techniques of forecasting demand. On completion of this lesson you will be able to: Identify factors that influence demand Recognize basic demand patterns Describe basic forecasting principles Explain the principles of data collection Compare and contrast basic forecasting techniques Define seasonality and the seasonal index Identify possible sources of and types of forecast error

Factors that Influence Demand Many factors influence demand. Often, it is not possible to identify all of them, or the effects they have. Some of the major demand influences include Business and economic conditions Competition Market trends Company plans for products, pricing and promotion. Other factors that affect demand in some situations include government or health regulations, climate conditions, seasonality, and population demographics. For example, a reasonably wealthy country that is experiencing a ‘baby boom’ may have increased demand for nurseryrelated and pre-school education products. In this case, the birth rate is a factor influencing demand. “Those who ignore the past are condemned to repeat it.” George Santayana

Example ABC Beverages has recorded the demand history for its premium freshly squeezed orange juice in the first quarter of 2003 (see Figure 1 below), which shows an abnormal spike in demand for February. Normal demand for the product remains steady at around 50000 litres per month. However, actual demand ‘spikes’ in February. This is mainly due to the success of a 6 week promotional © Copyright Leading Edge Training Institute Limited

4

Unit 1

Basics of Supply Chain Management period starting in February during which the company ran a ‘Buy 2 get 3rd free’ campaign. This is responsible for an increase of 30,000 litres in February and 15,000 in March. Orange Juice Demand Data 100000 80000 Special Promotion

Litres

60000

Seasonal Variation

40000

Trend Factor

20000

Normal Demand

0 -20000

Jan

Feb

Mar

F03

Figure 1 Freshly Squeezed Orange Juice Demand Data

The chart above also shows the affects of seasonal variation on demand for orange juice which has a negative effect in January and February, as demand usually drops in those two months. In March, seasonal demand usually increases. Sources of Demand It’s important to identify and monitor all sources of demand. These vary from industry to industry. It is easy to overlook lesser sources of demand when concentrating on the main customer. Other sources of demand include: Spare parts, for example, exhaust pipes in the car industry Promotions : for example, ‘buy one get one free’ promotion for baby wipes Intracompany demand: for example, a beverage concentrate manufacturing facility in England is unable to meet demand for several months. A plant in the same group, based in Mexico is able to produce what is required and ship over the product.

Patterns of Demand The best way to identify patterns of demand is to plot demand in a graph against a time scale. It will then be easy to visually identify demand shapes or consistent patterns of demand. Although actual demand varies, there are several underlying demand factors that often have a measurable effect on demand, depending on the type of product. These are: Trends Seasonality Random Variation Cycle The chart below shows a historical demand pattern. It shows quite large variations in demand. There are also clear patterns of demand. © Copyright Leading Edge Training Institute Limited

5

40

Unit 1

Basics of Supply Chain Management Seasonal Variation

35 30 25

Trend

20 15 10 5 Nov

Sep

Jul

May

Mar

Jan

Nov

Sep

Jul

May

Mar

Jan

Nov

Sep

Jul

May

Mar

Jan

0

Trends The red line on the chart shows the underlying trend which is gradually increasing from year to year. This is a linear trend. There are other types, for example, exponential or geometric. Seasonal Variation The demand pattern shows a recurring pattern of demand fluctuation based on the time of year. According to this data sales are always highest in mid-summer and at their lowest in January. Seasonality is most often apparent on a yearly basis like this. However, it can also occur on a weekly or daily basis. For example, demand in a restaurant will fluctuate depending on the time of day and day of the week. Many restaurants have their greatest demand on Friday and Saturday nights. The reasons for seasonality are varied. The weather, holiday seasons, or special events may all play a part. For example, demand for accommodation will increase dramatically in a city that is hosting a festival or major sporting event. Random Variation Many other factors affect demand during specific periods. These occur randomly. The actual demand may still remain quite close to the underlying patters of seasonal variation. In other points the variation may be much larger. Usually, this pattern of variation can be measured and to some extent predicted. Cycle Over several years or decades, economic cycles play a part in influencing demand. These cycles are often a slow moving but complex pattern of growth and recession which economists try to predict. Although these cycles affect demand, it is quite difficult to predict such cycles. Stable and Dynamic Demand Patterns When the general shape of the demand pattern for a particular product remains the same over time, the pattern is said to be stable. If the shape of demand is changing, it is called a dynamic demand patter. The more stable the demand, the easier it is to forecast.


