Purpose of the study . Improving the speed and efficiency of decision-making on the development of sales of products by manufacturing companies based on the application of quantitative estimates of the potential of retail outlets, using automated methods of intellectual data analysis of Big Data. In order to create new mechanisms for the intellectual management of enterprises (economic agents) in the concept of the digital economy, it is necessary to simulate the processes of their interaction in the organizational market environment from the perspective of a multi-agent approach. The approach describes cybernetic mechanisms of interaction of agents with the ability to adapt to the needs of the population based on the analysis of market situations in order to develop intelligent management systems for enterprises to increase their market potential and competitiveness. Today the paradigm of open complex systems is used to simulate and analyze the mechanisms and processes of interaction of agents in the market environment. The main modern functioning of enterprises are corporate information systems, telecommunication networks, Internet technologies, mobile communication systems, Big Data, technologies of intellectual analysis, forecasting and machine learning. In addition, information for modeling, research and analysis of the market and the behavior of agents can be collected from open sources on the Internet. In the field of mass sales, the result of the introduction of innovative technologies is the support of decision-making in the process of managing the sales of goods and services in order to synthesize effective business-strategies for the production and sale of goods, aimed at increasing the company’s profits. From the point of view of the manufacturer, it is necessary to carry out strategic and tactical planning for the deployment and maintenance of the sales network through retail outlets that implement or have the potential to realize the company’s products. In the overwhelming majority of cases, experts make the collection and analysis of objective data on outlets, and the models for building quantitative estimates of the potential of outlets are used only the internal data on the sale of goods or partners. Materials and methods . The solution is based on the analog method of assessing the attractiveness of the outlet, used in marketing. Data mining techniques (various clustering methods) and the method of mathematical statistics - variance analysis are also used to solve this problem. Modern methods of processing large volumes of data and their intellectual analysis allow us to offer new methods for quantifying potential based on the analysis of the whole diversity of data stored in the open information sources. Results . A multi-stage method of quantitative assessment of the potential of retail outlets in ruble equivalent was proposed. A method has been developed for dividing retail outlets into strata based on features describing the position of the retail outlets, competitive environment, transport accessibility, and a typical consumer. A modification of the K-means clustering method is proposed. Conclusion . The paper proposes an approach to solving the task of promoting a wide range of products by the manufacturer through existing distribution networks. A method for quantitative assessment of the potential of outlets is proposed. The proposed method is based on the analog method of comparing outlets and uses the methods of clustering outlets across a wide range of indicators. The results of approbation of the approach on the data for 33 regions of the Russian Federation are given. The results of the work are planned to be used in the future to solve the problem of building a matrix of consumer preferences for outlets.
Read full abstract