Determination of Logistic Risks in Seasonal Forecasts of Sales of mechanical engineering enterprises

Apply

Determination of Logistic Risks in Seasonal Forecasts of Sales of mechanical engineering enterprises

Journal Title: Маркетинг і цифрові технології - Year 2019, Vol 3, Issue 2

Abstract

The aim of the article. The data on which the researcher provides forecasts can have a different nature. It can have a steady or hopping dynamics. The data can be characterized by the current tendency to rise or fall, and also to have a chaotic appearance. It can be in the form of dynamic or static data, seasonal data or data without existing seasonality. All these factors influence the quality of the models obtained and, accordingly, the accuracy of forecasts. The aim of the study was to obtain forecasts of seasonal demand for all sub-industries of the machine-building, to evaluate the received forecasts for the proposed scientific-methodical approach and to determine the logistic risk of trust for each of the forecasts. The results of the analysis. Forecasts of seasonal demand for production of sub-industries of the machine-building and Groups of companies were obtained by the time series decomposition method. It is proposed to estimate the model of the decomposition of the time series regarding the possibility of obtaining a forecast in accordance with the reliability of forecasts of its two components: the forecast for the trend (excluding seasonality) and the forecast that into account seasonality. The trend model and its forecast are based on data adjusted for the season, that is, with the exception of seasonal fluctuations. These models are rated for accuracy, reliability and adequacy. Depending on the model's estimates, a conclusion is drawn about the model itself and the reliability of the forecasts for it. A model with a seasonal component was obtained by multiplying the trend values by the corresponding seasonal index. In order to evaluate the accuracy of the model, taking into account the seasonality, it is suggested to use the ratio of the standard error estimate to the data range. Proposed gradation for estimating forecasts is based on seasonality: high; medium and low. Regarding trend forecasts, i.e. without taking into account the seasonal component, the following grading of estimates is proposed: high reliability, when the model is accurate, reliable and adequate, and the determination coefficient R2 > = 0,70; average reliability 0,50 < = R2 < 0,70; low reliability 0,30 < = R2 < 0,50; unreliable R2 < 0,30. In the scientific use of logistics it is proposed to introduce the concept of "Logistic risk of forecasting demand". It substantially expresses the risk of losses from trusting forecasts of demand for products of the enterprise. A qualitative assessment is offered (low, medium, high, very high), depending on the reliability of forecasts for trends and seasonality, and quantitatively, depending on the magnitude of the error in the forecasts. It is proposed to evaluate logistic risks of forecasting demand according to the forecast gradations. From the results, the greatest logistic risk of forecasting the sale of products is inherent in the sub-industry "Production of computers, electronic and optical products" and in the Groups of its companies. The smallest logistic risk of forecasting the sale of products is inherent in the sub-industry "Production of electrical equipment" and Groups of its companies. In other sub-industries and Groups of companies, the logistic risks of forecasting the sale of products has similar characteristics. Conclusions and directions for further research. Production planning should be based on demand forecasts for products. The obtained forecasts of demand for products according to trends with the consideration of seasonality are proposed to be evaluated according to the level of logistic risk, in accordance with the levels of reliability of forecasts for the trend and forecasts, taking into account seasonality. The logistic risks of forecasting demand are divided into three levels: low, medium and high. By comparing the trends inherent in the industry, sub-industry and enterprise, it is possible to reduce the logistic risks of forecasting demand and avoid other logistic risks associated with the formation of inventories and components for manufacturing products.

Authors and Affiliations

Dmytro Yashkin

Keywords

seasonality forecasting logistic risks machine building demand for products

EP ID EP593955
DOI 10.15276/mdt.3.2.2019.7
Views 155
Downloads 0