6

Dependent and Independent Demand Dependent demand occurs when the demand for the product is derived from the demand for another product. For example, the sale of ice-cream cones and wafers is dependent on the sale of ice-cream. The sale of mobile phone chargers is dependent on the sale of certain types of mobile phone. It is not usually necessary to forecast demand for dependent items as this can be calculated from the forecast of the product they are dependent on. Independent items are usually end items of finished goods. However, this category also includes service parts and inter-company transfers where items are supplied to other plants in the same company. All independent demand items must be forecast.

1. All of the following have a measurable effect on demand except: A. Trends B. Seasonality Review Q

C. Random variation D. Gut feel

What to Forecast At each level of business planning the forecast requirements differ because the information needed to plan the business differs. For example, a detailed forecast of the amount of raw material required daily for the next 3 months will be of little use when formulating a strategic plan of where the business needs to go in the next 5 years. The following table links each level of business planning with the most appropriate time frame and forecast. Forecast

Time Frame

Strategic Business Plan

Market direction

Between 2 and 10 years

Production Plan

Product groups

Between 1 and 3 years

Master Production Schedule

End items and options

Months

Forecasting Principles There are four basic principles of forecasting which help to ensure more effective use of forecasts. These four principles are explained in the following paragraphs. Forecasts are usually wrong. Errors are inevitable and are to be expected. Even a forecast that is correct on average may be inaccurate over each period.


7

Unit 1


Unit 1

Basics of Supply Chain Management Forecasts should include an estimate of error. Accepting the first principle that forecasts are usually wrong, it’s important to find out by how much the forecast is in error. Every forecast should include an estimate of error. This is often a percentage of the forecast. The forecast may also be expressed as a range set between maximum and minimum values. Statistical estimates of forecast error can be made by studying the variability of demand about the average demand. Forecasts are more accurate for product groups . Even when a product group has very stable demand characteristics, the demand for the individual items of the group can vary a lot and is often random. For example, it is easier to estimate the average speed of all participants in a 100m sprint than to predict the exact finishing time of each runner. Fast running times will average out with slower times. Forecasts are generally more accurate for large groups of items than for the individual items in a group. In manufacturing, products are usually grouped in families according to the similarity of the processes and equipment used to make them. For example, a sweet factory will have different processes for boiled sweets and chocolates. Although there may be several varieties of boiled sweets, they are all made using a similar process and are a logical product group. The further into the future you forecast, the greater the inaccuracy of your forecast. Forecasts are generally more accurate for nearer time periods. The further into the future we look, the greater the uncertainty. Mainly this is due to the fact that we have more information about the near future. A company will know what promotional events are running in the near future. It will have a good understanding about the current economic climate and customer sentiment and will be able to extrapolate a fairly accurate picture of the near future from the current demand.

Data Collection TRAP Data As forecasts are usually based on judgements or calculations made from historical data it is essential to ensure high data integrity. It’s important that the data collected for historical demand follows the TRAP principles. The data must be Timely, Relevant, Accurate and appropriately Positioned. imely

Information must be available in time to be useful in decision- making. For example, a production line may fill in records on how much material was used in production. But if they are slow to record wasted materials, the system assumes the waste material is still available at the production point and therefore an insufficient reorder quantity would be specified.

elevant

Information should be as concise and relevant as possible and, where possible, in a consistent format.

ccurate

Most users of computer systems assume the data is accurate. Therefore it is important to ensure that the data is accurately entered and stored to avoid expensive mistakes as a result of inaccurate data.

ositioned

If the data is not stored in an area that is easily accessible by those who require the data it is unlikely to be timely and may be overlooked entirely.


