Basics of forecasting. Types of forecasts Forecasting methods in management theory

SUPPLY AND DEMAND FORECASTING Supply and demand forecasts are made using human resource planning methods and modeling techniques (see Chapter 25). AT large organizations The use of simulation is especially beneficial because

Section 3 Market Demand Forecasting Brief Explanations Brief Explanations Key Tools 16. Market Sizing and Market Creation (Evans) 17. HOOF Approach to Demand Forecasting (Evans) Useful Tools

Forecasting is a method that uses both past experience and current assumptions about the future to determine it. If forecasting is done well, the result will be a picture of the future, which can be used as a basis for planning.

Specialists have developed several specific methods for compiling and improving the quality of forecasts.

The main types of forecasts, often used in conjunction with the planning of the organization's activities, are:

Economic, the essence of which is to anticipate the general state of the economy and sales volume for a particular company of a particular product;

Forecasts of technology development, allowing to predict the economic feasibility of developing new technologies;

Forecasts of the development of competition provide for the strategy and tactics of competitors;

Forecasts based on surveys and research make it possible, using many branches of knowledge, to predict what will happen in complex situations;

Social forecasting is used to predict changes in people's social attitudes and the state of society.

There are the following forecasting methods:

1. Informal. Management uses various sources of written and oral information as an aid to forecasting and setting goals. Methods of collecting verbal (oral) information are often used in the analysis of the external environment. This includes information received from radio and television broadcasts, from consumers, suppliers, competitors, consultants, at sales meetings in professional organizations. Sources of written information about the external environment are newspapers, trade magazines, newsletters, professional journals, and annual reports. Some managers use data on the actions of competitors, obtained by means of industrial espionage.

2. Quantitative, used for forecasting, if there is reason to believe that the activity in the past had a certain trend that can be continued in the future, and when the available information is sufficient to identify statistically significant trends or dependencies. In addition, the manager must know how to use a quantitative model, and remember that the benefits of making a better decision should cover the costs of creating a model.

There are the following typical methods of quantitative forecasting:

Time series analysis, based on the study of events that occurred in the past, is the basis for planning. It can be carried out using a table or graph by drawing points on the coordinate grid corresponding to past events. This method is often used to estimate demand for goods and services, determine inventory requirements, and forecast sales patterns;

Causal (causal) modeling is an attempt to predict what will happen in similar situations by examining the statistical relationship between the factor being considered and other variables. This dependence is called correlation. The closer the correlation, the more predictive the model is.

3. Quality. If the quality of information is insufficient, or management does not understand complex methods, or when a quantitative model is too expensive, management may use quality models forecasting. At the same time, experts who turn to for help predict the future. The most common qualitative forecasting methods are considered to be; the opinion of the jury, the general opinion of the employees of the sales department, the model of consumer expectation and the method of expert assessments.

Jury opinion. This forecasting method involves combining and averaging the opinions of experts in relevant areas.

Aggregate opinion of sales staff. Experienced salespeople often correctly predict future demand. They are intimately familiar with consumers and take into account their recent actions faster than a quantitative model can be built. In addition, a good sales agent at a certain time period often "feels" the market, more accurately than quantitative models.

The consumer expectation model is a prediction based on the results of a customer survey of an organization.

Peer review method - a procedure that allows a group of experts to reach agreement.

NIZHNY NOVGOROD COMMERCIAL INSTITUTE

DEPARTMENT OF ORGANIZATION AND MANAGEMENT

Coursework in Management

on the topic: Forecasting in management: goals, forms, methods.

Is done by a student

group 3-MT

Butov I. A.

teacher Gilman S.A

Nizhny Novgorod 1999

Plan.

Introduction 3

1. The concept and essence of forecasting 5

2. Basic forms and methods of forecasting in a commercial organization 13

2.1.Classification of the main methods

forecasting 14

2.2. Main stages of expert forecasting 18

2.3 Exploratory forecasting 28

2.4 Normative forecasting 31

2.5 Scenario method 34

3. Ways to improve the efficiency and validity of forecasts 42

4. Conclusion 44

References 46

Annex 1 Technology levels 47

Introduction

The process of enterprise management is a continuous development of management decisions and their application in practice. The success of the case largely depends on the effectiveness of the development of these solutions. And before starting any business, it is necessary to determine the purpose of their actions. In the production process, enterprise managers very often have to face critical problems, and the final financial result of the enterprise will depend on how optimally the decision is made.

The need for a solution arises only if there is a problem, which in general is characterized by two states - given (desired) and actual (predicted), and it is forecasting that will be the starting point in the process of making a managerial decision. The mismatch between these states predetermines the need to develop a management decision and control over its implementation.

For forecasting to be most effective, goals must be specific and measurable. That is, for each goal, there must be criteria that would allow assessing the degree of achievement of the goal. Without these criteria, it is not possible to implement one of the main management-control functions. Based on this, we can conclude that a goal, the degree of achievement of which can be quantified, will always be better than a goal formulated only verbally (verbally).

Forecasting is a kind of ability to foresee, analyze the situation and its expected course and changes in the future. Since each decision is a projection into the future, and the future contains an element of uncertainty, it is important to correctly determine the degree of risks associated with the implementation of the decisions made. Risk calculation is also an integral part of forecasting, as a system for assessing possible losses and gains when making a given decision.

It should be noted that the development of a solution is a process of understanding the goals and means and mental discussion and performance of an action that precedes the actual implementation of this action. The volitional factor is one of the factors guiding the process of developing and making a decision. Since the decision may be different, the purpose of the will factor is precisely the choice of one specific solution. 1

In the process of forecasting, one should not be limited to solving problems of economic and mathematical modeling and choosing the optimal solution according to certain criteria from a finite set of alternative solutions.

