دانلود رایگان مقاله سیستم پیش بینی مبتنی بر مدل خاکستری برای جریان بازگشتی وسایل نقلیه پایان عمر

abstract

Due to legislation and economic reasons, firms in most industries are forced to be responsible and manage their products at the end of their lives. Management of product returns is critical for the stability and profitability of a reverse supply chain. Forecasting the return amounts and timing is beneficial. The purpose of this paper is to develop a forecasting system for discarded end-of-life vehicles and to predict the number of end-of-life vehicles that will be generated in the future. To create the forecasting system, grey system theory, which uses a small amount of the most recent data, is employed. The accuracy of the grey model is improved with parameter optimization, Fourier series and Markov chain correction. The proposed models are applied to the case of Turkey and data sets of twelve regions in Turkey are considered. The obtained results show that the proposed forecasting system can successfully govern the phenomena of the data sets, and high accuracy can be provided for each region in Turkey. The proposed forecasting system can be used as a strategic tool in similar forecasting problems, and supportive guidance can be achieved.

5. Conclusions

Forecasting the return flow of end-of-life products is a complicated problem that includes a number of known and unknown parameters that affect the return of products. Although several researchers have investigated product return forecasting, the return flow of end-of-life vehicles has rarely been studied. In this paper, a forecasting system based on grey system theory is designed and return flow of end-of-life vehicles in Turkey is predicted. The proposed forecasting system is composed of sub-models based on grey modelling and improvements by parameter optimization, Fourier series and Markov chain correction. In the experimental results, the original data of discarded vehicles of the different regions in Turkey are obtained from TURKSTAT. The results of the forecasting submodels are compared for each region, and the performance of the models differed for each region due to the characteristics of the related data sets. The proposed forecasting system achieved high accuracy for all regions in Turkey. The results of the future predictions provide guidance to the managers and practitioners of recovery and recycling systems.