8

Unit 1

Basics of Supply Chain Management Data Integrity There are numerous ways in which error can be introduced into company systems as a result of delayed or inaccurate data entry. More recent developments in data storage and transmission, such as bar coding and electronic data interchange (EDI) have helped improve data integrity. Bill of Material Error: A substitution may occur on any given BOM. If the change is not updated, the recorded amount of both the original component and the substituted component held in inventory will be incorrect. Work Order Error: When a Work Order (WO) is released, the Bill of Material (BOM) for that work order is locked at the time of WO release. Subsequent changes to the BOM must also be updated in the WO to maintain accurate records. Time Delays: Delays in updating data may affect the ability to cycle count correctly. In consequence, incorrect stock record adjustments may be performed. For example, a delay in scrapping material, the system may suggest material is available that has already been consumed in manufacturing. Data Entry Error: These occur particularly with manual data entry. For example, entering receipt of 1010 units instead of 1100 will introduce errors into the system that will impact inventory accuracy and planning. Data Collection Principles There are three important guidelines to consider when collecting data for forecasts: Record the data in the same format required by the forecast. If the purpose is to forecast demand on production, data based on demand, not shipments will be required. Shipments show how production responded to incoming orders but this is not a true indicator of demand as production may have under or over produced. The forecast period should be the same as the schedule period and the items in the forecast should be the same as those controlled by manufacturing. Record the circums tances related to the data. Record details of external events such as sales promotions, weather conditions or public holidays if they have a noticeable effect on the demand. Record the demand separately for different customer groups. Each customer group will have its own characteristics. For example, a busy city retailer may make several orders for a product in one week while a smaller outlet may only require one order a fortnight.

Forecasting Techniques There are many different ways to forecast. However, they fall into one of two categories: Qualitative forecasting Quantitative forecasting Qualitative Forecasting Qualitative forecasting relies on the experience and judgement of the people involved in the forecasting process. Future estimates are based on subjective assessments, intuition, and informed opinion, as, for example, in the Delphi method, which relies on the opinion of a panel of experts. These techniques are used to forecast business trends and potential demand for new © Copyright Leading Edge Training Institute Limited

9

Unit 1

Basics of Supply Chain Management products. They may be used extensively in medium and long range forecasting but are less appropriate for detailed production and inventory forecasting. Qualitative forecasting is useful where there is no reliable historical trend to work from, such as in very dynamic and changeable markets or when introducing a new product. Quantitative Forecasting In contrast, quantitative forecasting is based on mathematical formulae using historical data. Quantitative techniques are strongly influenced by the historical demand trends and are therefore most useful where extensive demand history is available and the demand is relatively stable. Both intrinsic and extrinsic factors may be assessed when using quantitative forecasting. These factors are described below. Extrinsic Technique s Extrinsic techniques, sometimes called causal techniques, are concerned with external influencers of demand. Examples of such influencers would include the weather, the disposable income of the target market, and changes in the demographic profile of the target market. For example, demand for a magazine aimed at professional women in their early twenties will be more likely to increase in the near future if the number of women graduating is increasing and if employment is also on the increase. Intrinsic Techniques Intrinsic techniques are based on internal factors that are mostly recorded and are usually readily available in the demand history. Forecasting that is reliant on intrinsic factors assumes that what happened in the past will happen in the future. There are many methods of extrapolating past data into the near future. These are all useful for forecasting, particularly in an environment where there is little random fluctuation in demand. Quantitative Forecasting Techniques At its simplest, quantitative forecasting involves one or two assumptions or rules, for example: Demand this month will be the same as last month. This is only useful in a few cases where there is little ongoing change in demand. Demand this month will be the same as the same month last year. This is useful if demand is relatively stable year to year but exhibits seasonal variation. The difficulty with forecasting based on either of these assumptions is the strong influence of random demand. For example, during the aftermath of 9/11 a great deal of uncertainty and fear led to a drop in air travel. Demand figures for November of that year would not have been an accurate predictor of airline ticket sales in the following year. Methods that average out history to discover underlying trends help to reduce the effects of random variation. Some methods that do this include moving averages, exponential smoothing, and seasonality.


10

Unit 1

Basics of Supply Chain Management Moving Averages It is often effective simply to forecast based on average demand in the preceding period. For example, a soft drinks company may forecast demand for April equal to the average demand for January February and March. Moving averages emphasise the underlying trend and smooth out the ‘noise’ of random demand fluctuation. Jan

Feb

Mar

Apr

May

22

25

27

25

25

27

29

27

27

29

28

The graphic to the right shows an example of moving averages. The average demand for January, February and March was 25. This is entered as the estimated demand for April.

Jun

The actual demand for April turns out to be 29, higher than the projected demand. The forecast for May is set as the average of the demand for February, March, and April. Each month’s forecast is based on the average of the three preceding months.

