CC BY
22
ЕКОНОМ1КА: реалп часу
№1(23), 2016
ECONOMICS: time realities
TEOPm I ПРАКТИКА ЕKOHOМIKИ ТА УПРАВЛ1ННЯ ПРOМИСЛOBИМИ ПВДПРИбМСТВАМИ
THEORY AND PRACTICAL ASPECTS OF ECONOMICS AND INDUSTRIAL ENTERPRISES MANAGEMENT
UDC 338.27
METHODS OF SALES FORECASTING IN A MARKETING INFORMATION SYSTEM OF THE INDUSTRIAL ENTERPRISES
O.G. Yankovoy, Ph.D., Professor
Odessa National Economic University, Odessa, Ukraine O.I. Yashkina, Ph.D. in Economics, Doctor of Economics, docent
Odessa National Polytechnic University, Odessa, Ukraine
To collect marketing information on industrial organization in the department of marketing functions marketing information system. The purpose of its functioning is to collect information about factors of marketing environment, analyzing, identifying the need for marketing research and their implementation, as well as forecasting of market reaction to marketing action company. The operation of a marketing information system reduces the risks of decision making in marketing.
Trend life cycle of high-tech products in modern conditions are such that the residence time of the goods in circulation is reduced, but the costs of development and implementation of products in production, on the contrary, increase. Therefore, in our opinion, the development of forecasting tools for marketing the enterprise information system that will provide accurate and reliable forecasts on future sales of products on the market is relevant.
Analysis of recent researches and publications
The scientific approaches to forecasting usually a handful of sales in two groups of methods: expert-based assessments of the experts concerning future sales [1-4]; statistical, based on extrapolation by statistical methods revealed trends for the future [5-7, 13]. In each of these groups there are many approaches for obtaining forecasts of different market situations and methods to assess the reliability of the obtained predictions.
Unsolved aspects of the problem
In our view, the scientific literature does not consider the issues of determination of the forecasting methods used in marketing information system, in connection with the stages of the innovation process.
Янковий О.Г., Яшкта O.I. Методы прогнозування збуту в маркетинговш тформацштй cucmeMi промысловых тдприемств.
У статт запропоновано науково-методичний тдхщ вибору методу прогнозування збуту шновацшно! продукцп в залежност вщ етапу шновацшного процесу. На кожному етат шновацшного процесу: фундаментальн дослщження, прикладн дослщження, експериментальн роботи, впровадження та дифузiя видшено види прогнозiв обсяпв витрат на шновацп та показниюв результативност шновацшно! дiяльностi. Для кожного з етатв шновацшного процесу визначен джерела отримання прогнозiв (експерти або дан поточного продажу) та пов'язан з ними методи прогнозування. Придшено увагу iмiтацiйним методам прогнозування на етат дифузп шновацш, за якими отримуються окремi прогнози збуту шновацшно! продукцп.
Ключовi слова: шновацшний процес, шновацшна продукщя, методи прогнозування збуту, експертн оцшки, дифузiя шновацш
Янковой А.Г., Яшкина О.И. Методы прогнозирования сбыта в маркетинговой информационной системе промышленных предприятий.
В статье предложен научно-методический подход выбора метода прогнозирования сбыта инновационной продукции в зависимости от этапа инновационного процесса. На каждом этапе инновационного процесса: фундаментальные исследования, прикладные исследования, экспериментальные работы, внедрение и диффузия выделены виды прогнозов объемов затрат на инновации и показателей результативности инновационной деятельности. Для каждого из этапов инновационного процесса определены источники получения прогнозов (эксперты или данные текущих продаж) и связанные с ними методы прогнозирования. Уделено внимание имитационным методом прогнозирования на этапе диффузии инноваций, по которым получаются отдельные прогнозы сбыта инновационной продукции..
Ключевые слова: инновационный процесс, инновационная продукция, методы прогнозирования сбыта, экспертные оценки, диффузия инноваций
Yankovoy A., Yashkina O. Methods of sales forecasting in a marketing information system of the industrial enterprises.
The paper proposes a scientific and methodical approach for choosing a method of innovative products' sales forecasting depending on the stage of the innovation process. For every stage of the innovation process (basic research, applied research, experimental studies, implementation and diffusion) forecast types of volumes of expenses for innovation and performance indicators of innovation are marked. For each of the stages of the innovation process sources of projections (expert review and sales data) and related methods of forecasting are identified. Special attention is paid to method of forecasting simulation at the stage of diffusion of innovations, which are obtained by the individual forecasts of sales of innovative products..