The concept and essence of forecasting

Forecasting is one of the main links in the management process. If we cannot imagine the expected course of an event, that is, we cannot predict this event, we will not be able to effectively make management decisions and optimize the operation of the enterprise.

World history has many examples of how some great people (statesmen of the past, military leaders, businessmen) made brilliant managerial decisions, for example, the decision to leave an empty Moscow to Napoleon. And these decisions were made for a reason, they contained elements of forecasting, more related to the art of forecasting. The word art in this case is more appropriate, because at that time there was no science of forecasting and management as such. Although some forecasts of the past did not have a clear justification, but were based on the gift of prediction, for example, the ancient teaching of "mantika" was a teaching about predicting future events through earthly events. It seemed to complement astrology, which bases its conclusions on the location of celestial bodies.

However, all this knowledge, with the exception of fairly accurate astronomical forecasts, cannot be considered science in the modern sense of the word, and we can attribute predictions confirmed by subsequent events, at best, to the art of forecasting. 1

But nevertheless, thanks to such forecasts, important management decisions were made, up to the formation of forecasting as a science. This happened in the middle of the 20th century. There is no clear date for the birth of forecasting as a science, but it was closer to the Second World War that the line between forecasting a possible technological process and just fantasy was hardly distinguishable.

Today, the forecast (according to G. Theil) is “a certain judgment regarding the unknown, especially future events.”

One of the first works that preceded the formation of forecasting as a modern science was a collection of technological forecasts published by the famous American metallurgical engineer Furnas in 1936. 1

Jifillan also made a significant contribution to the creation of the science of forecasting. In 1937, he predicted the creation of television as one of the ways to make money. To be more precise, in his work he analyzed the previously published forecasts made by Edison, Steinmetz, himself in the period 1910-1920. and found that at least 75% of them were correct.

In 1952, Gyfillan published a review of the state of technological forecasting, where he first emphasized the principle of demand-opportunity. In this work, he also lists the main stages of forecasting, which later received the name of exploration.

Speaking of technological forecasting, we mean not the traditional, but an expanded understanding of the term technology, which means a wide area of ​​purposeful application of the physical sciences, life sciences and behavioral sciences. 2

See Appendix 1 for technology levels.

Forecasting is impossible without the possession of broad information about the organization as a whole, obstacles to its functioning and development, etc., that is, about the most important points that affect the behavior of the organization and decision-making. Analysis of all this collected information allows us to draw conclusions about the development trends of both the organization itself and its environment, which are mainly of the nature of a forecast, that is, a system of reasoned ideas about the directions of development and the future state of the control object.

The basis of forecasts are: special surveys, other forecasts, probabilistic mathematical analysis and time series analysis, brainstorming, individual surveys of specialists, scenarios in case of unforeseen circumstances. 1

The object of any forecast in management can be economic, social, technical, organizational and other processes occurring both in the organization itself and in its environment. Required:

1) scientific analysis of these processes, identification and analysis of causal and other relationships between them, assessment of the current situation and identification of key problems that need to be addressed;

2) attempts to foresee the future of the organization, namely the conditions in which it will function, the difficulties and the tasks arising from them;

3) analysis and comparison of various options for the development of the organization, its personnel, production and scientific and technical potential.

Thus, forecasting in management means a scientific way of identifying the state and probable ways of development of an organization. Forecasts are developed in the form of qualitative characteristics, and in elementary cases in the form of statements about the possibility or impossibility of the occurrence of any event. These characteristics should include quantitative, point or interval indicators and the degree of probability of their achievement.

The forecast cannot be 100% accurate, it must be supplemented with certain assumptions. Sometimes, when there is not enough material to draw any definite conclusions, assumptions are used as a stand-alone strategy development tool.

As already mentioned, one of the goals of forecasting is to solve problems that have arisen in the course of the organization's activities. This can be seen in the decision-making process structuring scheme (Figure 1.1).

Particular attention should be paid to the description of the problem situation, to find those factors that need to be carefully analyzed and considered when solving. First of all, it is necessary to establish whether they are internal or external in relation to this organization, since the possibilities of influencing these two groups of factors are different.

Internal factors more dependent on the enterprise itself. These include: goals and development strategy, the state of the portfolio of orders, the structure of production and management, financial and labor resources, the volume and quality of work, etc. They form an enterprise single system, the interconnection and interaction of elements of which leads to the achievement of its goals. Therefore, a change in one or more factors can lead to a violation of the properties of the entire system. Therefore, measures of managerial influence should be aimed at maintaining the integrity of this system.

External factors. Since external factors shape the environment in which the organization operates, these factors are inert and cannot be influenced by managers. This is due to the fact that this environment is characterized by greater complexity, dynamism and uncertainty, which makes it difficult to take into account environmental factors when making organizational decisions. In parallel with this, and factors have a different impact on the work of the organization.

For example, suppliers, consumers, competitors, other organizations and institutions of society directly related to the field of activity in which this organization is engaged, have a direct impact on its work, the nature of emerging problems and their solutions. As an example, we can recall the problems of domestic enterprises that arose during the period of destruction of the former system of economic ties and changes in relations between suppliers and consumers of products. In some cases, this led to a halt in production, a radical change in the range of products, and the need to search for new suppliers. Changing tastes and priorities of consumers also cause many problems in an organization that has previously focused its production on meeting their needs. In this regard, the following questions arise: Change the range or quality of products? Looking for new markets? Whether to introduce new types of products and services? etc.