28

The mathematical formula for moving averages is quite simple:

Moving Average =

(Sum of the demand figures) ----------------------------------(The number of demand figures)

For example: (22 + 25 + 27) -----------------3

Moving Average for April =

=

25

1. Demand figures for January to June has been given below. Enter a forecast for July based on a moving average of the previous three months.

Review Q

Jan

Feb

Mar

Apr

May

Jun

34

41

46

44

49

51

Jul

Exponential Smoothing Exponential smoothing makes the calculation of a moving average simpler and reduces the amount of data needed. It can be used as a routine method of updating item forecasts and works well for stable items, particularly those with no trend or seasonality. It is an acceptable method for short range forecasting and can detect trends but will lag them. The technique involves using an average figure and the previous month’s actual demand and applying a weight factor, or smoothing constant to each figure before calculating the forecast demand. The formula for exponential smoothing is: New forecast =

old forecast

+

weighting factor(actual demand – old demand)

The weighting factor is often called alpha and is represented by the symbol ? © Copyright Leading Edge Training Institute Limited

11

Unit 1

Basics of Supply Chain Management The following table calculates the new forecast for a series of periods using exponential smoothing with a weighting factor of 0.2. Period

Old Forecast (OF)

Actual Demand (AD)

Weighting Factor: 0.2 (AD-OF)

New Forecast

1

4000

4400

80

4080

2

4080

3400

-136

3944

3

3944

2200

-348

3596

4

3596

5400

360

3956

5

3956

4200

48

4004

Table 1 Exponential Smoothing Example

2. Using the data from Table 1 above, calculate the new forecast for period 5, assuming the weighting factor has changed to 0.4 before the end of period 4.

Review Q

Period

Old Forecast

Actual Demand

5

3956

4200

New Forecast

Seasonality Seasonal demand patterns are evident in many consumer products. In summer months, the sale of sunglasses, suncream, cold drinks, and garden furniture tends to increase. During colder months, the demand for oil and electricity increases as the need for heat and light increases. Seasonality also refers to more frequently recurring demand patterns. Supermarket and restaurant sales are often highest at weekends and coming up to certain holidays. Canteens and cafes experience peak demand for during the early morning and midday for breakfast and lunch. Seasonal Index Forecasts are made for the average demand. If seasonality exists as a factor in demand, it can be calculated using the seasonal index. This is necessary in order to cut out the effects of seasonal variation so that you can compare sales in a high season with those in a low season. Seasonal Demand

Demand

Ju l Au g Se p Oc t No v De c

Ju n

Average

Ap r Ma y

Ja n Fe b Ma r

1800 1600 1400 1200 1000 800 600 400 200 0


12

The extent of seasonal variation in demand is indicated by the seasonal index, an estimate of the amount by which demand during the season will fall outside average demand. Throughout the year, demand for sunglasses might average around 1000 per month. However, the average demand in the month of June may be much higher, at 1650. Average demand for the month of October may fall to 475. The following formula calculates the seasonal index: Period average demand ---------------------------------------------Average demand for all periods

Seasonal index =

Using this formula, the seasonal index for June and October are calculated as follows: 1650 ----------1000

Index for June =

= 1.65

475 -----------1000

Index for October =

= 0.475

The period in question can be any length from daily to quarterly depending on the type of seasonal demand. The average demand for all periods is taken by totalling the demand for each period and dividing by the number of periods. The average demand for all periods is also called deseasonalized demand. 3. From the following demand data, calculate the seasonal index for each period against the average demand over the 6 months .

Review Q

Jan

Feb

Mar

Apr

May

Jun

Month

600

720

850

1100

1360

1650

Demand Seasonal Index

When the seasonal pattern is relatively stable, the seasonal index can be applied to an average demand in order to calculate a seasonal forecast using the following formula: Seasonal demand =

(seasonal index) x (deseasonalized demand)

For example, given that the seasonal index for June is 1.65, if we have predicted total demand for next year to be 13200, that’s an average demand of 1100 for each period. We can then calculate seasonal demand for June of next year as follows: June demand =

( 1.65 ) x ( 1100)

=


1815

13

Unit 1


Unit 1

Basics of Supply Chain Management 4. Using the seasonal indices you calculated in the last exercise, determine the seasonal demand for next year, given that the deseasonalized demand is 1100. Review Q

