2. Demand Management

Demand Management: The function of recognizing all demands for goods and services to support the market place. It involves prioritizing demand when supply is lacking. Proper demand management facilitates the planning and use of resources for profitable business results.
It occurs in the short, medium and long term. In the log term, demand projections are needed for strategic business planning of such things as facilities. In the medium term, the purpose is to project aggregate demand for production planning. In the short run, demand management is needed for items and is associated with master production scheduling.

Most of companies spend too few effort on establishing an proactive Demand Plan but rather simply aggregate the customer forecasts into monthly buckets - or worse keep the previous one - as an input for the Supply Plan. This increase the bullwhip effects and logistics people spend most of their time chasing orders and supplies, changing priorities and increasing significantly inventory with a negative impact on the On-Time-Delivery (OTD) performance.
The result in long fulfillment lead-time resulting in order delays or even loss of Sales. Companies spend many effort on production planning during M.P.S (Master Production Schedule) and in details in the M.R.P (Manufacturing Requirement Planning) with a poor result when the demand plan is not realistic. As the result operational performance is weak and management is put under pressure.

Logistics people value added is wasted in changing daily the production planning and checking the suppliers deliveries, as well as frequent meetings trying to follow unplanned orders. The inventory turn is then low, meaning that the return on dollar invested to fulfill orders are too low and company money is wasted on funding the demand plan discrepancy. And worst of all, unnecessary inventory built or unused raw material purchased become at the end obsolete and the financial impact is a direct write-off in the accounting book.

As a summary, inefficient demand planning results in order delays - so cash collection delay, losses of sales, high inventory and obsolescence which can lead to billion of dollar wasted.

4 Steps to Demand planning
This is done following 4 steps within a monthly in a continuous process:

1.Aggregate the customer forecasts, planned orders and backlog. Review the gaps, double entries and forecast in the past or in the frozen period (usually within the standard order lead-time).

2. Review and build a draft demand plan under the demand buckets:
Determine the market opportunities : trend, seasonality, new customer expectations, evolving competition, new products introduction,...in order to find the best demand areas
Determine the optimum product mix and customer mix to achieve a sustainable growth in the target demand areas
Determine the sales & marketing tactics to stimulate the demand in target demand areas : promotion, special events, price revision, distribution channel,...

3. Communicate the Demand Plan to the organisation:
Supply organisation : review demand plan in units to review capacity, efficiency and resources requirements
Finance organisation : review the assumption, product mix and turnover
Product organisation : review the product mix, new product introduction, product life cycles and demand plan in units

4. Review the scenarios during executive meeting:
Review the assumptions taken
Adjust the priorities
Adjust the demand plan according to the scenario amended

 Demand management activities
If material and capacity resources are to be planned effectively, all sources of demands must be identified. Demand management includes 4 major activities:

- Forecasting,
- Order processing,
- Making delivery promises,
- Interfacing between manufacturing planning and control and the marketplace.

 

Order processing occurs when a customer’s order is received. The product may be delivered from finished goods inventory or it may be made or assembled to order. If goods are sold from inventory, a sales order is produced authorizing the goods to be shipped from inventory.

 If the product is made or assembled to order, the sales department must write up a sales order specifying the product. A copy stating the erms and conditions of acceptance is sent to the customer. Another copy, sent to the aster planner, is authorization to go ahead and plan for manufacture. He must know wat to produce, how much, and when to deliver

Data Collection and Forecast Techniques

Forecasts are usually based on historical data manipulated in some way using either judgment or a statistical technique. Thus, the forecast is only as good as the data on which it is based. To get good data, three principles of data collection are important.

1. Record data in the same terms as needed for the forecast. This is a problem in determining the purpose of the forecast and what is to be forecast. There are three dimensions to this:

Suppose a firm makes a bicycle that comes in three frame sizes, three possible wheel sizes, a 3-, 5-. or 10-speed gear changer, and with or without deluxe trim. In all, there are 54 (3 x 3 x 3 x 2) individual end items sold. If each were forecast, there would be 54 forecasts to make. A better approach is to forecast (a) total demand and (b) the percentage of the total that requires each frame size, wheel size, and so on. That way there need be only 12 forecasts (three frames, three wheels, five gears, and the bike itself).

In this example, the lead time to make the components would be relatively long in comparison to the lead time to assemble a bike. Manufacturing can make the components according to component forecast and can then assemble bikes according to customer orders. This would be ideal for situations where final assembly schedules are used.

2. Record the circumstances relating to the data. Demand is influenced by particular events, and these should be recorded along with the demand data. For instance, artificial bumps in demand can be caused by sales promotions, price changes, changes in the weather, or a strike at a competitor’s factory. It is vital that these factors be related to the demand history so they may be included or removed for future conditions.

3. Record the demand separately for different customer groups. Many firms distribute their goods through different channels of distribution, each having its own demand characteristics. For example, a firm may sell to a number of wholesalers that order relatively small quantities regularly and also sell to a major retailer that buys a large lot twice a year. Forecasts of average demand would be meaningless, and each set of demands should be forecast separately.

Considerations in Forecast System Design:
1. Determine the information that needs to be forecasted. This includes defining the source of the historical data to be provided & the periods over which the data will be collected.
2. Assign the responsibility for the forecast to a person whose performance will be measured on the accuracy of actual sales to the forecast.
3. Following forecast system parameters shall be set:
a) Forecast Horizon: it is a length of time into the future over which a forecast will be prepared.
b) Forecast level: the level of detail to be provided in the forecast, such as business unit, product family, subfamily, model & brand or SKU.
c) Forecast periods: the time period to be used for forecasting purposes such as years, quarters, months, weeks, days or hours.
d) Forecast Frequency: the planned manner in which the forecast will formally be reviewed & potentially revised.
e) Forecast Revision: the way in which changes to the forecast will be recorded, such as original forecast, revised forecast, subsequently revised forecast, current forecast.
4. Select appropriate forecasting models & techniques depending on the volatility of the demand for a given product or service.
5. Collect data for input to forecasting models.
6. Test models for forecast accuracy.
7. Record actual demand information against the forecast.
8. Report forecast accuracy.
9. Determine the root cause for variance bet’n forecast & actual data.
10. Periodically assess the forecast system for performance so that changes can be made to the forecasting approach where necessary.

Forecast Techniques

Qualitative Techniques
Qualitative techniques are projections based on judgment, intuition, and informed opinions. By their nature, they are subjective. Such techniques are used to forecast genera! business trends and the potential demand for large families of products over an extended period of time. As such, they are used mainly by senior management. Production and inventory forecasting is usually concerned with the demand for particular end items, and qualitative techniques are seldom appropriate.

When attempting to forecast the demand for a new product, there is no history on which to base a forecast. In these cases, the techniques of market research and historical analogy might be used. Market research is a systematic, formal, and conscious procedure for testing to determine customer opinion or intention. Historical analogy is based on a comparative analysis of the introduction and growth of similar products in the hope that the new product behaves in a similar fashion. Another method is to test-market a product.