There is also a second group of external factors, which is even more unmanageable by the managers of the organization. It (this group) has an indirect impact on the activities of the organization. It includes:

the state of the economy of the country (or region)

level of scientific, technical and social development

socio-cultural and political environment

significant events for the organization in other countries

other factors

The economic state of the country (region) affects the work of the organization through such environmental parameters as the availability of capital and labor, price levels and inflation, labor productivity, buyers' income, government financial and tax policies, etc. The specific impact will be something like this: inflation leads to a reduction in purchasing power and reduces the demand for the products produced by the organization; an increase in the level of prices for products of related industries causes a corresponding increase in production costs in the organization, which results in an increase in prices for its products and can cause an “outflow” of a certain group of consumers; with a reduction in their income, buyers change the composition and structure of consumption, which can also affect demand; the level of scientific and technological development in the country has a strong influence on the structure of the economy, on the processes of automation of production and management, on the technology by which products are manufactured, on the composition and structure of the organization's personnel, and, most importantly, on the competitiveness of products and technologies. Accounting for numerous and diverse environmental factors, choosing the main ones among them and foreseeing possible changes in their mutual influence is the task of forecasting.

An analysis of the factors that led to the emergence of a problematic situation makes it possible to determine the resources (including temporary ones) that will be associated with the solution of the problem.

In the decision-making process, an assessment of the actions that are taken at its various stages takes place. So at the stage of recognizing the problem, the target setting is most often used, by the deviation from which the problem is judged.

The decision-making phase begins with the collection and processing of the information necessary to develop a course of action. Usually, when solving complex problems, it is not possible to limit oneself to only the information provided by existing reporting systems; therefore, it takes time and resources to inform the solution of the problem.

When predicting managerial decisions, as a rule, the question arises: “Which way of decision to choose from acceptable options?”. The one that is most useful or preferable to meet the objectives of the organization will be selected. The quality of managerial decisions depends on how justified they are chosen, and this, in turn, determines the competitiveness of the organization, the speed of its adaptation to changes in the economic situation and, ultimately, efficiency and profitability.

Basic forms and methods

forecasting in

commercial organization

Classification of the main forecasting methods.

The main stages of expert forecasting.

Exploratory forecasting.

Normative forecasting.

scripting method.

1.Classification of the main methods

forecasting.

The generally accepted and main forecasting methods are:

1. Expert forecasting;

2. Survey forecasting;

3. Normative forecasting;

4. Scenario method.

Technological forecasting is divided into exploratory (sometimes also called exploratory) and normative.

Exploratory forecasting is based on the orientation towards the opportunities presented, the establishment of trends in the development of situations on the basis of information in the development of the forecast.

Moving in the technology space from lower-level technologies to higher-level technologies is referred to as exploratory forecasting. Or otherwise it can be said that the needs and goals should correspond to the means and capabilities of a commercial organization.

An example of exploratory forecasting is forecasting in the field of electronics, when the predicted process is presented as a successive movement of technologies, starting from quantum electrodynamics and ending with instantaneous worldwide communications. 1

Normative forecasting is focused on the mission of the organization, on those needs and goals that it seeks to achieve. Normative forecasting corresponds to the movement in the space of technologies from technologies to more high levels to technologies of a lower level, that is, from needs and goals to the means of their implementation.

An example of normative forecasting can be forecasting in the field of outer space, when the predicted process is presented as a sequential transfer of technologies from understanding the problem of outer space as an environment that should serve for the benefit of man, to specific means of solving it - the conditions for nuclear fission and the amount of energy released during this and etc.

Within the framework of technological forecasting, such tasks as the development of forecasts in the field of economic, commercial, social and political activities are solved.

One of the main problems of the accuracy and efficiency of forecasts is the most useful combination of exploratory and normative forecasting methods. This is a consequence of the differences in the methods used. So for exploratory forecasting, it is typical to use such methods as:

extrapolation;

modeling;

method of historical analogy;

script writing;

other methods;

based on the analysis of accurate empirical data. When using exploratory forecasting methods, preference is given to quantitative information. The use of qualitative (non-quantitative) information in exploratory forecasting is also possible.

An example of this is the use of intuitive methods, the same method of scenarios or the method of expert curves, which make it possible to determine the emerging trends in changing the situation, based not only on empirical data, but also on the experience of highly qualified experts.

The main methods used in normative forecasting are, first of all, the methods of PATTERN, Delphi, Glushkov, Pospelov, etc.

The currently widely used toolkit, goal trees, first appeared as part of the PATTERN method (planning justification through the scientific and technical evaluation of quantitative data), developed in 1936 for aeronautics and space.

New types of forecasting include forecasting using feedback, intuitive methods, "bypass" methods, etc. But the main ideas used in the development of forecasts are quite fully represented in exploratory and normative forecasting.

It should also be noted that a very important point (both for the collection process and for the processes of data analysis and processing) is to determine whether the information is quantitative or non-quantitative (qualitative).

Quantitative information, if reliable enough, has the advantage that it allows the use of accurate mathematical methods and models and determines the development trends of the situation with a certain accuracy, indicating confidence intervals, possible errors in calculations, etc. But even more significant is the circumstance that the range of problems for which it is possible to develop adequate mathematical models turns out to be much narrower than the set of situations in which it is necessary to make real decisions.

Much more often when developing forecasts, one has to deal with qualitative information.

When developing a forecast, they include situations where data are presented in the form of verbal (verbal) descriptions, when estimates are obtained using verbal or verbal-numerical scales, when there is information only about comparative estimates of alternative options.

There are also situations when the obtained quantitative information cannot be "fitted" into any of the existing mathematical models, and can also be analyzed using specially developed methods of qualitative analysis.

In recent years, expert forecasting has been developed, focused to a greater extent on working not only with quantitative, but also with qualitative information obtained directly from experts.