Keywords: innovative process, innovative products, sales forecasting methods, expert review, diffusion of innovations
Teopia i nparrma eKOHOMÍKH Ta ynpaBmHHS npoMHC^oBHMH nignpneMCTBaMH
The aim of the article is the provision of scientific and methodological tools for identifying methods to make predictions in the marketing information system of the industrial enterprise depending on the stage of the innovation process. The main part
The innovation process consists of several stages: basic research, applied research, experimental work, deployment and diffusion. The first three stages is R&D, the last two commercial phase of the innovation process in which results R&D is perceived or ignored by the market. The information provided by the experts at the first three stages, on projected sales of innovative products, the potential of technology transfer and other indicators of innovative activity has a high degree of "fuzziness", that is, the likelihood that experts in these stages is quite low. In the last two stages of the innovation process, the experts forecasts are more reliable.
At each stage of the innovation process addressed the specific tasks involved and the experts, knowledge, opinions, practical experience and skills which are useful for determining the direction of further actions, management decisions, reducing risks in business, etc. Experts provide projected costs of innovation, and assess the predicted performance indicators of innovative activity (Fig. 1).
At each stage of the innovation process, in addition to the basic research phase, experts provide forecasts on the volume of sold innovation products. For planning of the enterprise activity is extremely important to have such predictions for a certain period. Thus, at the stage of applied research to determine accurate amount of future sales of innovative products that will hit the market in a few years, it is extremely difficult. Such forecasts have a low accuracy, therefore, the formation of strategic development plans of the enterprise for them is high risk.
Predictions of innovation expenditures
Fundamental research
Applied research
1. The cost projections for further research and development studies.
2. The cost projections for the development and implementation of innovative technologies.
Forecasts of performance indicators innovative activity
1. Forecasts of the number of implemented new technological processes.
2. Projections of the number of developed new products.
/ /
"Research &
Development"
(R&D) /
/ /
Invention
Industrial design
Utility model /
1. The cost projections for the development of innovative technologies.
2. The cost projections for implementation of innovative technologies.
1. Forecasts of the number of implemented new technological processes.
2. Projections of the number developed new types of production and volume of sale.
/ / The cost projections for implementation of innovative technologies.
Innovative product / ->
Projections cost on sales of innovative products. The costs forecasted for the promotion of innovative products.
/ /
Introduction
/
s
Mass production The cost projections for future capture of market innovation of the enterprise.
/ S
1. Forecasts of the volume sold innovation products.
2. Forecasts of technology transfer.
Forecasts of the volume sold innovation products.
Fig. 1. Forecasts of expenditure on innovation and indicators of innovative performance at each stage of the
innovation process
Source: Own elaboration
EKOHOMIKA: peoniï nacy
№1(23), 2016
ECONOMICS: time realities
At the stage of experimental works the situation is more certain. The company is exploring the possibility of introducing innovations in the less distant future. At the implementation stage predictions of volume of sales of innovative products are the most reliable. The company has all the necessary information about the market, about consumers, about the methods of marketing and promotion methods.
At the stage diffusion of innovation (mass production) forecasts are also credible since there is existing information about the sale of the goods, its dynamics and market trends.
Fig. 2. Source and methods of obtaining the predictions at each stage of the innovation process
Source: Own elaboration
Table 1. Package methods of forecasting sales innovative products according to the known information
To the implementation stage predictions based on expert assessments. Only experts can foresee the future sales of innovative products, based on theoretical knowledge and experience in a specific field of science or production. After the implementation of the marketing department you receive current information that may be used to obtain more reliable predictions (Fig. 2).
It is proposed to choose the tools produce forecasts according to the known information (table 1).
Known information Methods Innovation process stage Approaches
Expert assessments of future sales for several periods Forecasting models for trend dynamics Applied research R&D Introduction 1. Forecasting by linear trend with the expectation of uniform growth of sales. 2. Forecasting on a parabolic or exponential trend while waiting for the "avalanche" of sales growth. 3. Forecasting exponential and a logarithmic trend in the expectation of slow sales growth.
