Business Forecasting One of the steps, say the very first one, in the process of management is planning. Planning is understood as the process of setting goals and choosing the means to achieve these goals. Planning is essential for, without it, managers cannot organise people and resources effectively. Meaning and Definition Forecasting is fundamental to planning. Forecasts are statements about future, specifying the volume of sales to be achieved and equipment, materials and other inputs needed to realise the expected sales.

A popular definition of forecasting is that, it is estimating the future demand for products and services and the resources necessary to produce these outputs. Starting point in forecasting is sales or demand forecasting. Sales forecasts trigger all other forecasts in production function. Need for Sales Forecasting Following are some of the reasons, why operations managers must develop forecasts : 1. New Facility Planning Strategic activities such as designing and building a new factory or designing and implementing a new production process, might take a long time, say five years.

This requires long range forecasts of demand for existing and new products, so that, operations managers can Lave the necessary lead time to build the processes to produce the products and services when needed. 2. Production Planning The rate of production must vary to meet the fluctuating demand from time to time (usually month to month). A time period of several months may be necessary to change the capacities of production processes Intermediate range demand forecast, helps operations managers get the lead time necessary to provide the capacity to produce the products to meet the variable monthly demands. 3.Work Force Scheduling Where the demand for products and services varies from week to week, it is necessary to vary the work force levels to meet the varying demands by using overtime, lay-offs or hiring. For this, operations managers need short range demand forecasts to enable them to have the lead time necessary to provide work force the changes to produce products or service to meet the weekly demands. 4.

Financial Planning Sales forecasts are the driving force in budgeting Budgeting is used by many operations managers to plan and control the financial performance of their production department. Types of ForecastsThere are long-term forecasts as well as short-term forecasts. Operations managers need long-range forecasts to make strategic decisions about products, processes and facilities. They also need short-term forecasts to assist them in making decisions about production issues that span, only the next few weeks.

Since forecasting forms an integral part of planning and decision-making, production managers must be clear about the horizon of forecasts – month or year, for example. Additionally, they must also be clear about the method of forecasting and unit of forecasting (gross rupee sales, individual product demand, etc).Application of Long-range and Short-range Demand Forecasts (a) Long range Forecast Long range forecasts provide, operations managers, with informatior decisions such as the following : 1.

Selecting a product design. The final design is dependent on expected sales volume. If the demand is high, the design should be such that the product can be mass-produced to ensure low-cost manufacture. 2. Selecting a production processing scheme (i.

e. , process design) for a new product. 3. Selecting a plan for the long range supply of scarce materials. 4.

Selecting a long range production capacity plan. . Selecting a long range financial plan for acquiring funds for capital investment. 6. To build new buildings and to purchase new machines. 7.

To develop new sources of materials and new sources of capital funds (finance). (b) Short-range Forecast Short range sales forecasts provide operations managers, with information to make important decisions such as the following : 1. How much inventory of a particular product (i. e. , finished goods) should be carried -next month? 2.

How much of each product should be scheduled for production next week? giver th quantity in the sales forecast and the quantity in inventory) 3. How much of each raw material should be ordered for delivery next week? (giver tr quantity of products in the production schedule and the quantity in inventory) 4. How many workers should be scheduled to work on regular time basis and on overumi basis next week? (given the production schedule and the number of workers available How many maintenance workers should be scheduled to work next week end? (fftwr the production schedule and the past experience regarding break downs). Fig 3. Forecasting Model A Forecasting Model Forecasting is more them a technique – it is a system. Most successful organisations, view it as a system. As seen in Fig 3.

1, forecasting model begins with information inputs. Forecasts are possible, only when a history of past data exists. An established TV manufacturer, for example, can use past data to forecast the number of picture screens required for next week’s TV -assembly schedule. Past data are available for products, which are already being produced. Suppose, the TV manufacturer decides to offer a totally new model.

Because, past data are not available, forecasting needs to be based on the manager’s skill, experience, judgement and established techniques. Data are used to forecast sales in terms of quantity and value of sales. Sales forecast is translated into demand for factory capacities, funds, facilities and others (long-term forecasting); for workforce, materials, department requirements, inventories and others (intermediate-range forecasting) and short-term forecasting involving forecasting for specific labour skill required, machine hours, cash and inventories. When the long-range, intermediate-range and short-range orecasts are operationalised, the consequence is production of goods and services. Forecasting Methods Forecasting methods can be broadly divided into two main categories: •Quantitative methods •Qualitative or judgemental methods In some situations, a combination of methods may be used.

In quantitative methods, also called time series methods, past data are used in making a forecast for the future. This process looks like driving a car by looking through the rear mirror. However, where forecast is-made for short-periods, quantitative methods are more appropriate.

Qualitative or judgemental methods rely on an expert’s opinion in making a prediction for the future. These methods are useful for intermediate to long-range forecasting tasks. The use of judgement in forecasting sounds unscientific and adhoc. But, where new products are sought to be introduced, there are few alternatives other than using the informed opinion opinion of knowledgeable people. However, to obtain better results, judgemental methods are used in conjunction with other categories of methods. Detailed description of the forecasting methods follows.Quantitative Methods These methods seek to identify patterns in the past data.

In order to systematically analyse data, managers use a time series analysis. In this, analysts plot demand data on a time scale, study the plots and look for consistent shapes or patterns. (See Fig.

3. 2) Demand pattern becomes continuous when, it is constant and does not consistently increase or decrease. The sales of a product in the mature stage of its life cycle may show a horizontal demand pattern.