There are several other methods of qualitative forecasting. One, called the Delphi method, uses a panel of experts to give their opinion on what is likely to happen. Few Qualitative techniques are given below
1. delphi method: forecast is developed by a panel of experts who anonymously answer a series of questions; responses are fed back to panel members who then may change their original responses
- very time consuming and expensive
- new groupware makes this process much more feasible

2. market research: panels, questionnaires, test markets, surveys, etc.

3. product life-cycle analogy: forecasts based on life-cycles of similar products, services, or processes

4. expert judgement by management, sales force, or other knowledgeable persons

Quantitive Techniques


Quantitative forecasting methods represent the relationship demand and one or more independent variables
. Using quantitative forecasting is more objective than using qualitative forecasting.

Whenever you are going to conduct a quantitative forecast, you will need to collect the historical data which is relevant to your study in order to predict future conditions. This data should be checked for anomalies by plotting it and looking for outliers. If an anomaly is identified, it should be documented and removed from the dataset.

Quantitative forecasting can be categorized into two types of models. The first type, causal models, uses independent variables instead of (or as well as) time in order to generate a forecast. The second type, time series models, creates a demand profile with time as the independent variable.
Moving average
Exponential smoothing
Regression analysis
Adaptive smoothing
Graphical methods
Econometric modeling
Life-cycle modeling

Extrinsic Techniques
Extrinsic forecasting techniques are projections based on external (extrinsic) indicators which relate to the demand for a company’s products. Examples of such data would be housing starts, birth rates, and disposable income. The theory is that the demand for a product group is directly proportional, or correlates, to activity in another field. Examples of correlation are:

Sales of bricks are proportional to housing starts.
Sales of automobile tires are proportional to gasoline consumption.

Housing starts and gasoline consumption are called economic indicators. They describe economic conditions prevailing during a given time period. Some commonly used economic indicators are construction contract awards, automobile production, farm income, steel production, and gross national income. Data of this kind are compiled and published by various government departments, financial papers and magazines, trade associations, and banks.

The problem is to find an indicator that correlates with demand and one that preferably leads demand, that is, one that occurs before the demand does. For example, the number of construction contracts awarded in one period may determine the building material sold in the next period. When it is not possible to find a leading indicator, it may be possible to use a nonleading indicator for which the government or an organization forecasts. In a sense, it is basing a forecast on a forecast.

Extrinsic forecasting is most useful in forecasting the total demand for a firm’s products or the demand for families of products. As such, it is used most often in business and production planning rather than the forecasting of individual end items.

Intrinsic Techniques
Intrinsic forecasting techniques use historical data to forecast. These data are usually recorded in the company and are readily available. Intrinsic forecasting techniques are based on the assumption that what happened in the past will happen in the future. This assumption has been likened to driving a car by looking out the rear-view mirror. While there is some obvious truth to this, it is also true that lacking any other “crystal ball,” the best guide to the future is what has happened in the past.

Since intrinsic techniques are so important, the next section will discuss some of the more important techniques. They are often used as input to master production scheduling where end-item forecasts are needed for the planning horizon of the plan.

Sources of Demand:
Many companies have multiple customers who can be classified into intermediate customers (wholesalers, retailers etc.) & ultimate customers (the end user/consumer of the product/service). When forecasting a demand it is necessary to consider all sources of demand throughout the supply chain which are:
a) Consumers: the ultimate users of the service or the product. “A person who purchases a good or service for his or her own use (not for resale)”.
b) Customers: people who will receive an invoice & pay for a product.
c) Referrers: people who prescribe or recommend products or services to others.
d) Dealers & Distributors: intermediaries who act on behalf of a supplier.
e) Intercompany: purchases from the affiliated business units of same company.
f) Service Parts: items subjected to independent demand for service reasons.
 Consumer may or may not be customers or customers may or may not be consumers. Independent demand items are the saleable products or service levels & are typically the parts that are master scheduled.

Demand management in MPC system

The position of the demand management in the MPC is shown in the above fig.
It is the key connection to the market place in the front end of the MPC system.

The Demand management is a gateway module in MPC system, providing the link to the market place, SOP & MPS. The communication between DM & market place are two way communications - gathering information from the customer & informing customer the status of the order.

The information provided to SOP is used to develop sales & operation plan covering a year or more in duration at a high level of aggregation. Both the sales order & forecast information is provided to the MPS system. It is in the MPS system that short-term, product specific manufacturing plans are developed & controlled as actual demand becomes available and information is provided to the provide delivery promises and order status to customers.

Planning & Control
The planning part of manufacturing planning & control involves determining the capacity that is required to meet the actual future demands. The control part determines how the capacity will be converted into products as the orders come in. The company executes the plan as actual demand information becomes available. The control function determines how the company will modify the plans in light of forecast errors and other change in assumptions.

Forecast & Plan
The difference between the patterns of demand & the response by the company points out the important distinction between forecast & plans. In demand management, forecasts of the quantities & timing of customer demand are developed.  These are estimates of what might occur in the market place. Manufacturing plans that specify how the firm will respond are based on these forecasts.

Exponential Smoothing


It is not necessary to keep months of history to get a moving average. Therefore, the forecast can be based on the old calculated forecast and the new data. This formula puts as much weight on the most recent month as on the old average. If this does not seem suitable, less weight could be put on the latest actual demand and more weight on the old forecast. One advantage to exponential smoothing is that the
new data can be given any weight wanted. The weight given to latest demand is called a smoothing constant (a). It is always expressed as a decimal from 0 to 1.0. 
                                        New forecast = (a) (latest demand) + (1 - a) (previous forecast)

Example
The old forecast for May was 220, and the actual demand for May was 190. If alpha (a) is 0.15, calculate the forecast for June. If June demand turns out to be 218, calculate the forecast for July.

Answer
June forecast = (0.15)(190) + (1 — 0.15)(220) = 215.5
July forecast = (0.15)(218) + (0.85)(215.5) = 215.9


Exponential smoothing provides a routine method for regularly updating item forecasts. It works quite well when dealing with stable items. Generally, it has been found satisfactory for short-range forecasting. It is not satisfactory where the demand is low or intermittent.

Exponential smoothing will detect trends, although the forecast will lag actual demand if a definite trend exists.

The above figure shows a graph of the exponentially smoothed forecast lagging the actual demand where a positive trend exists. Notice the forecast with the larger a follows actual demand more closely.

If a trend exists, it is possible to use a slightly more complex formula called double exponential smoothing. This technique uses the same principles but notes whether each successive value of the forecast is moving up or down on a trend line. Double exponential smoothing is beyond the scope of this text.

A problem exists in selecting the “best” alpha factor. If a low factor such as 0.1 is used, the old forecast will be heavily weighted, and changing trends will not be picked up as quickly as might be desired. If a larger factor such as 0.4 is used, the forecast will react sharply to changes in demand and will be erratic if there is a sizable random fluctuation. A good way to get the best alpha factor is to use computer simulation. Using past actual demand, forecasts are made with different alpha factors to see which one best suits the historical demand pattern for particular products.