Jan

Feb

Mar

Apr

May

Jun

Month Seasonal Demand

Forecast Accuracy It is commonly accepted that the forecast will never be exactly right. Even if the overall average demand for a product group is accurately predicted over the year, the breakdown of demand for each product in the group may be quite far out and the actual demand each month may vary significantly from the average demand. This poses a problem when actual demand exceeds forecast demand as it may affect customer service. Most companies hold safety stock to ensure against stockouts when demand is higher than forecast. The forecast can be wrong in two ways: either through random error or forecast bias. Random Error When a forecast had random errors the actual demand will vary above and below the average demand for the year but the total variation from the average will be close to zero. Random variation such as this can be measured using mean absolute deviation (MAD) which is covered in a later lesson. Once the random variation is known it is possible to: Judge the reasonableness of the error. Make plans to accommodate for expected error. Set appropriate safety stock levels. Forecast Bias When a forecast has a persistent tendency to err in a particular direction it is said to be biased. In the chart below, the forecast shows a positive bias; it is nearly always higher than the actual demand. This can be due either to bias on the part of the forecaster or bias built into the business process. It is more likely that the bias is due to the forecaster if the error is in one direction for all items. However, if the error is in one direction for a specific set of items over a period of time it may be due to the business process.


14

Unit 1

Basics of Supply Chain Management 1600 1400 1200 Actual Demand 1000

Forecast Demand

800 600 400 Jan

Feb

Mar

Apr

May

Jun

Fixing Forecast Bias Often, subjective bias on the part of the forecaster is introduced in order to safeguard against certain issues. For example, the forecast may be increased to match performance objectives within the forecaster’s functional area. It may be adjusted to create a higher safety stock in response to problems in production. Usually, the bias tends to increase inventories, which leads to a high risk of inventory obsolescence and carries associated costs of storing, managing, and insuring such inventory. When subjective bias of this kind has been identified, the simplest remedy may simply be to reduce all the forecast figures by a percentage. The exact percentage may be determined by examining historical forecast accuracy. In some cases, forecast bias may be built into the process for specific products. For examp le, if the business process has ignored increased growth trends in a particular product group, the forecast will tend to be consistently low for that product group. Correcting process bias can be complex and time-consuming. Each item must be examined to identify the cause of the bias and the process must then be adjusted to correct this bias. Tracking Forecast Accuracy An accurate forecast of demand is important to ensure efficient allocation of resources within an organization. Inaccuracies in the demand forecast will cause problems at all levels of the organization and may impact customer service. It is particularly important that detailed short-term forecasts used for tactical and operational planning are accurate as errors here will increase inventory and potentially lose sales and customers. One way to measure forecast accuracy is to examine its converse concept: forecast error. To calculate the forecast error, examine the forecast and actual demand figures for each SKU and calculate the amount by which the forecast figure was in error. In the table below, the forecast error for each SKU and the total forecast error were calculated by subtracting the forecast figure from the forecast figure and recording the absolute value. © Copyright Leading Edge Training Institute Limited

15

Unit 1

Basics of Supply Chain Management FrescaJuice Forecast Accuracy F03 Actual

Forecast Ltrs

Actual Ltrs

Absolute Error

Jul-03

Jul-03

Jul-03

Orange Juice Grapefruit Juice Breakfast Juice

2,000

1,920

80

800

750

50

550

700

150

Lemon Juice Cranberry Juice

200

150

50

600

640

40

Apple Juice

900

1,300

400

5,050

5,460

410

Total

Table 2 Absolute Error

When you divide the absolute error figure by the actual demand and multiply by 100, you see the forecast error as a percentage of the total demand. Table 3 below displays the absolute error as a percentage of the actual demand for each SKU.