2.Main stages

expert forecasting.

With this method of forecasting, most of the problems that arise in the development of forecasts can be solved. There are several main stages in expert forecasting, which are shown in Fig. 2.2.1. 1

Block diagram of the main stages of forecast development.

Fig.2.2.1

1. At the stage of preparation for the development of the forecast, the following tasks should be solved:

organizational support for the development of the forecast was prepared,

the task for the forecast is formulated,

the working and analytical support groups were formulated,

an expert committee was formed,

prepared methodological support for the development of the forecast,

prepared information base to make a forecast

computer support for the development of the forecast was prepared.

After a decision is made to develop a forecast, it is necessary to appoint executors for this development. This group of workers is entrusted with organizational support for the development of the forecast. They must also provide methodological and informational support.

A high-quality expert forecast can be developed only when it is well prepared, if competent specialists are involved in its development, when reliable information is used, when the estimates are obtained correctly and processed correctly.

To develop a high-quality forecast, it is necessary to use modern technologies that accompany and support the development process.

Experts who are professionally familiar with the object of expertise are invited to the composition of the expert commission. If a multidimensional assessment of an object is required, or heterogeneous objects should be assessed and this requires specialists of different professions, then the expert commission should be formed in such a way that it includes specialists who are able to professionally assess all the main aspects of the predicted problem.

The task of the analytical group is the methodical preparation of the forecasting process. The analytical group includes specialists with professional knowledge and experience in predictive development. The development of the forecast must be carried out methodically competently, the methods used must correspond to the nature of the forecasted situation and the nature of the information to be obtained, analyzed and processed. Also, the development of a forecast should be clearly regulated, that is, the working group should prepare the necessary documentation, which includes: an official decision to conduct a forecast, the composition of an expert commission, a schedule for developing a forecast, contracts with specialists involved in its development, etc. Specialists should be provided with all the necessary information about the object of forecasting. A dedicated analytical review prepared by the reflection team on the foreseeable problem may be useful. When working with multivariate forecasts, one has to deal with large amounts of information, which, moreover, must be analyzed and processed in accordance with the forecast development technology used. This cannot be done without a computer and appropriate software.

2. When analyzing retrospective information about the forecasting object, a clear separation of quantitative and qualitative information is assumed. Quantitative information (rather reliable) is used for extrapolation of the dynamics of changes in predicted parameters, to determine the most likely trends in their changes. Qualitative information is classified, systematized and serves as the basis for expert assessments and is used to develop expert forecasts. When developing a forecast, it is necessary to analyze the internal conditions of the object of forecasting, a meaningful analysis of their features and dynamics of development.

If mathematical, simulation, analog, etc. are developed. models of the functioning of the forecasting object and changes in internal conditions, then the necessary data are entered into them and, on their basis, calculations are made to assess the most probable changes in the internal conditions of the forecasting object.

When developing a forecast, external conditions, the external environment of the functioning of the object of forecasting should be given no less attention than internal ones.

The internal environment, as an internal condition of the forecasting object, includes: intra-organizational processes, technology, personnel, organizational culture, management of functional processes. External includes the general external environment and the direct business environment of the organization.

3. Determination of the most probable options for the development of internal and external conditions of the object of forecasting is one of the central tasks of developing a forecast. At this stage of forecast development, based on the analysis of internal and external conditions and all available information about the object of forecasting, information as a result of the work of the expert commission, a list of possible alternative options for changing internal and external conditions is preliminarily determined. After their preliminary assessment, alternative options are excluded from the list, the feasibility of which in the forecast period is doubtful or the probability of their implementation is below a pre-set threshold. The remaining alternatives are subjected to a more in-depth assessment in order to identify alternatives for changing internal and external conditions, the implementation of which is most likely.

4. At this stage of the development of the forecast, the most active work of experts is expected to identify and evaluate the key events that are expected to occur in the forecasted period of time.

The previous stage of development of the forecast provides the information necessary for the analytical team to conduct an examination. Experts are provided with information on the most likely change in internal and external conditions, based on the previous analysis, questions are formulated that should be answered as a result of the examination, and the most likely scenarios for the development of events are outlined.

Depending on the nature of the object of forecasting, on the nature of the assessments and judgments that must be obtained in the process of conducting an examination, specific methods for organizing and conducting an examination are determined. Examinations can be single-round and multi-round, anonymous and providing for an open exchange of opinions, etc.

A variety of methods are used in the comparative evaluation of objects, in predicting the quantitative and qualitative values ​​of the parameters of the predicted object, ranging from various modifications of the Delphi method to various procedures of the brainstorming method. The nature of the expert information that is supposed to be used in the development of the forecast imposes certain requirements on the choice of a specific method for organizing and conducting an examination. If the predicted object is quite complex and multifaceted, then it is advisable to use complex methods of organizing and conducting an examination during the examination to develop a forecast, the analytical group can use questionnaires and interviews.

5. The information prepared at the previous stages, including that received from experts, is used in the direct development of the forecast. As a rule, cases are unlikely when it is known in advance in which direction the changes in internal and external conditions will occur, which strategy will be chosen by the organization in a particular development of events. After all, the development of the organization in the predicted future depends on various factors, as well as on their combination and interaction. From this we can conclude that in strategic planning and in other cases of using forecasts, it is necessary to consider various alternative scenarios for the development of events, both favorable and unfavorable.

At the previous stages, the most probable changes in the main internal and external conditions that determine the course of predicted events were determined. For the most probable alternatives, their changes, the most probable alternatives for the development of predicted events should be developed.