Forecast Decissions

Leading Indicators:
There are several types of forecasts that are based on external indicators factors that determine the demand performance of a related item. E.g. the demand for new housing will have direct impact on the demand for the constriction materials, interior furnishings & appliance purchases. A causal factor is a factor that creates demand for a specific product.

A good example of the use of causal factors in forecasting is the development of a forecast for a crop pesticide based on causal factors. Published information defines the number of acres of a particular crop. That will be planned next season. The company then estimates the amount of corps that will be treated with pesticide. They then estimate their share of the market & determine the amount of pesticide normally used per acre.

New Product Introduction:
The challenge for a new product is to forecast demand for a product that has never been sold before. An example of the need for this kind of forecast is the area of ‘”provisioning” of aircraft spare parts. The question arises as how many spare parts to carry in each regional maintenance center to support the fleet in active service with a minimum of aircraft downtime. Before the regional jet had completed it’s flight testing, decisions had to be made about the initial & replenishment stocking levels for all spare & service parts for the entire aircraft fleet. Forecasts were made on the likely number of flight hours before various spare parts would need to be replaced.

A preventive maintenance schedule was developed based on predicted mean time between failure statistics that had been estimated by design engineering. Stocking levels were then established using safety stocks based on the potential error in the forecast. As soon as the aircraft entered active service, actual data were used to replace estimates & stocking levels were adjusted.

Some companies are introducing products that are in the same family as is a previous product, or replacement products & they can use historical analogy as a good indicator of future sales. The impact of introducing new product on the existing sales is important. When the customer is provided with choices he may buy the new model & sales. Historical analogy allows a company to minimise & manage the risk associated with the new product introduction.

Focus Forecasting:
Focus forecasting is “a system that allows the user to simulate the effectiveness of numerous forecasting techniques, enabling selection of the most effective one”. It was invented by Bernard Smith when he was working for Servistar in 1972. He was responsible for forecasting 20000 independent demand items, nuts, bolts, screws & fasteners. Focus forecasting is computer based simulation technique that compares the forecasts generated using any of 14 simple strategies.

The computer selects the best strategy to forecast this item at this moment in time. It does this through a process of simulation. The computer goes back through 3 periods at a time & pretends they did not happen & then projects the 3 periods using each of the simple strategies. Whichever strategy gives best fit to the actual results becomes the strategy selected to forecast this item at this point in time. The simulation can result in different strategies being applied to an item every period.

Following are the assumptions of focus forecasting:
a) The most recent past is the best indicator of the future.
b) One forecasting model is better than others.
c) All forecasting models will be compared for all items forecasted.
d) Recent history will be forecasted for each item with each model.
e) The model which achieves the closest fit to actual product sales will be selected.
f) This model will be used to forecast this item this time.
Focus forecasting still used in Servistar, which forecasts 2,50,000 SKUs every month.

Data Issues for forecasting:
The accuracy of the forecast depends upon the following factors which are involved in data collection:
a) Availability of data: It is the access to historical information concerning the level of sales activity for a certain item. It is important to preserve the sales history when new systems are implemented.
b) Consistency of data: the data collected should be comparable from period to period. It is critical that the data are captured in the same terms. E.g. the item nos. should be the same over the entire period from which the sales information is being collected.
c) Amount of history: A product that is subject to seasonal demand at least to seasons of demand are required to establish the amount of seasonal activity.
d) Forecast frequency: It is the periodic interval bet’n the successive generation of the forecast. Mostly it is a month.
e) Cost & Time Issues: It refers to the cost associated with the generation of the forecast. There is an inherent cost associated with the acquisition, storage & processing of forecast information. This cost must be compared with the value of the forecast. There is also a time factor associated with a larger database; it takes more processing time to get results. The forecasting system needs to perform within prespecified cost & time standards.
f) Recording true demand: it is actual demand that is recorded for a product or service as opposed to the dates on which the company can support the customer’s request. True demand occurs at the level of the ultimate customer-the consumer or end user of the product or service.
g) Order date vs. ship date: When there is a significant lead time from requirement to delivery this information becomes very critical. The organization needs to develop 2 forecasts one at the level of order input received (a bookings forecast) & the other at the level of anticipated shipment (shipment forecast). This information also relates to the quantity ordered vs. the quantity shipped.
h) Product Unit vs. Financial Units: Sales & operations do not speak the same language.
i) Level of aggregation: It refers to the use of different forecasts & data at different levels within the organization. The level of overall business activity is expressed in the business plan, groups of products are expressed in the SOP & detailed schedules are expressed as specific units.
j) Custom Partnering: It improves communications thereby increasing accuracy of forecast. When the customers & the suppliers are partners the information is shared.

Planning Horizons & Time Periods:
a) Calendar: Mostly the future sales are forecasted by a month. If the sales forecasts by calendar month & operation forecasts by the number manufacturing days in a given month or quarter. In view of the possible variations, the need to establish an agreed upon company calendar for consistency & communication purpose is apparent.
b) Time periods & planning intervals: Common alternatives are days, weeks, months, quarters or years. The planning cycle refers to the frequency of revision to the plans are developed. Business plan is developed on an annual basis, a SOP on a monthly basis & a master production schedule on a
weekly or daily basis.
c) Planning Horizons: It is the length of time into the future for which the forecast will be generated. The planning horizon is equal to the length of the period multiplied by the number of periods. As a general rule the further into the future that the forecast is made, the less accurate the forecast will be. The forecast should cover a sufficient time for the specified planning purpose.

Data Preparation & collection:
The accuracy of source data is fundamental to the generation of sales forecasts. Collection & recording of the data is the most important aspect. The data should be collected in the same term as the forecast requires. If the forecast is expressed in weekly intervals then the data should be collected in weekly intervals.

It is important to record customer requested quantities not merely shipments since shipments are constrained by product or resource availability. Shipments may not reflect the true level of sales that were available to the company. The shipments are the measure of supply not demand.

Additionally the sales numbers do not include lost sales-sales that would have been realized if the product or service had been available at the time of customer need. It is important to estimate the level of lost sales to create a true forecast of potential sales that can be generated.

It is also important to track the original customer requested date vs. promised date. In a service organization lost sales may not be recorded. The customer can cancel the order or look for the alternative if he is dissatisfied with the service.
Forecast can be generated at many levels but will be more accurate at higher levels than at lower levels. Forecasts can be aggregated up the pyramid from the detail level to the aggregate level.

Dealing with Outliers:
“A data point that differs significantly from other data for a similar phenomenon. For example, if the average sales for a product were 10 units per month, and one month the product had sales of 500 units, this sales point might be considered an outlier”. Some forecasting system permits users to remove outliers from the data. In some cases the outliers may be a singular anomaly (a one time event) unlikely to be repeated at any time in the future. Regardless of whether or not this information is used for forecasting or sales projections including the outlier will distort the demand. The outlier may be caused by the events that have a probability of recurrence. It is always easier to remove the problem outlier than to
adjust the forecast model that may need to be changed.