FrescaJuice Forecast Accuracy F03 Actual

Forecast Ltrs

Actual Ltrs

Absolute Error

% Error

Jul-03

Jul-03

Jul-03

Jul-03

Orange Juice Grapefruit Juice Breakfast Juice

2,000

1,920

80

4

800

750

50

7

550

700

150

21

Lemon Juice Cranberry Juice

200

150

50

33

600

640

40

6

Apple Juice

900

1,300

400

31

5,050

5,460

410

8

Total

Table 3 Forecast error as a percentage of actual demand

Absolute(Actual - Forecast) % Forecast Error =

100 x Actual Demand


16

Unit 1

Basics of Supply Chain Management Gathering Forecast Information The forecast, as an estimate of future demand can be determined in many ways: using historical data and mathematical formulae, using subjective opinion and informal sources, or any combination of these approaches. The forecast may use data from inside the company such as past sales or orders received in each period. This information can be projected into the future taking into account growth factors or economic trends, to achieve a forecast estimate. Many companies gather externa l information to assist in the forecasting process, such as market surveys and market research. The three main areas of research are market intelligence, market changes, and market demand. Such research involves consulting with the market to identify what it believes it wants. Methods include street polls, supermarket stands to gauge reaction to a product and focus groups. Market Intelligence This approach involves comparing intelligence of the market, gathered wherever possible, with the statistical forecast to identify if any changes must be made. This may be an individual or cross- functional team responsibility. Knowing what people want to buy is essential to the business of forecasting. Market Changes Market changes may be temporary, for example as the result of promotions by an organization or its competitors, or more permanent, for example, changes in government regulations that impact on product demand as in the UK where beef on the bone was banned as a result of BSE fears. Market Demand Market demand is the total volume that will be bought by a defined customer group, in a specified location, during a particular period of time under specific environmental conditions and marketing effort. A shift in market demand can often be detected by market surve ys and research. A typical example is the clothing industry where basic demand changes with each season.


17

Unit 1

Basics of Supply Chain Management Summary Lesson 2 covered the factors influencing demand and the principles and techniques of forecasting demand. You should be able to: Identify factors that influence demand Recognize basic demand patterns Describe basic forecasting principles Explain the principles of data collection Compare and contrast basic forecasting techniques Define seasonality and the seasonal index Identify possible sources of and types of forecast error

Further Reading Introduction to Materials Management, JR Tony Arnold, CFPIM, CIRM and Stephen Chapman CFPIM APICS Dictionary 10th edition, 2002


18

Unit 1

Basics of Supply Chain Management Review The following questions are designed to test your recall of the material covered in lesson 2. The answers are available in the appendix of this workbook. 6. The following are major influences on a firm’s demand for product and services except: A. Master Production Schedule B. General business and economic trends C. The firm’s promotional activities D. Market trends 7. All of the following are fundamentals of forecasting except: A. Forecasts are generally inaccurate B. Forecasts for sub-assemblies are more accurate C. Forecasts are more accurate in the near term D. Forecasts should include an estimate of error 8. When a company has to rely on external indicators when forecasting, the forecasting technique for calculating the data is called: A. Qualitative forecasting B. Extrinsic forecasting C. Intrinsic forecasting D. Causal forecasting 9. Which forecasting technique uses the following formula: New forecast = old forecast + ? (old forecast – actual demand)? A. Weighted moving average B. Seasonal index C. Exponential smoothing D. Focus forecasting 10. In the month of June a product sells 300 units. The product in question has an annual demand of 2400. What is the seasonal index for this product for June ? A. 1.0 B. 1.5 C. 1.75 D. 2.0 © Copyright Leading Edge Training Institute Limited

19

Unit 1

Basics of Supply Chain Management 11. Which is the best description of forecast bias? A. A forecast has a persistent tendency to err in a particular direction B. The standard deviation is consistently positive C. The mean absolute deviation (MAD) = the forecast error D. The sum of the errors is less than the MAD 12. Tracking forecast accuracy is useful for A. Monitoring the quality of the forecast B. Determining the variation in the production plan C. Measuring whether the schedule is being met D. Measuring the material plan


20

Unit 1

Basics of Supply Chain Management What’s Next? Lesson 2 covered a variety of techniques and tools that can be used to improve forecasting. At this point you have completed two of the 10 lessons in Unit 1. You should review your work before progressing to the next lesson which is: Supply Chain Management Basics – Lesson 3 Master Planning.


21

Unit 1

Basics of Supply Chain Management Appendix


22

Unit 1

Basics of Supply Chain Management Answers to Review Questions Lesson 2 Review 1. D Gut Feel A gut feeling is an internal hunch or judgment made about demand. It does not have any effect on demand. 2. Moving Average for July

Jan

Feb

Mar

Apr

May

Jun

Jul

34

41

46

44

49

51

48

This was calculated by dividing the sum of demand for April, May and June by 3 3. New forecast for period 5, assuming the weighting factor has changed to 0.4 before the end of period 4.