Suppose one of the goals of developing a forecast is to determine the dynamics of development of quantitative indicators and parameters, then the amount of information obtained at the previous stages of developing a forecast is used (quantitative and qualitative) and the corresponding extrapolation methods (determining changes in predicted indicators and parameters in the future), curves of their change are calculated within the predicted time frame. But we do not always have the necessary information to use quantitative extrapolation methods. This feature is characteristic of the current stage of the economic life of Russia, the lack of statistical data necessary for the calculations, since the previous economic dependencies and patterns have changed. Therefore, as a rule, the only way to extrapolate indicators and parameters for a predicted period of time is the method of constructing expert curves. These curves reflect an assessment of the dynamics of changes in the predicted values ​​of indicators

And parameters by experts. They (experts) determine the critical points at which the trend in the values ​​of predicted indicators and parameters may change under the influence of various factors. And then, at each of the critical points located on the time axis, the expected values ​​of the predicted indicators and parameters are evaluated, as well as the nature of their change in the interval between the two critical points.

An example of a discrete expert curve is shown in Figure 2.2.2.

Fig.2.2.2.

When developing a variant forecast, extrapolation of the predicted values ​​of indicators and parameters for various variants of the initial conditions and for various variants of possible alternative variants of the dynamics of their changes should be extrapolated. At the same time, each alternative version of the developed forecast can be accompanied by a description of the predicted development of events.

6. A priori and a posteriori assessment of the quality of the forecast. Evaluation of the quality of the forecast is one of the central problems in the process of developing managerial decisions. The degree of confidence in the developed forecast largely affects the decision and affects the effectiveness of management decisions made using the developed forecast.

However, assessing the quality of a forecast is a rather difficult task not only at the moment when the forecast has just been developed (a priori estimate), but also at the moment when the predicted event has already occurred (a posteriori estimate). It should also be noted here that a qualitative forecast can be used in different ways when making a decision.

If the management of the organization does not have a significant impact on the course of events, but only monitors it, then after the end of the forecast period, it is only necessary to compare the values ​​of the predicted indicators and parameters with those obtained in reality. This makes it possible to evaluate the quality of the developed forecast a posteriori.

After the development of the forecast, criteria must be defined by which the accuracy of the forecast can be assessed. Usually, two methods are used to evaluate the forecast: differential and integral.

The integral method assumes a generalized assessment of the quality of the forecast based on the assessment of the quality of the forecast according to particular criteria. With the differential method, sets of estimates of individual components of the quality of the forecast are evaluated, which have a fairly clear objective meaning. These criteria can be: the clarity and clarity of the task for the forecast, the correspondence of the forecast to the task, the timeliness of the development of the forecast, the professional level of the development of the forecast, the reliability of the information used, etc.

An example of the use of the integral method is the criterion "integral quality of expert forecast".

The quality of an expert forecast is determined by such criteria as:

competence (or, more generally, quality) of an expert;

quality of information provided to experts;

the quality of expert information coming from experts;

level of forecast development technology.

If the forecasting period has already ended, then it is necessary to compare the predicted values ​​of indicators and parameters with those obtained as a result of the actual course of predicted events.

And here the question comes to the fore - by what criterion to evaluate the quality of the forecast a posteriori. As an example of criteria for assessing the accuracy of a forecast, the following formula can be given:

K 1 \u003d │X-I │ K 2 \u003d │lnX / And │,

where X is the predicted value of the indicator score,

U is the true value of the indicator score.

7. After the forecast is prepared and presented to the management of the organization, the customer, etc. the stage of post-forecast work with the prepared material begins.

Variant development of the forecast involves the development of a forecast under various alternative options for conditions and assumptions. And they can change. Events that seemed unlikely yesterday are happening today, and those that seemed most likely do not happen.

Therefore, an integral part of modern forecasting technology is periodically, depending on the ongoing changes, monitoring of the implementation of the predicted course of events. Monitoring allows timely detection of significant deviations in the course of events. If they can have a fundamental impact on the further course of events in terms of making important strategic decisions, then the forecast should be subject to adjustment.

Adjustments can be of different levels of significance, complexity, labor intensity, etc. If they are not very significant, then this problem can be solved at the level of the analytical group accompanying the development of the forecast. If the adjustments are more significant, then additional involvement of individual experts may be required, and in especially important cases, if there are significant changes, additional work of the expert commission with a possible change in its composition. The latter is necessary, in particular, in those cases when the correction of the forecast requires the involvement of specialists of a different professional orientation.

3. Survey forecasting.

One of the main methods used in exploratory forecasting is the extrapolation of time series-statistical data about the object of interest to us. Extrapolation methods are based on the assumption that the law of growth that took place in the past will continue in the future, taking into account corrections due to the possible saturation effect and the stages of the object's life cycle.

Among the curves that accurately reflect the change in the predicted parameters in a number of common situations is the exponent, that is, a function of the form:

where t is time,

a and b-parameters of the exponential curve.

Among the most famous exponential curves used in forecasting is the Pearl curve, derived from extensive research in the field of growth of organisms and populations, and having the form:

y=L/(1+ae-bt),

where L is the upper limit of the variable y.

No less common is the Gompertz curve, derived from the results of research in the field of income distribution and mortality (for insurance companies), which has the form:

where k is also the exponent parameter.

The Perl and Gompertz curves were used to predict such parameters as the increase in the efficiency of steam engines, the increase in the efficiency of radio stations, the increase in the tonnage of merchant fleet ships, etc.

Both the Pearl curve and the Gompertz curve can be classified as so-called S-shaped curves. Such curves are characterized by an exponential or close to exponential growth at the initial stage, and then, when approaching the saturation point, they take on a flatter shape.