Decomposition of data:
Decomposition is “a method of forecasting where time series data are separated into up to three components: trend, seasonal, and cyclical; where trend includes the general horizontal upward or downward movement over time; seasonal includes a recurring demand pattern such as day of the week, weekly, monthly, or quarterly; and cyclical includes any repeating, non seasonal pattern. A fourth component is random, that is, data with no pattern. The new forecast is made by projecting the patterns individually determined and then combining them”. Steps in decomposition are:
1. Purify the data,
2. Adjust the data,
3. Take out the baseline & components,
4. Identify demand components,
5. Measure the random error,
6. Project the series,
7. Recompose.

Forecast Error & Tracking

Monitoring forecast feedback & measuring forecast performance are parts of the forecasting process. Monitoring forecast feedback alerts the forecaster to processes that are out of control & how far they are out of control. In the concept of tracking signals a demand filter identifies errors that exceed some predetermined range or trip value. Demand filtering checks actual demand against some limit & refers the data to a person to determine whether or not action should be taken. Whichever tracking signal is used the system will generate an exception report to alert someone that there is a forecast error. It is important to know why error has occurred. The most essential element in tracking the forecast is to hold people accountable for forecast accuracy.

Forecast Accuracy:
The avg. difference between the forecast value & the actual value. The difference bet’n the actual demand
& the forecast demand.

Forecast Accuracy =  (Actual - Forecast) / Forecast

The forecast accuracy should be based on the forecast frozen at a period equal to the supply lead time.

Forecast Error:
Forecast error is the difference between actual demand and forecast demand. Error can occur in 2 ways:


Bias:

“A consistent deviation from the mean in one direction (high or low). A normal property of a good forecast is that it is not biased”.
In terms of forecasting bias is the tendency of the forecast to be either above or below the actual observations. With this concept if the computed bias is –ve the forecast is consistently too low; if the computed bias is –ve the forecast is consistently too high. The +ve & -ve errors
cancel each other out when the bias is computed. Bias is a measure of general tendency or direction of error. Bias is calculated as the total error divided by the no. of periods.

Bais exists when the cumulative actual demand varies from the cumulative forecast. This means the forecast average demand has been wrong. The forecast should be changed to improve its accuracy.

The purpose of tracking the forecast is to be able to react to forecast error by planning around it or by reducing it. Often there are exceptional onetime reasons for error. These reasons relate to the discussion on collection and preparation of data and the need to record the circumstances relating to the data;

Cumulative actual demand may not be the same as forecast. Consider the data in the Figure. Actual demand varies from forecast, and over the six-month period, cumulative demand is 120 units greater than expected.
Bias exists when the cumulative actual demand varies from the cumulative forecast. This means the forecast average demand has been wrong. In the example in the Figure, the forecast average demand was 100, but the actual average demand was 720 ± 6 = 120 units. The figure shows a graph of cumulative forecast and actual demand.
Bias is a systematic error in which the actual demand is consistently above or below the forecast demand. When bias exists, the forecast should be changed to improve its accuracy.
Errors can also occur because of timing. For example, an early or late winter will affect the timing of demand for snow shovels although the cumulative demand will be the same.
Tracking cumulative demand will confirm timing errors or exceptional one-time events. The following example illustrates this. Note that in April the cumulative demand is back in a normal range


Random variation: In a given period, actual demand will vary about the average demand. The difference are random variations.
The variability will depend upon the demand pattern of the product. Some products will have a stable demand, and the variation will not be large. Others will be unstable and will have a large variation.

Mean Absolute Deviation

Forecast error must be measured before it can be used to revise the forecast or to help in planning. There are several ways to measure error, but one commonly used is mean absolute deviation (MAD).

Consider the data on variability in the side figure.  Although the total error (variation) is zero, there is still considerable variation each month.

Total error would be useless to measure the variation.
One way to measure the variability is to calculate the total error ignoring the plus and minus signs and take the average.

 This is called mean absolute deviation:

mean implies an average,
absolute means without reference to plus and minus,
deviation refers to the error

 

Normal distribution
The mean absolute deviation measures the difference (error) between actual demand and forecast. Usually, actual demand is close to the forecast but sometimes is not. A graph of the number of times (frequency) actual demand is of a particular value produces a bell-shaped curve. This distribution is called a normal distribution and is shown in the site figure.

There are two important characteristics to normal curves: the central tendency, or average, and the dispersion, or spread, of the distribution. In the site figure, the central tendency is the forecast. The dispersion, the fatness or thinness of the normal curve, is measured by the standard deviation. The greater the dispersion, the larger the standard deviation. The mean absolute deviation is an approximation of the standard deviation and is used because it is easy to calculate and apply.

The mean absolute deviation is an approximation of the standard deviation and is used because it is easy to calculate and apply.
From statistics we know that the error will be within:
- ± 1 MAD of the average about 60% of the time,
- ± 2 MAD of the average about 90% of the time,
- ± 3 MAD of the average about 98% of the time.

Tracking Signal
The Tracking Signal or TS is a measure that indicates whether the forecast average is keeping pace with any genuine upward or downward changes in demand. Depending on the number of MAD’s selected, the TS can be used like a quality control chart indicating when the model is generating too much error in its forecasts. 

Bias exists when cumulative actual demand varies from forecast. The problem is in guessing whether the variance is due to random variation or bias. If the variation is due to random variation, the error will correct itself, and nothing should be done to adjust the forecast. However, if the error is due to bias, the forecast should be corrected. Using the mean absolute deviation, we can make some judgment about the reasonableness of the error. Under normal circumstances, the actual period demand will be within ± 3 MAD of the average 98% of the time. If actual period demand varies from the forecast by more than 3 MAD, we can be about 98% sure that the forecast is in error.

A tracking signal can be used to monitor the quality of the forecast. There are several procedures used, but one of the simpler is based on a comparison of the cumulative sum of the forecast errors to the mean absolute deviation.
                                                                 The TS formula is: (algebraic sum of forecast errors)/MAD
“The ratio of the cumulative algebraic sum of the deviations between the forecasts and the actual values to the mean absolute deviation. Used to signal when the validity of the forecasting model might be in doubt”.

Tracking signals are used to measure forecast bias & are computed by dividing the cumulative sum of the errors by the MAD. Bias will be shown if the results were consistently –ve or +ve. The calculation result should stay close to zero & should vary bet’n being –ve & +ve. A value called a trip value is the predetermined threshold at which an action message is generated, indicating potential forecast bias. One commonly used value is 4 in either direction. The absolute value of the tracking signal can be used as the alpha factor in exponential smoothing. This is called adaptive smoothing because the value of the alpha factors adapts to the forecast accuracy

Forecast Performance

Companies striving for operational excellence and a competitive advantage realize the impact forecasting has upon the ability of a company to satisfy its customers and to simultaneously manage its resources. Effective forecasting helps management resolve the dilemma of more demanding customer requirements and greater shareholder expectations. To resolve this dilemma, managers are expected to provide better customer service with fewer resources. In this environment, the importance of effective forecasting is elevated. In manufacturing and distribution companies, a forecast is not simply a projection of future business; it is a request for product (or a request for resources to ensure supply of a product). In simple terms, this is how the forecast works: If a product is in the forecast, you can expect the product (or resource) to be available. If it is not in the forecast, you should not expect the product
to be available. With this concept of a forecast as a request for product, forecast accuracy becomes crucial to ensuring satisfactory, or exemplary, customer service. Forecast accuracy also becomes critical to the proper utilization of resources. For example, when product is requested and not sold or the sale is delayed, resources have been tied up unnecessarily. When a product has not been forecasted but the company must still meet the customer demand, often this is accomplished at a considerably higher cost – a poor use of resources.