Period

Old Forecast

Actual Demand

New Forecast

5

3956

4200

4054

This was calculated by multiplying the difference between forecast and actual demand by the weighting factor of 0.4 and adding this to the old forecast figure. 4. Seasonal index for each period against the average demand over the 6 months.

Jan

Feb

Mar

Apr

May

Jun

Month

600

720

850

1100

1360

1650

Demand

0.6

0.72

0.85

1.1

1.36

1.65

Seasonal Index

The seasonal index for each month is calculated by dividing the average demand for the month by the average demand over the entire season. 5. Seasonal demand for next year based on deseasonalized demand of 1100.

Jan

Feb

Mar

Apr

May

Jun

Month

660

792

935

1210

1496

1815

Seasonal Demand

This is calculated by multiplying the deseasonalized demand by the seasonal index for each month. © Copyright Leading Edge Training Institute Limited

23

Unit 1

Basics of Supply Chain Management 6. A The Master Production Schedule (MPS) is driven by market demand (as set down in the forecast and production plan). It does not influence market demand. 7. B Forecasts are most accurate at the aggregate level and tend to be less accurate for subassemblies. For this reason, it is important to forecast at the product group level rather than the sub-assembly level. 8. B Extrinsic forecasting relies on external factors. An extrinsic forecast is based on external factors that will influence demand. For example, the number of new houses built will impact on the demand for flooring. Extrinsic forecasts are useful for large aggregations such as total company sales. 9. C Exponential smoothing uses a smoothing constant or weighting factor, often called alpha (? ). The alpha factor smoothes variation between latest actual demand and forecast demand. 10. C To calculate the seasonal index, you divide the period average demand by the average demand for all periods in the season. In this example, the average demand for all periods in the season is 200, so the seasonal index for June is 300 / 200 or 1.5. 11. A Forecast bias is evident when actual demand varies consistently higher or lower than the forecast. When bias occurs in the forecast the forecast is incorrect and must be adjusted. 12. A A tracking signal is used to measure the quality of the forecast and determine whether to adjust the forecast. There are many methods of tracking forecast accuracy, including forecast error as a percentage of demand.


24

Unit 1

Basics of Supply Chain Management Glossary Term

Definition

bill of material (BOM)

A listing of all the subassemblies, intermediates, parts, and raw materials needed for a parent assembly, showing the required quantity of each. It is used with the MPS to determine items that must be ordered. Also called formula or recipe.

Delphi method

A qualitative forecasting technique where the opinions of experts are combined in a series of iterations. The results of each iteration are used to develop the next, so that convergence of the experts' opinion is achieved.

dependent demand

Demand that is directly related to or derived from the bill of material structure for another item or end product. Dependent demand should be calculated rather than forecast. Some items may have both dependent and independent demand at the same time.

exponential smoothing

A weighted moving average forecasting technique in which past records are geometrically discounted according to their age with the heaviest weight assigned to most recent data. A smoothing constant is applied to avoid using excessive historical data.

extrinsic forecast A forecast based on a correlated leading indicator, for example, estimating furniture sales based on house builds. Extrinsic forecasts are more useful for large aggregations like total company sales. independent demand

Demand for an item that is unrelated to the demand for other items. Examples include finished goods and service part requirements.

intrinsic forecast A forecast based on internal factors, such as an average of past sales. lead time

Lead time is the span of time required to perform a process.

master production schedule (MPS)

The anticipated build schedule for those items assigned to the master scheduler. The master scheduler maintains this schedule and it drives material requirements planning. It specifies configurations, quantities and dates for production.

moving average

An arithmetic average of a certain number of the most recent records. As each new record is added, the oldest record is dropped. The number of periods used for the average reflects responsiveness versus stability.

random variation

A fluctuation in data that is caused by random or uncertain events.


25

seasonality

A repetitive pattern of demand from year to year or month to month (or other time period) showing much higher demand in some periods than in others.

trend

General upward or downward movement of a variable over time, for example in product demand.

work order

an order to the machine shop for tool manufacture or equipment maintenance or an authorization to start work on an activity or product.


26

Unit 1


Basics of Supply Chain Managment (Lesson 2)

Recommend Documents