Current sales data with the seasonally adjusted Prediction by the decomposition method of time series Diffusion (mass production) Forecasting by trend models without seasonality and seasonally adjusted
Expert assessment of future sales and other factors that affect them Forecasting with regression models of the relationship Introduction Diffusion Forecast sales of innovative products based on other factors (cost of marketing, cost of sales, product prices, etc.)
Expert assessment of market capacity, communication efficiency, predictions of the perception market the innovations simulation model of innovation distribution Introduction Diffusion Forecasting sales volumes for the saturation of the market models: — diffuse model Bass the proliferation of goods and technologies; — Gompertz model; — Pearl-R. model.
The data on projected sales and projected changes of interrelated indicators Forecasting methods for saving lagged correlation Introduction Diffusion Forecasting sales of innovative products in connection with the forecasts of other market factors
Source: Own elaboration
Teopia i nparrma eKOHOMÍKH Ta ynpaBmHHS npoMHC^oBHMH mgnpHeMCTBaMH
Obtaining forecasts using trend dynamics models of a time series using the decomposition method of a time series and using pairwise and multiple regression models is common in the scientific literature [8, 9]. Consider a simulation model of diffusion innovation.
To apply a simulation model of diffusion innovation is at the stage of implementation or diffusion innovation products company. At these stages innovation process, as a rule, more accurately known data on the capacity of the market innovative products and the effectiveness of communication channels.
Simulation models of diffusion innovations in society are based on the classification of future customers innovative products on consumer behaviors. Thus, diffuse model of F. Bass take into account two communication channel for the diffusion innovations in society - advertising and word-of-mouth [10]. All future consumers of innovative products Bass were divided into innovators and imitators. Innovators buy innovative products under the influence of advertising, simulators make purchases under the influence review innovators.
The Bass model consists of differential equations
f(t) 1 - F(t)
=p+qF(t),
(1)
where F(t) - distribution function; f(t) - density distribution; p - innovation coefficient; q - imitation coefficient.
The volume of sales innovative products in time, S(t), is a function that depends on the density distribution f(t)
S(t) = mf(t), where m - market capacity.
Substituting the solution of the differential equation (1) into the formula S(t), we obtain
S(t)= m
(p+q)2
-(p+q)t
(2)
p (i+V^q»)2 p
The time of peak sales on the model of Bass is determined by the formula
t* = lnq - lnp
p+q
(3)
To apply the Bass model you need to determine the size of the market - m, innovation coefficient - p, and imitation coefficient - q. The innovation coefficient is considered to be the effect of external influence or advertising effect. Usually its value ranges [0-0,03]. The imitation coefficient is considered to be the effect of "word of mouth". Its value varies in the range [0-0,4].
Using the Bass model, the business can predict future sales by simulating the input data. To determine the effect of advertising and the effect of personal contact to build the Bass model by using special marketing research. For example, the effect of advertising is determined by pre-tests, and the effect of personal communication, or the coefficient of imitation, with the help of focus groups.
Gompertz and Pearl-R. model is also ^-shaped logistic curves. For curve Gompertz get a trend in the development industries and a series of new products. Gompertz model analytically expressed by the formula [11]:
y = kabt, (4)
where a,b - positive parameters with b< 1;
k - function asymptote.
In that case, when the projected "avalanche" growth of sales at the stage of market growth, the forecasts used Pearl R. curve. Logistic curve, or Pearl-R. curve - increasing function, often is given by the following formula:
k
y =-Ibt, (5)
1+ae bt
where a, b — positive parameters;
k - function asymptote.
In both models k - potential market capacity. The model parameters determine the rate of market growth. In these models, not divided separately the effect of advertising and the effect of "word of mouth". To determine the parameters of the models, usually based on the forecast of sales in the "zero" point and after a certain period of time.
Forecasting sales of innovative products in relation to other market factors enables to obtain a forecast that is balanced with the determining factor of the sale. For example, the forecast sales of components for computers interconnected with the forecast growth in demand for computers. To obtain balanced forecasts it is recommended to apply conservation methods lagged correlation [12].
For a measure that indicates the deviation coefficient of correlation lags computed from the actual points, from the same ratio, but calculated with the accession of the forecast points, was adopted by the following value:
K= Rx1x2(nn- v2(nn (6)
where
V2(nn
- lagged correlation coefficient
calculated on the actual levels of the economic indicator series.
rx x (nn - lagged correlation coefficient
calculated on the actual levels of the economic indicator series with the accession of the forecast points.