First and foremost, we need to measure forecast accuracy if we wish to improve. Measurements are used to make improvements to the specific forecast as well as to the demand planning process. An effective demand planning process measures forecast accuracy in different ways for different purposes. A detailed measure of forecast accuracy at the item level identifies individual products that are outside an established tolerance. This enables us to review – and correct – the individual product forecasts. The earlier a significant forecast error is identified, the quicker we can respond to the real market demand.

Aggregate measures at a product family level are used to determine whether there is a problem with the total (aggregate) product forecast. When a problem surfaces, a more detailed forecast review can be initiated. An aggregate measure is also used as an indicator of the quality of the forecast. It answers the following questions:
Is the forecast reliable?
Is the forecast getting better or worse?
Does the demand planning process need attention?
The aggregate forecast is the sum of the individual item forecasts. As such it also serves as a test for reasonableness. Further, it is used to quantify the overall marketing,sales and business plans. A recurring forecast problem is an indication of a poor demand planning process or, possibly, a tremendously uncertain market.

If the company has an effective demand planning process and significant product forecast errors persist, then a review of tactics and strategies to deal with the uncertainties is required. Since actual customer demand will almost always vary from the forecast either in quantity, timing, or both, it is necessary for companies to have tactics and strategies to deal with these variations. These tactics and strategies include: carrying a buffer inventory, varying delivery lead time, maintaining the ability to flex capacity, and managing demand. Most companies utilize a combination of these approaches. However, it is important that the approach chosen occurs through a managed process.

Because these decisions have considerable impact on customer service, cost, and resource requirements, it should not be left up to chance which tactics or strategies are utilized. Sales and Operations Planning is the management process used by many companies to ensure a managed approach to these tactics and strategies.

Measurement Criteria - An effective forecast measurement process meets certain criteria:
1. It should be “owned” by those responsible for achieving the forecast. Sales and Marketing should develop (or at least agree with) the measurement methodology.
2. The forecast measurements should be easily understood. Unfortunately, it is common to find forecast measurements that are so complicated that sales and marketing simply disavow the validity of the measurement and, thereby, take no accountability for the forecast.
3. The measurements should identify forecast problems quickly and easily at both the aggregate and detail levels which helps to prioritize items requiring forecast review.

Measuring Forecast Performance

Forecast accuracy
is the difference bet’n the actual sales recorded in a given period & that predicted by
the sales forecasting process. Forecast at the product level should be monitored during the monthly SOP meeting. The rate at which actual sales are booked against predicted sales is a measure of
forecast accuracy. Forecasts for specific products should be monitored by the master scheduler by
comparing the actual sales for a specific product to the predicted level of sales. Forecast accuracy should be monitored by developing acceptable levels of forecast error by product group &/or by specific product.

Items that have stable demand & high volume should have lower acceptable errors than those
with less stable demand & lower volume. A trend analysis of the variability in the forecast will help
determine the stability of the item being forecasted.

Forecast variability can be measured by determining the magnitude of the standard deviation or the
mean absolute deviation. These measures assume a normal distribution of sales about the mean. The
higher the magnitude of the std deviation the greater the spread of the normal distribution & the greater
the range of forecast variability.

The next issue in forecast performance measurement is the development of sample size. This measure
is based on analysis of the population being studied. Depending on the group & how the sample is
developed small nos. of observations can predict the behavior of larger groups of the population. This is
the theory behind customer service surveys, opinion polls, market analysis & statistical samplings.

Tracking signal is a way to determine when a forecasting technique for an item needs to be reviewed. A
tracking signal is activated when the running some of forecast errors when compared to the mean
absolute deviation of a distribution exceeds the no.4. Tracking signals are useful for determining the
turning points in a product life cycle as they indicate that the actual sales are no longer following the
historical trend.

Outliers are defined as data points that do not fit the overall pattern & trend in a population. It is important
to separate the true outliers from the data points that do have casual factors. A true outlier is either a
mistake in data entry or a one time event that will never be repeated. It’s important not to remove outliers
that are caused by factors that could potentially be repeated.

Demand filters are used as a means for controlling variation by providing a check that limits the amount
of variation from one period to the next.

Forecasting Monitoring: Forecasts should be monitored on a regular & ongoing basis. It is necessary
to establish the lower & upper control limits to the forecast accuracy. In some companies forecast is the
responsibility of mktg dept & in other companies it is the responsibility of operations dept. In general
terms the ownership of the forecasting process is best positioned in sales & mktg dept. A defined group
is assigned the responsibility of generating the forecast & that they are measured on the accuracy of the
forecast that they generate on the regular basis.

Forecasting Demand

Forecast is an estimate of future demand. A forecast can be constructed using quantitative methods, qualitative methods, or a combination of methods, and it can be based on extrinsic (external) or intrinsic (internal) factors. Various forecasting techniques attempt to predict one or more of the four components of demand: cyclical, random, seasonal, and trend”.

Forecast will contain some element of error, therefore it should be expressed as a value, with a +/- % of error associated with it. To improve the forecast process the error % should be tracked & monitored.

Forecast error is the difference between actual demand and forecast demand, stated as an absolute value or as a percentage”. “Forecast management is the process of making, checking, correcting, and using forecasts. It also includes determination of the forecast horizon”.

Forecasts depend upon what is to be done. They must be made for the strategic business plan (SBP), the production plan (PP) and the master production schedule (MPS). The purpose, planning horizons and level of detail vary for each.

The strategic business plan is concerned with overall markets and the direction of the economy over the next two to ten years or more. Its purpose is to provide time to plan for those things that take long to change. For production, the strategic business plan should provide sufficient time for resource planning: plant expansion, capital equipment purchase, and anything requiring a long lead time to purchase. The level of detail is not high, and usually forecasts are in sales units, sales dollars, or capacity. Forecasts and planning will probably be reviewed quarterly or yearly.

Production planning is concerned with manufacturing activity for the next one to three years. For manufacturing, it means forecasting those items needed for production planning, such as budgets, labor planning, long lead time, procurement items, and overall inventory levels. Forecasts are made for groups or families of products rather than specific end items. Forecasts and plans will probably be reviewed monthly.

Master production scheduling is concerned with production activity from the present to a few months ahead. Forecasts are made for individual items, as found on a master production schedule, individual item inventory levels, raw materials and component parts, labor planning, and so forth. Forecasts and plans will probably be reviewed weekly.