Many of the mentioned processes can be described using the corresponding differential equations, the solutions of which are the Pearl and Gompertz curves we have considered.

As an example, we can cite a differential equation that describes the increment in the amount of information (knowledge) I depending on the number of researchers N, the average productivity coefficient of one researcher q per unit time t, and C- a constant coefficient characterizing the dynamics of changes in the amount of information. It looks like this:

.

Integrating this differential equation, we obtain a formula for the amount of information:

In general, the dynamics of changes in predicted indicators and parameters over time can be represented as:

,

where y(t) is a trend function describing the parameter change trend,

e(t) is a random function that characterizes the deviation of the predicted variable from the trend.

Extrapolation uses regression and phenomenological models. Regression models are built on the basis of the established patterns of development of events using special methods for selecting the type of extrapolating function and determining the values ​​of its parameters. In particular, the least squares method can be used to determine the parameters of the extrapolating function.

Assuming the use of one or another extrapolation model, one or another distribution law, it is possible to determine confidence intervals that characterize the reliability of predictive estimates.

Phenomenological models are built on the basis of the conditions of maximum approximation to the trend of the process, taking into account its features and limitations, and accepted hypotheses about its future development.

With a multi-factor forecast in phenomenological models, it is possible to assign large weighting factors to factors that in the past had a greater influence on the development of events in the past.

If, when forecasting, a retrospective period is considered, consisting of several periods of time, then, depending on the nature of the forecasted indicators, less distant from the moment of forecasting on the time scale, etc. It should also be taken into account that often, when forecasting, experts' assessments of the near future may be overly optimistic, and assessments of the more distant future may be overly pessimistic.

If several different technologies can participate in the predicted process, each of which is represented by a corresponding curve, then the envelope of partial curves corresponding to individual technologies can be used as the resulting expert curve.

4. Normative forecasting.

Normative forecasting is an approach to developing a forecast based on the goals and objectives that an organization sets for itself in the forecast period. The main method used in normative forecasting is the method of horizontal decision matrices, when the priority of the implementation of the projects proposed to achieve the set goals is determined.

Usually two-dimensional and three-dimensional matrices are used. Most often, horizontal decision matrices are used to determine the optimal allocation of resources under given constraints. At the same time, cash, labor, its quality and qualifications, equipment, energy resources, etc. can act as resources.

In particular, one dimension of the horizontal decision matrix may correspond to the main problems that arise in achieving the goal, the second dimension to the resources that may be required to solve these problems.

The agreed matrices of the lower hierarchical levels of problems are combined into matrices of higher levels up to the main matrices for the strategic problems of the organization.

In a three-dimensional horizontal decision matrix, one dimension, for example, may correspond to commercial missions (sales areas), the second to resources, the third to time. Resources, in turn, can be divided into financial, commercial, sales, production, equipment, etc. resources.

Vertical decision matrices are designed to track the vertical movement of technologies. The vertical decision matrix for intra-company planning according to the recommendations of the Stanford Institute may look something like this (Fig. 2.4.1.):

Fig.2.4.1.

In particular, a three-dimensional vertical decision matrix called "Overall Scheme for the Development of the National Space Program System" was developed by North American Aviation.

For a more rational selection of projects for implementation, operations research methods can be used, such as:

linear programming, which allows one to formulate an optimization problem in the form of linear constraints (inequalities or equalities) and a linear objective function;

dynamic programming, designed to solve multi-stage optimization problems;

integer programming, which allows solving optimization problems, including problems of optimal resource allocation, with discrete (integer) values ​​of variables, etc.

The normative forecasting tools include methods for constructing goal trees, methods of the PATTERN type, etc.

In this case, each of the goals under consideration is assigned quantitative weighting factors, and for each project, the contribution to the achievement of each of the goals is estimated, if it is non-zero. The degree of contribution is subsequently multiplied by the target's weighting factor. This procedure can be illustrated by the following example (Figure 2.4.2) 1 :

Fig.2.4.2.

Naturally, for implementation, it is advisable to choose a project that represents the greatest value.

5. SCENARIO METHOD.

In the development of managerial decisions, the scenario method is widely used, which also makes it possible to assess the most probable course of events and the possible consequences of the decisions made.

Scenarios for the development of the analyzed situation developed by specialists allow, with one level of certainty or another, to determine possible development trends, relationships between acting factors, to form a picture of possible states that the situation may come to under the influence of certain influences.

Professionally developed scenarios allow you to more fully and clearly determine the prospects for the development of the situation, both in the presence of various control actions, and in their absence.

On the other hand, scenarios of the expected development of the situation make it possible to realize in a timely manner the dangers fraught with unsuccessful managerial actions or unfavorable developments.

It is argued that the need to foresee the most probable development of the situation first arose with the emergence of industrial production, since there was no need for this with seasonally repeating agricultural production.

It is difficult to fully agree with this point of view, since from time immemorial mankind has fought, from time to time carried out grandiose construction. And without an idea of ​​the possible development of the situation, such targeted actions would hardly have been possible.

At the same time, we often find prototypes of the scenario method at different times in different countries.

So Kutuzov, who gathered a military council in Fili, and listened to various options for possible actions, evaluated various scenarios for the development of the war with the French, proposed by the military leaders.

He compared their strengths and weaknesses and came to a difficult, but perhaps the only right decision to leave Moscow, dooming it to fires and destruction.

However, subsequent developments proved him right. The scenario he preferred for the development of events fully justified itself.

A statesman occupying a responsible position and a businessman making an important decision for the fate of the project, a financier analyzing the stock market, a surgeon on the eve of a complex unconventional operation, a designer laying the foundations of a fundamentally new facility when making important decisions, as a rule, try to predict a possible scenario for the development of events in order to make a decision that ensures success.