The value K is called deviation criterion lagged correlation. The best pair of trends for forecasting economic indicators invited to consider one whose deviation criterion lagged correlation minimum.
Conclusions
In marketing information system of the industrial enterprise necessarily apply methods of forecasting sales of innovative products, at each stage of the innovation process have certain features. To receive forecasts of sales of innovative products it is necessary to apply the methods of forecasting by expert estimations, methods of forecasting by statistical methods, and also methods make predictions by methods of simulation modeling.
e
ЕКОНОМ1КА: реалП часу
№1(23), 2016
ECONOMICS: time realities
Abstract
On the industrial enterprise the process of collecting information about factors of marketing environment, its analysis, determination of the need for marketing studies and their implementation, as well as forecasting market reaction to the marketing activities of the enterprise is provided by marketing information system. Using forecasting techniques in marketing information system management decisions should be taken at all stages of the innovation cycle. Purpose of the article is to provide scientific and methodological tools of determination of forecasting methods in marketing information system for industrial enterprises, depending on the stage of the innovation process.
When choosing the method of forecasting sales of innovative products it is offered to perform depending on the stage of the innovation process. At each stage of the innovation process (basic research, applied research, experimental studies, implementation and diffusion) forecast types of volumes of expenses for innovation and performance indicators of innovation are marked. For each stage of the innovation process by forecasting sources (experts or current sales) and related methods of forecasting. At the stage of diffusion forecasting sales of innovative products is recommended predict by simulation methods. Some forecasts of sales of innovative products are provided by model Gompertzian, Pearl-Reed model or a model of innovation diffusion Bass. Projections interrelated with other factors are obtained by the method of lagged correlation.
Forecasting sales of innovative products demands the use of methods of expert estimates, forecasting methods for the statistical methods and methods for example by means of simulation.
JEL Classification: С53, D48. Список л^ератури:
1. Власова М.Л. Социологические методы в маркетинговых исследованиях [Текст]: учеб. пособие для вузов / М.В. Власова: Гос. ун-т - Высшая школа экономики. - М.: Изд. Дом ГУ ВШЭ, 2006. -710с.
2. Голубков Е.П. Маркетинговые исследования: теория, практика и методология / Е.П. Голубков. -М.: Финпресс, 2005. - 464 с. - С. 263-269.
3. 1лляшенко С.М. Сучасш тенденцп застосування 1нтернет-технологш у маркетингу / С.М. 1лляшенко // Маркетинг i менеджмент шновацш. - 2011. - № 4, Т. II. - С. 64-74.
4. Мазманова Б. Методические вопросы прогнозирования сбыта / Б. Мазманова // Маркетинг в России и за рубежом. - 2000. - № 1. - С. 25-28.
5. Бернс Э.С. Основы маркетинговых исследований с использованием Microsoft Excel / Э. С. Бернс, Р. Ф. Буш: Пер. с англ. - М.: Издательский дом «Вильямс», 2006. - 704 с.
6. Моосмюллер Г. Маркетинговые исследования с SPSS / Г. Моосмюллер, Н. Н. Ребик - М.: ИНФРА-М, 2007. - 158 с.
7. Хили Дж. Статистика. Социологические и маркетинговые исследования. 6-е изд. / Дж. Хили; Пер. с англ. под общ. ред. А.А. Руденко. - К.: ООО «ДиаСофтЮП»; СПб.: Питер, 2005. - 638 с.
8. Янковой А.Г. Многомерный анализ в системе STATISTICA / А.Г. Янковой - Одесса: Оптимум, 2001. - Вып. 1. - 216 с.
9. Яшина О.1. Прогнозування рядiв динамiки в маркетингових дослщженнях / О.1. Яшина // Вюник соцiально-економiчних дослщжень ОДЕУ. - 2006. - № 23. - С. 366-371.
10. Басс Ф. (Bass F.M., 1969) A new product growth for model consumer durables // Management Science. - 1969. - V. 15. - P. 215-227.
11. Гомперц Б. (Gompertz B., 1825) On the nature of the function expressive of the law of human mortality, and on a new mode of determining the value of life contingencies Philosophical Transactions of the Royal Society of London 1825, Vol. 115, P. 513-583.
12. Янковий О.Г. Прогнозування взаемопов'язаних показниив соцiально-економiчного розвитку Украши / О.Г. Янковий, О.1. Яшина // Статистика Украши. - 2006. - № 3. - С. 61-66.
