General Regression Neural Network and Artificial-Bee-Colony Based General Regression Neural Network Approaches to the Number of End-of-Life Vehicles in China

Establishing the number of vehicles that will reach the end of their useful lives in the coming years will substantially affect recycling management and recycling policy. Thus, how to construct a reasonable, accurate model to forecast a product's end of life is important for recycling management. To improve forecast accuracy for vehicle end of life, this paper proposes two approaches: a general regression neural network (GRNN) and an optimized GRNN based on an artificial bee colony. These approaches are applied to forecast the number of end-of-life vehicles (ELVs) in China. In addition, the proposed models are used to predict the number of ELVs that will appear in China from 2016 to 2020 by combining the forecasting data for the main factors that influence the number of such vehicles. Theoretical and simulation results indicate that the described approaches are effective and feasible. This paper provides practical data support and a better theoretical model for researchers, government managers, and industrial engineers faced with the problems posed by ELVs.