Linear trend emerges when, demand increases or decreases from one period to the next.The sales of products in the growth stage of the product life cycle tend to show am upward trend, whereas those in decline, tend to show a downward trend. The cyclical pattern pertains to the influence of seasonal factors that have impact on demand, either positively or negatively. For example, the demand for woolen wear will be in high in winter and low during summer. Common Time Series Models The most common and relatively easiest methods for developing a forecast from past data are simple moving averages, weighted moving averages, exponential smoothing and regression analysis. Simple Moving Average (SMA)In this model, the arithmetic average of the actual sales for a specific number of recent past time periods is taken as the forecast for the next time period.

SMA = Sum of demands for all periods Chosen number of periods SMA = ? D / n = 1/n (D1 +D2 +D3+ … Dn) Where, n = the chosen number of periods i = 1 is the oldest period in the n-period average i = n is the most recent period Dt = the demand in the ‘i’th period Weighted Moving Average (WMA) WMA is like the SMA model described above, except that, instead of an arithmetic average of past sales, a weighted average of past sales is the forecast for the next time period.

A WMA allows for varying, not equal weighting of old demands n Thus, WMA = ? CtDt Where Dt is the demand during time period ‘t’,Ct is the weight t=1given to that demand and *n’ is the chosen number . of periods Also, O ? C ? 1 and n ? Ct = 1 t=l Exponential Smoothing Models In these methods, the forecasted sales for the last period are modified by information about the forecast error of the last periods.These modifications of the last year’s forecasts are the forecasts for the next time periods. In these methods, the weight assigned to a previous period’s demand decreases exponentially as that data gets older. Thus, recent demand data receive a higher weight than does the older demand data. Regression Analysis Regression analysis is a forecasting technique that establishes a relationship between variables – one dependent and other(s) independent. In simple regression, there is only one independent variable.

In multiple regression there is more than one independent variable.If the historical data set is a time series, the independent variable is the time period and the dependent variable in sales forecasting, is sales. A regression model does not have to be based on a time series, in such cases, the knowledge of future values of the independent variable is used to predict future values of the dependent variable. Regression is normally used in long-range forecasting, but, if care is taken in selecting the number of periods included in the historical data and that data set is projected only a few periods into the future, then regression may also be used for short-range forecasting.Weakness of Time Series Analysis Time series analysis basically depends on past data. This dependence on historical data is itself one of the weaknesses of the time series analysis, and the validity of forecast depends upon the similarity between past trends and future conditions. Any significant departure from historical trends will weaken the forecast dramatically.

Unfortunately, departures from historical trends seem to be occurring with increasing frequency. A second potential weakness in time series analysis is that, it provides quantitative answers.Managers must take care that, they do not place too much confidence in these results. The use of numbers and equations often gives a misleading appearance of scientific accuracy. Qualitative or Judgemental Methods A qualitative forecast is one, that is not based exclusively on a mathematical model.

These methods are usually based on judgement about the causal factors that underlie the sales of particular products or services and on opinions about the relative likelihood of these causal factors being present in the future. Judgemental methods are useful when historical data are not available.In the absence of past data, statistical methods have no validity.

Past data, even when they exist, may not be representative of future conditions. Qualitative forecasts are the only alternatives available. Further, in these days of management science and computers, qualitative forecasts assume greater relevance. The most popular judgmental forecasting methods are: Executive Committee Consensus, Delphi Method, Survey of Salesforce, Survey of Customers, Historical Analogy and Market Research.

Executive Committee Consensus/Jury of Executive’s OpinionHere, a committee of executives from different departments is constituted and is entrusted with the responsibility of developing a forecast. The Committee may use many inputs from all parts of the organisation and may have staff analysts provide analysis as needed. Such forecasts tend to be compromised ones, not reflecting the extremes that might be present, if they were prepared by individuals. However, that is the most commonly used method of forecast. The Delphi Method First developed fey Rand Corporation, Delphi is the n ought after method of forecast.The method seeks to eliminate the undesirable consequent of group thinking which does do exist when experts meet in committees. Delphi method draws on a pool of experts from both inside and outside the organisation. Members are so drawn that, each one is an expert in one aspect of the problem and none is conversant with all aspects of the issue.

In general, the method proceeds on the following lines: 1. Each expert in the group, makes independent predictions in the form of brief statements. 2. The coordinator edits and clarifies these statements. 3.The coordinator provides a series of written questions to the experts, that include feedback supplied by the other experts. 4. Steps 1 to 3 are repeated several times, till consensus is obtained.

As many as six rounds may be needed to reach the convergence. Survey of Salesforce/Field Expectation Method Individual members of salesforce are required to submit sales forecasts of their respective regions. These estimates are combined to form an estimate of sales for all regions. Managers must then, transform this estimate into a sales forecast to ensure realistic estimates.This is a popular forecasting method for companies that have a good communication system in force and that have salesforce who sells directly to customers. Survey of Customers/User’s Expectation Method In this method, estimates of future sales are obtained directly from customers. Individual customers are surveyed to determine what quantities of the company’s products they intend to purchase in each future time period. A sales forecast is determined by combining individual customers’ responses, and where customers are limited in number, this method is highly useful.

Historical AnalogyThis method ties the estimate of future sales of product to knowledge of a similar product’s sales. Knowledge of one product’s sales during various stages of its product life cycle is applied to the estimate of sale for a similar product. For example, an assumption can be made that, colour television would follow the general sales pattern experienced with black and white television. Where a product is new, this method is particularly useful. Market Surveys In the market research method, questionnaires, telephone talks or field interviews form the basis for predicting market demand for products.Market surveys are ordinarily preferred for new products or for existing products to be introduced into new market segments. Choice of a Model We have, till now, discussed both quantitative as well as qualitative techniques of forecasting. Managers now have a problem of selecting a particular model out of many.

The criteria for selecting a technique include accuracy, cost, ease of application and specific requirements of a planning situation. It has been proved that not one method meets all the criteria, but a combination of the techniques will be more effective in any given situation. bibliography production management.

by L R Potti