It is believed that the first scenarios for predicting the development complex systems used by Herman Kahn. The first scenarios developed were mostly descriptive.

Subsequently, the scenario method has been largely developed through the use of more accurate qualitative-quantitative models.

The scenario method involves the creation of scenario development technologies that provide a higher probability of developing an effective solution in situations where it is possible, and a higher probability of minimizing expected losses in situations where losses are inevitable.

Currently, various implementations of the scripting method are known, such as:

obtaining a consensus

iterative procedure of independent scenarios,

use of interaction matrices, etc.

The method of obtaining a consensus opinion is, in essence, one of the implementations of the Delphi method, focused on obtaining the collective opinion of various groups of experts on major events in a particular area in a given period of the future.

The disadvantages of this method include insufficient attention paid to the interdependence and interaction of various factors influencing the development of events, the dynamics of the development of the situation.

The method of iterative combination of independent scenarios consists in the compilation of independent scenarios for each of the aspects that have a significant impact on the development of the situation, and the repeated iterative process of coordinating scenarios for the development of various aspects of the situation.

The advantage of this method is a more in-depth analysis of the interaction of various aspects of the development of the situation.

Its disadvantages include the insufficient development and methodological support of scenario coordination procedures.

The method of mutual influence matrices, developed by Gordon and Helmer, involves the determination, based on expert assessments, of the potential mutual influence of events in the population under consideration.

Estimates linking all possible combinations of events by their strength, distribution in time, etc., make it possible to refine the initial estimates of the probabilities of events and their combinations. The disadvantages of the method include the laboriousness of obtaining a large number of estimates and their correct processing.

The paper proposes a methodology for compiling scenarios, which involves a preliminary definition of the space, parameters that characterize the system.

The state of the system at a point in time t is a point of S(t) in this parameter space. Determination of possible trends in the development of the situation makes it possible to determine the probable direction of evolution of the position of the system in the space of the identified parameters S(t) at various points in time in the future S(t+l), S(t+2), etc.

If there are no control actions, then it is assumed that the system will evolve in the most probable direction.

Control actions are equivalent to the action of forces capable of changing the direction of the trajectory S(t).

Naturally, the control actions should be considered taking into account the limitations imposed by both external and internal factors.

The proposed technology for developing scenarios involves considering the position of the system at discrete times t, t+1, t+2, ... .

It is assumed that the point corresponding to the system S(t) in the parameter space is located in a cone that expands with distance from the initial time t.

At some time t+T, the system is expected to be located in the section of the cone corresponding to time t+T.

Control actions lead to a shift in the position of the system in the parameter space. In this case, it is also advisable to consider only discrete points, paying the most attention to the most probable points. In such an analysis, it is necessary to anticipate the possibility of additional internal stresses between the elements of the system, since they can also change the position of the system in the parameter space.

To assess stresses, appropriate indicators can be used, in particular, of an economic or social nature, as well as threshold values ​​of indicators, when exceeded, the position of the system can change significantly.

Control actions in some cases can be aimed at preventing exceeding the threshold values ​​of indicators, if our goal is to maintain stability.

In some cases, it is possible to purposefully strive to exceed the threshold values ​​of indicators, if this corresponds to the tasks set for the system.

One of the most important results of using this version of the scenario method, as well as its other varieties, is a better understanding of the analyzed situation and the main patterns and features of its development.

Noteworthy is a variation of the scenario method proposed by Abt, Foster, and Rea.

Indeed, a deeper understanding of the situation obtained in the process of developing a forecast suggests, as a next step, the development of a system of influences that can change the considered scenarios for the development of the situation. And the probable future may turn out to be corrected.

The method developed by the authors provides for the selection of only those variables that are directly related to the development of the analyzed system, whether it is a control system for environment or a process control system in existing production, etc.

Next, it is planned to develop sufficiently detailed scenarios to identify the dangers that threaten the system and the necessary counteraction to them. It is envisaged to select among the many possible scenarios the most suitable for further analysis, as well as procedures for using computers to develop undistorted scenario forecasts.

Let's consider the listed procedures in more detail. Before proceeding with the development of the scenario, it is supposed to analyze the situation with the definition of the main acting forces, the main relationships between the main factors acting in it, the necessary detail and structuring of the situation.

The selection of variables in this method involves the use of experts.

Experts' forecasts of the development of the situation are analyzed, with the possible use of content analysis, and the variables that are part of the logical reasoning of experts, and their interrelations, are highlighted.

The main task in this case is to obtain a set of essential variables that fully determine the development of the analyzed situation.

The next step is to determine for each variable the appropriate scale in which it could be measured.

Since in real situations, along with quantitative variables, qualitative ones are also used, it is planned to develop a verbal-numerical scale for each variable, containing both numerical values ​​of gradations and their meaningful description.

Quantitative values ​​of variables allow more reliable identification of possible hazards.

If the variables are continuous, then it is advisable to highlight their characteristic values ​​for use in the analysis of the situation.

In some cases, information about variables can be presented in the form of a thesaurus, which reflects the basic information, both quantitative and descriptive, which allows a sufficiently complete representation of the variable.

An unjustified increase in the number of variables makes it difficult to analyze the situation, while their excessive generalization (aggregation) also makes the analysis difficult.

The main task of the script is to provide a clue to understanding the problem. When analyzing a specific situation, the variables characterizing it take on the appropriate values ​​- certain gradations of the verbal-numerical scales of each of the variables.

All values ​​of pairwise interactions between variables that have mutual influence in the development of a given situation are determined.

This interaction between variables is usually represented in matrix form.