13. Yashkina O. Modelling of automobile building enterprises' innovations diffusion: [Електронний ресурс] / O. Yashkina // Економжа: реали часу. Науковий журнал. - 2014. - № 1 (11). - С. 147154. - Режим доступу до журн: http://economics.opu.ua/files/archive/2014/No1/147-154.pdf.
References:
1. Vlasova, M.L. (2006). Sotsiologicheskiye metody v marketingovykh issledovaniyakh [Sociological Methods in Marketing Research]. Vyisshaya shkola ekonomiki (High School of Economics). M.: Izd. Dom GU VSHE.
2. Golubkov, E.P. (2005). Marketingovyye issledovaniya: teoriya, praktika i metodologiya [Marketing research: theory, practice and methodology]. M.: Finpress. pp. 263-269.
Теорiя i практика економжи та yпрaвлiння промисловими пiдприeмствaми
3. Illyashenko, S.M. (2011). Suchasni tendentsiyi zastosuvannya Internet-tekhnolohiy u marketynhu [Modern trends in the use of Internet technologies in marketing]. Marketing I menedzhment InnovatsIy (Marketing and Innovation Management), 4 (II), pp. 64-74.
4. Mazmanova, B. (2000). Metodicheskiye voprosy prognozirovaniya sbyta [Methodological issues of sales forecasting]. Marketing v Rossii i za rubezhom (Marketing in Russia and abroad), 1. pp. 25-28.
5. Berns, E.S. and Bush, R.F. (2006). Osnovy marketingovykh issledovaniy s ispolzovaniyem Microsoft Excel [Fundamentals of marketing research using Microsoft Excel]. M.: Izdatelskiy dom "Vilyams" (Moscow: Publishing House "Williams").
6. Moosmyuller, G. and Rebik, N.N. (2007). Marketingovyye issledovaniya s SPSS [Marketing research SPSS]. Moscow: INFRA-M.
7. Hili, Dzh. (2005). Statistika. Sotsiologicheskiye i marketingovyye issledovaniya [Statistics. Sociological and marketing research]. (6th ed.). K.: OOO "DiaSoftYuP"; SPb.: Piter.
8. Yankovoy, A.G. (2001). Mnogomernyy analiz v sisteme STATISTICA [Multivariate analysis system STATISTICA]. Issue 1. Odessa: Optimum.
9. Yashkina, O.I. (2006). Prohnozuvannya ryadiv dynamiky v marketynhovykh doslidzhennyakh [Forecasting time series in marketing research]. VIsnik sotsIalno-ekonomIchnih doslidzhen ODEU (Journal of Social and Economic Research ONEU), 23, pp. 366-371.
10. Bass, F.M. (1969). A new product growth for model consumer durables. Management Science. V. 15, pp. 215-227.
11. Gompertz, B. (1825). On the nature of the function expressive of the law of human mortality, and on a new mode of determining the value of life contingencies Philosophical Transactions of the Royal Society of London, Vol. 115, pp. 513-583.
12. Yankovoy, A.G. and YashkIna, O.I. (2006). Prohnozuvannya vzayemopovyazanykh pokaznykiv sotsial'no-ekonomichnoho rozvytku Ukrayiny [Prediction of interrelated indicators of socio-economic development of Ukraine]. Statistika UkraYini (Statistics Ukraine), 3, pp. 61-66.
13. Yashkina, O. (2014) Modelling of automobile building enterprises' innovations diffusion. Ekonomics: time realities. scientific journal (Economy: time realities. Scientific journal), 1 (11), pp. 147-154. Retrieved from http://economics.opu.ua/files/archive/2014/No1/147-154.pdf.
Надано до редакцш^' колегп 26.12.2015
Янковий Олександр Григорович / Oleksandr G. Yankovyi
yankovoy_a@ukr. net
Яшина Оксана 1вашвна / Oksana Yashkina
nomer27@ukr. net
Посилання на статтю / Reference a Journal Article:
Methods of sales forecasting in a marketing information system of the industrial enterprises [Електронний ресурс] / O. Yankovyi, O. Yashkina //Економта: реалй часу. Науковий журнал. - 2016. - № 1 (23). - С. 43-48. - Режим доступу до журн.: http://economics.opu.ua /files/archive/2016/nl.html