Demand patterns
A pattern is the general shape of a time series. Pattern shows that actual demand varies from period to period.  If historical data for dem
and are plotted against a time scale, they will show any shapes or consistent patterns that exist. A pattern is the general shape of a time series.Although some individual data points will not fall exactly on the pattern, they tend to cluster around it.

The above fig shows a hypothetical historical demand pattern. The pattern shows that actual demand varies from period to period. There are four reasons for this: trend, seasonality, random variation, and cycle.

- Trend: The trend can be level, having no change from period to period, or it can rise or fall;
In trend method a forecast is calculated by inserting a time value into the regression equation. The regression equation is determined from the time-serieas data using the “least squares method”.
F=a+bt
This graph illustrates a linear trend, but there are different shapes, such as geometric or exponential. The trend can be level, having no change from period to period, or it can rise or fall.
 
- Seasonality: Each year’ s demand fluctuating depending on the time of year;
The demand pattern in the figure shows each year’s demand fluctuating depending on the time of year. This fluctuation may be the result of the weather, holiday seasons, or particular events that take place on a seasonal basis. Seasonality is usually thought of as occurring on a yearly basis, but it can also occur on a weekly or even daily basis. A restaurant’s demand varies with the hour of the day, and supermarket sales vary with the day of the week

- Random variation: It occurs where many factors affect demand during specific periods and occur on a random basis;
Random variation occurs where many factors affect demand during specific periods and occur on a random basis. The variation may be small, with actual demand falling close to the pattern, or it may be large, with the points widely scattered. The pattern of variation can usually be measured, and this will be discussed in the section on tracking the forecast.

- Cycle: Over a span of several years, wavelike increases and decreases in the economy influence demand. Forecasting of cycles is a job for economists.

PRINCIPLES OF FORECASTING
Forecasts have 4 major characteristics:

- Forecasts are usually wrong,
- Every forecast should include an estimate of error,
- Forecasts are more accurate for families or groups,
- Forecasts are more accurate for nearer time periods.

Forecasts are usually wrong. Forecasts attempt to look into the unknown future and, except by sheer luck, will be wrong to some degree. Errors are inevitable and must be expected.

Every forecast should include an estimate of error. Since forecasts are expected to be wrong, the real question is, “By how much?” Every forecast should include an estimate of error often expressed as a percentage (plus and minus) of the forecast or as a range between maximum and minimum values. Estimates of this error can be made statistically by studying the variability of demand about the average demand.

Forecasts are more accurate for families or groups. The behavior of individual items in a group is random even when the group has very stable characteristics. For example, the marks for individual students in a class are more difficult to forecast accurately than the class average. High marks average out with low marks. This means that forecasts are more accurate for large groups of items than for individual items in a group.
For production planning, families or groups are based on the similarity of process and equipment used. For example, a firm forecasting the demand for knit socks as a product group might forecast men’s socks as one group and women’s as another since the markets are different. However, production of men’s and women’s ankle socks will be done on the same machines and knee socks on another. For production planning, the forecast should be for (a) men’s and women’s ankle socks and (b) men’s and women’s knee socks.

Forecasts are more accurate for nearer time periods. The near future holds less uncertainty than the far future. Most people are more confident in forecasting what they will be doing over the next week than a year from now. As someone once said, tomorrow is expected to be pretty much like today.
In the same way, demand for the near term is easier for a company to forecast than for a time in the distant future. This is extremely important for long lead-time items and especially so if their demand is dynamic. Anything that can be done to reduce lead time will improve forecast accuracy.

Stable versus dynamic
The demand pattern that retain the same general shape are called stable and those that do not are called dynamic. Dynamic change can affect the trend, seasonality or randomness of the demand. The more stable the demand, the easier it is to forecast.

Dependent versus independent demand
A product is independent when is not related to the demand for any other product. Dependent demand for a product occurs where the demand for the item is derived from that of a second item. Requirements for dependent demand are calculated
and for independent demand need be forecast.

Forecasting Inaccuracy & Accuracy

Reasons for forecasting Inaccuracy
Lack of participation by functional managers in the development of & the execution of a forecast system.Forecast requires alignment with one another to ve effective.
Too difficult to understand. If a forecasting tool is too difficult to understand it will most likely fail.
The more complex is the forecasting tool, the more isolated users are from the process& the more work is required to maintain it. Such a tool is also likely to be more costly to run & maintain.
Lack of compatibility bet’n the forecasting system & the capabilities of the organisation. If the techniques are not understood & the results are not trusted, managers are unlikely to follow the forecast & rely instead on informal techniques or gut feelings.
Data may be inaccurate either by collection errors or classification errors, including errors in sampling methods, data classification, measurement, aggregation, time collection & overlooking important data.
Some data are inappropriate for forecasting. Some items should not be forecast such as dependent demand or parts within the final assembly schedule.
Lack of monitoring. Some companies do not check the accuracy of the forecast. They do not track actual sales against projected sales & hence have no way of assessing current forecast performance or establishing targets for improvement.

Demand filters
“A standard that is set to monitor sales data for individual items in forecasting models. It is usually set to be tripped when the demand for a period differs from the forecast by more than some number  of mean absolute deviations”. Demand filters are a means of controlling variations by providing a check that limits the amount of the variation from one period to the next. A demand filter typically is a ratio of the new demand to the avg. of the old demand. Stable items should have tight demand filters to maintain more control as significant variations in demand should be examined & explained. Demand filters also prevent order input errors by establishing reasonableness of the order quantities being placed. Any input error will trip the demand filter & hence will be flagged for review & correction.

Improving Forecasting Accuracy
Steps to improve forecast accuracy
Assign responsibility & accountability for forecasts.
Set realistic goals for accuracy.
Tie performance reviews to those goals.
Provide good forecasting goals.
Forecast at the right level of the product structure.
Track forecast accuracy regularly.
Review & improve areas of poor forecast performance.

Intrinsic & Extrinsic Factors

Intrinsic Factors
Internal factors are those under the direct control of the company itself. They are:
a) Product Life Cycle Management:
Stage 1: In the introductory phase accurate forecasting of demand is essential. Since there is no previous sales history, qualitative methods are used for forecasting.
Stage 2: In the growth phase the product grows & must be forecasted using the best available methods. Typically a mix of qualitative & quantitative measures are used. The data in this phase exhibit a positive trend.
Stage 3: In the maturity phase the product is established in the marketplace. In this phase the majority of profit can be expected from product sales. Demand should be fairly predictable based on historical performance so quantitative forecasting methods can be used. Data in this phase are largely flat so moving averages & exponential smoothing techniques are typically used, unless the demand is seasonal.
Stage 4: In the decline phase the product demand is dwindling (gradually decreasing) & the company has to take make decision about not supporting the product. Some companies actually sell the product line to another company prepared to undertake the sales & support functions. Accurate forecasting of declining product is essential to minimize inventory exposure & obsolescence issues. The data in this phase typically exhibit a negative trend.
b) Planned price changes:
Many companies use the concept of hedging to acquire stocks of products that are likely to be subject to future price changes.
c) Changes in the sales force:
In many companies there is a direct correlation between the size of the sales force & the level of sales revenue. Assuming the market is not saturated the company could expect to receive double the sales
revenue from doubled sales force.
d) Resource Constraints:
If a company historically faced resource constraints then the sales history reflects only what was capable of being delivered not the true level of actual demand from the customers. Therefore it is important to note the periods of constrained output so that true demand may be recognized not just what was sold or shipped in a given period.
e) Marketing & Sales Promotion:
In many consumer goods companies products are often sold during special promotional periods. In these promotional periods special offers or pricing is created in order to stimulate additional demand for the product to try to gain the market share from the competition & become a more dominant supplier to the market. The promotion creates artificial demand that is transitory at best.
f) Advertising:
Higher rates of advertising for higher sales.