After developing and presenting the scenario using variables and evaluating their interaction and internal consistency, it is possible, using verbal-numerical scales, to proceed to presenting the scenario in the form of a meaningful description.

This form is often more convenient when preparing a progress report. Sometimes it is expedient to include in the composition of the scenario the prehistory of the development of the analyzed situation.

A distinctive feature of the presented method is its multivariance, i.e. consideration of several alternative options for the possible development of the situation, taking into account baseline scenarios.

By grouping scenarios into classes, you can determine a rational strategy for influencing the situation.

As a rule, data on several possible scenarios for the development of a situation is more informative than one single scenario and contributes to making more effective decisions.

A feature of this method is also that it is possible to estimate the values ​​of the interaction of variables only at the boundaries of the area of ​​​​admissible values, and not over the entire area, as is assumed in the method using the matrix of mutual influences.

The use of special programs for computers, as well as random number generators, followed by cutting off impossible situations to generate alternative scenarios, expands the horizon for analyzing possible situations in the future.

The developed wide range of possible alternative scenarios for the development of the situation allows you to more fully define critical situations for decision-making, as well as determine the possible consequences of the proposed; alternative solutions in order to compare them and choose the most effective one.

A professionally developed and periodically updated forecast is an integral part of the process of developing and making important management decisions.

Ways to improve efficiency

and validity of forecasts

To the greatest extent, the effectiveness of the forecast depends on how useful they are for planning and executing business operations. Forecasts are useful when their components are carefully considered and the limitations contained in the forecast are frankly named. There are several ways to do this.

Ask yourself what the forecast is for, what decisions will be based on it. This determines the required forecast accuracy. Some decisions are dangerous to make, even if the possible forecast error is less than 10%. Other decisions can be made without fear, even with a much higher error tolerance. Determine the changes that must occur in order for the forecast to be reliable. Then carefully evaluate the likelihood of the relevant events. Define the components of the forecast. Think about data sources, determine how valuable past experience is in making a forecast. Isn't the change so fast that an experience-based forecast would be useless? Does the data on similar products (or development options) provide a basis for making predictions about the fate of your product? How easy or inexpensive will it be to obtain reliable information about past experiences? Determine how structured the forecast should be. When forecasting sales, it may be appropriate to identify separate parts of the market (developing customers, stable customers, large and small customers, the likelihood of new customers, etc.). 1

Also, by increasing the efficiency of forecasts is the use of break-even analysis. This analysis determines the point at which total revenue equalizes with total costs, that is, the point at which the enterprise becomes profitable.

The break-even point refers to the situation at which the total income becomes equal to the total costs. To determine the break-even point, three main factors must be taken into account: the selling price of a unit of production, variable costs per unit of production, and general fixed costs per unit of production:

,

where BEP is the breakeven point;

TFC - total fixed costs;

P is the price of a unit of production;

VC-variable costs per unit of output.

Conclusion.

From the foregoing, we can conclude that under the current conditions of the functioning of a market economy, it is impossible to successfully manage a commercial firm without effective forecasting of its activities. The extent to which forecasting is accurate and timely, as well as consistent with the problems posed, will ultimately depend on the profits received by the enterprise.

In order for the forecast effect to be as useful as possible, it is necessary to create so-called forecasting departments at medium and large enterprises (for small enterprises, the creation of these departments will be unprofitable). But even without such departments it is impossible to do without forecasting. In this case, the forecast must be obtained by the forces of managers and the specialists involved in this process.

As for the forecasts themselves, they must be realistic, that is, their probability must be high enough and correspond to the resources of the enterprise.

To improve the quality of the forecast, it is necessary to improve the quality of the information necessary for its development. This information, first of all, should have such properties as reliability, completeness, timeliness and accuracy.

Since forecasting is a separate science, it is advisable (to the extent possible) to use several forecasting methods when solving a problem. This will improve the quality of the forecast and will allow you to identify "pitfalls" that may not be noticed when using only one method.

It is also necessary to correlate the received forecast with the precedents in solving this problem, if such took place under similar conditions of functioning of a similar organization (competitor). And with a certain adjustment, in accordance with this precedent, make decisions.

List of used literature.

M. Mescon, M. Albert, F. Hedouri. Fundamentals of Management.-Moscow "Delo" 1994

General course of management in tables and graphs.-ed. B.V. Prykina.-Moscow "Banks and exchanges" 1998.

Management of organizations.-tutorial, ed. Z.P. Rumyantseva.-Moscow Infra-M 1995.

Management (MODERN RUSSIAN MANAGEMENT).-ed. F.M. Rusinova.-Moscow FBK-Press 1998.

Economic strategy of the firm.-ed. A.P. Gradova.-St. Petersburg "Special Literature" 1995.

B.G. Litvak. Management decisions.-Moscow EKMOS 1998.

V.R. Vesnin. Fundamentals of Management.-Moscow "Triada LTD" 1996.

Attachment 1.

Technology levels.

Technologies are divided into 8 levels, starting from the emergence of the first idea (the level conditionally called "Scientific resources") and ending with its wide implementation in society (the level conditionally called "society").

The main levels of technology are presented in Table 1 in ascending order of their social significance and the stage of implementation of the idea.

The transition from a lower level of implementation of an idea to a higher one is called technology transfer. Important ideas that have been realized are characterized by a consistent movement of technologies from the lowest to the highest level.

Table 1.

1 N.L. Kardanskaya. Making a managerial decision. page 10

1 B.G. Litvak. Management decisions. page 112 Management Methods forecasting major financial indicators (2)Abstract >> Finance

That it embodies the developed strategic goals in form specific financial indicators; enables... , integral methods. As part of the financial management these methods are mainly used to evaluate and forecasting financial...