Introduction to Forecasting

Forecasting is the process of making predictions of the future based on past and present data and most commonly by analysis of trends. A commonplace example might be estimation of some variable of interest at some specified future date.

Every business enterprise interested in planning its activities must have clear idea about the demand for its product .Important business planning decisions, including the strategies to be followed, the amount of capital that is likely to be necessary, labor requirement and skills, the necessary distribution and after-sale service networks, sales incentives, sourcing of raw material, etc. are all critically dependent on the perception of the demand of its product. If this perception is substantially faulty, most of these decisions of the enterpriser likely to prove to be erroneous and lead to avoidable losses. A reasonably correct estimate of demand on the other hand can prove to be the key for a successful venture.

Every organization invariably engages in annual planning exercise. The heads of various functional areas such as marketing, production, materials and finance take part in this exercise with specific objectives. The marketing function provides data on sales that the organization should target in coming year. This is primarily achieved through forecasting. Based on this inputs, the production function prepares an annual production plan and projects various requirements on the basis of this plan. The material function prepares a procurement plan to match the requirements projected by the production function. Finally, on the basis of all these, the finance function undertakes cash planning and funds management. Therefore, forecasting plays a vital role in every organization

What is Demand Forecasting?

Demand is a key factor in business, but it’s not always constant. Demand can be defined as the market’s perceived need for a product or service. This translates into willingness to purchase the product or service. Supply, on the other hand, is the quantity of that product or service available to meet the demand. It’s often determined by the number of suppliers of a product or service in a market, and their individual production capacities. The biggest challenges for operations management is managing demand.

Types of Forecasts

• Economic forecasts – Predict a variety of economic indicators, like money supply, inflation rates, interest rates, etc.
• Technological forecasts – Predict rates of technological progress and innovation.
• Demand forecasts – Predict the future demand for a company’s products or services.

Demand trend types on the basis of regularity, are

• growing demand – A growing demand appears on a chart as a positive sloping line. Each high point and low point tends to be higher than the last. New products that are being launched or ones that are being heavily promoted typically show a fairly sharp upward trend.
• declining demand – A declining demand appears as a negative sloping line.
• stagnating demand – A chart depicting a relatively straight line with no distinctive upward or downward trend over a nine year period.
• cyclical demand – A cyclical demand is one that repeats itself over time, and can be long term or short term. Short-term cycles include seasonal fluctuations.
• randomly fluctuating demand – Random fluctuations are considered “noise” that makes forecasting demand more difficult. Random fluctuations can occur within any of the trend types.

The formulation of appropriate and useful production policy is an important aspect for an enterprise. This involves determination of level of production, manpower requirements equipment and inventory level etc. All these decisions are basically related to the size of production which in turn can be determined from potential demand of the product. Thus, the starting point of decision related to production strategy is the product demand forecast for a specified period. To know what a business should perform we must know its future Sales. In the absence of this information , both short and long term planning will rest on the foundation which is much less substantial than sand. A poor job of demand forecasting will lead to an ineffective production planning and towards an inventory that is either too large or too small.

In a literal sense forecasting means prediction. Forecasting may be defined as a technique of translating past experience in the prediction of things to come. It tries to evaluate the magnitude and significance of forces that will affect future operating conditions in an enterprise.

In the words of Garfield, “Production is an integral part of any of any scientific generalization that holds the relationship between two or more factors. The generalization must hold not only with respect to past observations related to the same phenomenon but also for all future observations related to the same phenomenon. Production is even more organically related to these that those generalization which establish a definite time sequence in the occurrence of certain factors; Due to dynamic nature of market phenomenon demand forecasting has become a continuous process and requires regular monitoring of the situation. Demand forecasts are first approximations to production planning. These provide foundation s upon which plans may rest and adjustments may be made.

“Demand forecast is an estimate of sales in monetary or physical units for a specified future period under a proposed business plan or program or under assumed set of economic and other environmental forces, planning premises outside the business organization for which the forecast estimate is made.”

Sales forecast is an estimate based on some past information, the prevailing situation and prospect of future. It is based on an effective system and is valid only to some specified period. The following are some main components of a sales forecasting system:

• Market Research Operations to get the relevant and reliable information about the trends in the market
• A data processing and analyzing system to estimate and evaluate the sales performance in the various markets
• Proper co-ordination of first and second steps above, and then to place the findings before the top Management for making final decisions

Seasonality

Seasonality is a characteristic of a time series in which the data experiences regular and predictable changes which recur every calendar year. Any predictable change or pattern in a time series that recurs or repeats over a one-year period can be said to be seasonal. It is common in many situations – such as grocery store or even in a Medical Examiner’s office—that the demand depends on the day of the week. In such situations, the forecasting procedure calculates the seasonal index of the “season” – seven seasons, one for each day – which is the ratio of the average demand of that season (which is calculated by Moving Average or Exponential Smoothing using historical data corresponding only to that season) to the average demand across all seasons. An index higher than 1 indicates that demand is higher than average; an index less than 1 indicates that the demand is less than the average.

Cyclic behaviour

The cyclic behaviour of data takes place when there are regular fluctuations in the data which usually last for an interval of at least two years, and when the length of the current cycle cannot be predetermined. Cyclic behavior is not to be confused with seasonal behavior. Seasonal fluctuations follow a consistent pattern each year so the period is always known. As an example, during the Christmas period, inventories of stores tend to increase in order to prepare for Christmas shoppers. As an example of cyclic behaviour, the population of a particular natural ecosystem will exhibit cyclic behaviour when the population increases as its natural food source decreases, and once the population is low, the food source will recover and the population will start to increase again. Cyclic data cannot be accounted for using ordinary seasonal adjustment since it is not of fixed period.