External (Extrinsic) Factors:
The external factors fall outside the control of the company. They are:
a) New Customers:
There is usually a direct correlation between the number of new customers & the level of sales activity. External factors such as casual factors, leading indicators & correlation analysis are useful in forecasting total company demand or demand for families of products.
b) Plans of Major Customers:
A supplier to Wal-Mart will be affected when Wal-Mart opens a new store. There will be an immediate demand for inventory to fill the selves in the new store & an ongoing additional demand for replenishment.
c) Government Policies:
Government policies cam have a large impact on projected sale revenue.
d) Regulatory Concerns:
Regulatory concerns are issues whereby legislation. Regulation or both have been used to control the sale of certain products or services to customers. E.g. Tobacco products & guns.
e) Economic Conditions:
The general health of economy has a large impact on the sales revenue of many companies. E.g. during economic recession people are unsure of the future so they avoid making commitments to major products.
f) Environmental Issues
g) Global Trends:
The world is shrinking fast today & previously protected markets are being opened up to the harsh light of expert competition.

Manufacturing Planning and Control (MPC) system

A production (or manufacturing) planning and control (MPC) system is concerned with planning and controlling all aspects of manufacturing, including materials, scheduling machines and people, and coordinating suppliers and customers. An effective MPC system is critical to the success of any company. An MPC system's design is not a one-off undertaking; it should be adaptive to respond to changes in the competitive arena, customer requirements, strategy, supply chain and other possible problems (Vollmann )

There are 5 levels in the manufacturing planning and control (MPC) system:

- Strategic business plan,
- Production plan (sales and operations plan),
- Master production schedule,
- Materiel requirements plan,
- Purchasing and production activity control.

Each level varies in purpose, time span and level of detail. Since each level is for different time span and purposes, each differs in the following:

- Purpose of the plan,
- Planning horizon: the time span from now to some time in future for which plan is created,
- Level of detail: the detail about products require for the plan,
- Planning cycle: the frequency with which the plan is reviewed.

At each level, 3 questions must be answered:
- What are the priorities: how much of what is to be produced and when?
- What is the available capacity: what resources do we have?
- How can differences between priorities and capacity be resolved?

 Strategic Business Plan (SBP)
The strategic business plan is a statement of the major goals and objectives the company expects to achieve over the next 2 to 10 years or more. It is a statement of the broad direction and show the kind of business the firm wants to do in the future.

The development of the SBP is the responsibility of senior management. Each department produces its own plans to achieve the objectives set by the SBP. These plan will be coordinated with one another and with the SBP. The level of detail is not high. It is concerned with general market and production requirements (total market for major product groups) and not sales of individual items.
Strategic business plans are usually reviewed every six months to 1 year.

Production Plan (PP)
Given the objectives set by SBP, production management is concerned with the following:

- The quantities of each product group that must be produced in each period;
- The desired inventory levels;
- The resources of equipment, labor, and material needed in each period;
- The availability of the resources needed.

The level of detail is not high. The production plan will show major product groups or families.
Production planners must devise a plan to satisfy market demand within the resources available to the company. For effective planning, there must be a balance between priority and capacity.
Along with the market and financial plans, the PP is concerned with implementing the strategic business plan. The planning horizon is usually 6 to 18 months and is reviewed each month or quarter.

The Master Production Schedule (MPS)
The master production schedule is a plan for the production of individual end items. It breaks down the PP to show, for each period, the quantity of each end item to be made. Inputs to the MPS are PP, the forecast for individual end items, sales orders, inventories and existing capacity.
The level of detail is higher. The MPS is developed for individual end items.
The planning horizon usually extends from 3 to 18 months but depends on the purchasing and manufacturing lead times . Usually the plans are
reviewed and changed weekly or monthly.

Material Requirements Planning (MRP)
The material requirements planning is a plan for the production and purchase of the components used in making the items in the MPS. It shows the quantities needed and when manufacturing intends to make or use them.
The level of detail is high. The MRP establishes when the components and parts are needed to make each end item.
The planning horizon is at least as long as the combined purchase and manufacturing lead times. It usually extends from 3 to 18 months.

Purchasing and Production Activity Control (PAC)
Purchasing and production activity control use the MRP to decide the purchase or manufacture of specific items. Purchasing and PAC represent the implementation and control phase of MPC system. Purchasing is responsible for establishing and controlling the flow of raw materials into the factory. PAC is responsible for planning and controlling the flow of work through the factory.
The planning horizon is very short, from a day to a month. The level of detail is high since it is concerned with individual components, workstations and orders. Plans are reviewed and revised daily.

Capacity management
At each level in the MPC system, the priority plan must be tested against the available resources and capacity of the manufacturing system. If the capacity cannot be made available when needed, then the plans must be changed. There can be no valid, workable PP unless this is done.
Over several years, machinery, equipment and plants can be added to or taken away from manufacturing. However, in the time spans involved from PP to PAC, these kind of changes cannot be made. Some changes, such as changing the number of shifts, working overtime, subcontracting the work, can be accomplished in these time spans.

Moving averages

One simple way to forecast is to take the average demand for the last three or six periods and use that figure as the forecast for the next period. At the end of the next period, the first-period demand is dropped and the latest-period demand added to determine a new average to be used as a forecast. If a longer period is used, the forecast does not react as quickly. The fewer months included in the moving average, the more weight is given to the latest information and the faster the forecast reacts to trends. However, the forecast will always lag behind a trend.

Moving averages are best used for forecasting products with stable demand where there is a little trend or seasonality. They are also useful to filter out random fluctuation. One drawback to using moving averages is the need to retain several
periods of history for each item to be forecast.