Why do we forecast

Since forecasting activity typically precedes a planning process one can identify specific reasons for the use of forecasting in organizations. Organizations face a different set of issues while they engage in planning and in each of these, forecasting plays an important role as a tool for planning process. The key areas of application of forecasting are summarized below:

• Dynamic and complex environment: Only if an organization has complete control over market forces and knows exactly what the sale of its products is going to be in the future is there no role for forecasting.
• Short term fluctuation in production: A good forecasting system will be able to predict the occurrence of short fluctuations in demand. Therefore, from this knowledge, organizations can avoid knee-jerk reactions to the unfolding reality. Production planning decisions could utilize this information and develop plans that minimize the cost of adjusting the production system for short term fluctuations.
• Better material management: Since the impending events in an organization are predicted through a forecasting system, organizations can benefit from better material management and ensure better resource availability.
• Rationalized man-power decisions: A forecasting system provides useful information on the nature of resources required, their timing and magnitude.

Therefore, organizations could minimize hiring and lying off decisions. Moreover, better planning on overtime and idle time could also be done based on this information

• Basis for planning and scheduling: With proper forecasting, planning and scheduling activities can be done on a rational basis.
• Strategic decisions: Forecasting plays an important role in long term strategic decision making. This includes planning for product line decisions

When it comes to demand forecasting, the major variables that need to be taken into account, are

• demand – Demand is what you’re trying to forecast, and historical data can help. Changes in demand can be charted as historical trends that are rising, falling, stable, cyclical, or random.
• supply – Supply is another variable that will affect the forecasting of demand for your organization’s products and services. Customer demand may be met by a number of suppliers, including your organization. The number of alternative suppliers in the market and their speed in bringing products to market may influence your estimate of future demand.
• product characteristics – Demand for a product may be affected by characteristics of the product. Demand for commodities is relatively stable, and forecasts for such mature products can cover longer time frames.
• competitive environment – The competitive environment as a factor in demand forecasting refers to the actions of a company and its competitors. How your company is performing in the market will affect your demand forecast.

The steps to creating a demand forecast is

• establish purpose – The first thing you need to define is why you’re conducting the forecast. What is its objective? How will it be used?
• select items – Second, you determine the products for which you’re attempting to forecast demand. You need to decide whether you’ll be forecasting just for this one product, or for competing and complementary products as well.
• determine time horizon – Third, you decide on the time frame of the forecast. Is it going to be short range, mid-range, or long range? Short range usually means less than three months, but can extend to a year; it may be most suitable for job scheduling. Mid-range forecasts range from three months to three years, and are appropriate for business planning and budgeting. Finally, when the time frame of the forecast is greater than three years, it becomes a long-range forecast. This would be appropriate for new products.
• select model – Fourth, you choose among different forecasting models. Different techniques are appropriate for forecasting demand for different products, depending on their nature and on the circumstances. In some cases, it may be possible to use more than one technique. The available models fall into two categories: qualitative and quantitative.
• gather data/make forecast – Fifth, you collect the data you need and produce the forecast. You may want to refine the forecasting model depending on the particulars of what you’re forecasting.
• validate results – Finally, you seek to confirm your results by using multiple methods of forecasting and alternative sources. Forecasts are often wrong, to a greater or lesser degree. A forecast that’s considered “accurate” has a typical margin of error of plus or minus five percent. The margin of error for a more speculative forecast may be plus or minus 20 percent. You can achieve better accuracy and reduce the probability of large errors by combining forecasts that use alternate methods.

Importance of Forecasting

Production and distribution are two main activities of a business enterprise. Demand forecasts tries to maintain a balance between production and distribution policies of the enterprise. With decentralization of functions and increase in the size of the organizations, forecasting of demand is of great value for proper control and co-ordination of various activities.

An efficient demand forecast helps the management to take suitable decisions regarding plant capacity, raw material requirements space and building needs and availability of labor and capital. Production schedules can be prepared in conformity with demand requirements minimizing inventory, production and other related costs.

Demand forecasting also helps evaluating the performance of the sales department. Thus, demand forecasting is a necessary and effective tool in the hands of management of an enterprise to have finished goods of right quality and quantity at right time with minimum cost.

Steps in Forecasting: The following are the main steps in demand forecasting

• Determine the objective of forecast,
• Select the period over which the forecast is to be made,
• Select the technique to be used for forecasting,
• Collect the information to be used,
• Make the forecast.

Techniques of forecasting

Implicit in forecasting is that there exist a pattern in the past demand data which can be extrapolated or generalized for the future with the desired measure of certainty. The demand pattern though regular is found to be stable in statistical sense. Since the only input to the forecasting system is the past history of the demand of an item, no direct information concerning the market, the industry, the economy, the sale of competition and complementary products, products price changes, advertising campaigns and so on is used. Forecasting methods involve construction of suitable mathematical relationship to describe the appropriate demand pattern. Management experts have developed many forecasting techniques to help managers to handle the increasing complexity in management decision making it is tricky and experimental process. No one method of forecasting can be applied to all enterprises. In many cases the decisions are based on a combination of several, if not all of these approaches. Final forecast generally include the contributions of many men of varied experience. The use of particular method depends upon the nature of the enterprise, the products manufactured, information system in use.

Elements of Forecasting: Forecasting consists basically of analysis of the following elements;

Internal factors:

• Past
• Present
• Proposed or future

External Factors:

• Controllable: (a) Past (b) Present (c) Future
• Non controllable: (a)Past (b) Present (c) Future