Simple moving average: At = (Dt + Dt-1 + Dt-2 + ... + Dt-N+1 )/N

where N = total number of periods in the average forecast for period t+1: Ft+1 = At

Key Decision: N - How many periods should be considered in the forecast
Higher value of N - greater smoothing, lower responsiveness
Lower value of N - less smoothing, more responsiveness 

The more periods (N) over which the moving average is calculated, the less susceptible the forecast is to random variations, but the less responsive it is to changes 
A large value of N is appropriate if the underlying pattern of demand is stable 
A smaller value of N is appropriate if the underlying pattern is changing or if it is important to identify short-term fluctuations

Example

                                     January               92              July                     84

                                     February              83              August                81

                                     March                 66              September            75

                                      April                  74              October                 63 

                                      May                  75              November               91

                                       June                 84              December              84

 

Suppose it was decided to use a three-month moving average on the data shown in above Figure. Our forecast for January, based on the demand in October, November, and December, would be:

                                         63 91 + 84               

                                                               .  = 79

                                                  3

  • Now suppose that January demand turned out to be 90 instead of 79. The forecast for February would be calculated as:

                                          91 + 84 + 90

                                                                  = 88

                                                   3

Weighted moving average

A weighted moving average adjusts the moving average method to reflect fluctuations more closely by assigning weights to the most recent data, meaning, that the older data is usually less important. The weights are based on intuition and lie between 0 and 1 for a total of 1.0

Quantitative & Qualitative Forecasting Techniques:

Quantitative Methods:
Based on historical information that is usually available within the company. Various techniques are:

Trend Analysis:
A method for forecasting sales data when a definite upward or downward pattern exists. Model includes double exponential smoothing, regression & triple smoothing.

Seasonal Adjustment:
Seasonal models take into account the variation of demand from season to season. Adjustments can be made to a baseline forecast to predict the impact of a seasonal demand.

Decomposition:
“A method of forecasting where time series data are separated into up to three components: trend, seasonal, and cyclical; where trend includes the general horizontal upward or downward movement over time; seasonal includes a recurring demand pattern such as day of the week,
weekly, monthly, or quarterly; and cyclical includes any repeating, non seasonal pattern. A fourth component is random, that is, data with no pattern. The new forecast is made by projecting the patterns individually determined and then combining them”.

Graphical Methods:
Plotting information in a graphical form. It is relatively easy to convert a spreadsheet into a graph that conveys the information in a visual manner. Trends & patterns are easier to spot & extrapolation of previous demand can be used to predict future demands.

Econometric Modeling:
A set of equations intended to be used simultaneously to capture the way in which dependent and independent variables are interrelated.

Life Cycle Modeling:
“A quantitative forecasting technique based on applying past patterns of demand data covering introduction, growth, maturity, saturation, and decline of similar products to a new product family”.

Qualitative Methods:
Based on subjective information such as intuition or informed opinion. This type of forecast is essential for new products where no historical information is available. This type of forecast is primarily used for medium & long term planning. Qualitative techniques include the use of information gathered from:

Expert Opinion:
The opinions of experts in the particular area are sought. Experts give their views on current trends & likely future developments that may have an impact on the general economy or a specific industry or market.

Market Research: Conducted thru surveys.

Focus Groups: Consists of panels of customers who are asked to provide their opinions about a product or service.

Historical Analogy: The sale of new product or service is compared with the sales of a previous similar product or service.
It is assumed that the sales patterns associated with the previous product or service can be transferred to the new product or service.

Delphi Method: “A qualitative forecasting technique where the opinions of experts are combined in a series of iterations (repetitions). The results of each iteration are used to develop the next, so that convergence of the experts’ opinions is obtained”. This method is based on the knowledge & judgment of a small group of experts. In many companies a mixture of both historical information (analyzed by quantitative technique) combined with qualitative input (from groups of experts) is useful in establishing a more accurate forecast.

Panel Consensus: A group of people provides opinion about the future & a facilitator brings the group to a consensus. The groups as whole would make better decisions than would each member individually.

Seasonality


Many products have a seasonal or periodic demand pattern: skis, lawnmowers, bathing suits, and Christmas tree lights are examples. Less obvious are products whose demand varies by the time of day, week, or month. Examples of these might be electric power usage during the day or grocery shopping during the week. Power usage peaks between 4 and 7 p.m., and supermarkets are most busy toward the end of the week or before certain holidays.

A useful indication of the degree of seasonal variation for a product is the seasonal index. This index is an estimate of how much the demand during the season will be above or below the average demand for the product. For example, swimsuit demand might average 100 per month, but in July the average is 175 and in September, 35. The index for July demand would be 1.75 and for September, 0.35.

Example:

Seasonal Forecasts
The equation for developing seasonal indices is also used to forecast seasonal demand. If a company forecasts average demand for all periods, the seasonal indices can be used to calculate the seasonal forecasts. Changing the equation around we get:
                                                    Seasonal demand = (seasonal index)*(deseasonalized demand)

Usage of seasonal Index
1.Calculate the seasonal index from past years(year 1 & 2)
2.Forecast the demand for Year 3 using some quantitative method – moving average/exponential factor
3.Calculate the average demand for year 3 (Total demand in point 2/no of months)
4.Use the seasonal index to find out the demand for each month in year 3 (Amount in point 3* Seasonal index for the month)

Example: A company forecasts an annual demand of 420 units for next year.Calculate the forecast for quarterly sales if the seasonal Index for the first three quarters are 1.28,1.02 & 0.75 respectively.
Ans:
Forecast average quarterly demand =   420/4 = 105 units
seasonal Index for 4th quarter = 4 - (1.28 + 1.02 + 0.75) =0.95

Expected quarter demand = (seasonal index) (forecast quarterly demand)
Expected first-quarter demand = 1.28 x 105 = 134.4 units
Expected second-quarter demand = 1.02 x 105 = 107.1 units
Expected third-quarter demand = 0.75 x 105 = 78.75 units
Expected fourth-quarter demand = 0.95 x 105 = 99.75 units
Total forecast demand = 420 units

Deseasonalized Demand

Forecasts do not consider random variation. They are made for average demand, and seasonal demand is calculated from the average using seasonal indices.Below fig shows both actual demand and forecast average demand. The forecast average demand is also the deseasonalized demand. Historical data are of actual seasonal demand, and they must be deseasonalized before they can be used to develop a forecast of average demand.

Also, if comparisons are made between sales in different periods, they are meaningless unless deseasonalized data are used. For example, a company selling tennis rackets finds demand is usually largest in the summer. However, some people play indoor tennis, so there is demand in the winter months as well. If demand in January was 5200 units and in June was 24,000 units, how could January demand be compared to June demand to see which was the better demand month? If there is seasonality, comparison of actual demand would be meaningless. Deseasonalized data are needed to make a comparison.

The equation to calculate deseasonalized demand is derived from the previous seasonal equation and is as follows:
                                                    Deseasonalized demand = (actual seasonal demand)/(Seasonal index)
 
Example
A company selling tennis rackets has a January demand of 5200 units and a July demand of 24,000 units. If the seasonal indices for January were 0.5 and for June were 2.5, calculate the deseasonalized January and July demand. How do the two months compare?
Answer :
Deseasonalized January demand = 5200/0.5=10,400 units
Deseasonalized June demand = 24,000/2.5=9600 units
June and January demand can now be compared. On a deseasonalized basis, January demand is greater than June demand.
Deseasonalized data must be used for forecasting. Forecasts are made for average demand, and the forecast for seasonal demand is calculated from the average demand using the appropriate season index.

The rules for forecasting with seasonality are:
Only use deseasonalized data to forecast.
Forecast deseasonalized demand, not seasonal demand.
Calculate the seasonal forecast by applying the seasonal index to the base forecast.