Research On Sales Forecast Of Great Wall Automobile Based On BP Neural Network

With the entry into the World Trade Organization in 2001,China's auto industry has undergone drastic changes,and automobile sales have seen a spurt of growth,rising from 3million in 2002 to 24 million in 2015.China was once a global automobile manufacturer Sales volume champion.With the strong support of relevant national policies and the continuous implementation of the "Thirteenth Five-Year Plan" and the "Made in China 2025" plan,the development of independent brand automobiles has become an important part of China's strategy to achieve a strong manufacturing country.Great Wall Automobile,as a well-known domestic independent brand automobile in China,has achieved a series of achievements in recent years.The company's Haval brand has sold 938,000 units in 2016,and Haval H6 has maintained the sales champion in the domestic SUV field for many years.Automobiles have gradually become a banner of domestic independent brand automobiles.However,with China 's economy entering a new normal and the rise of domestic brands Geely and Changan,the sales of Great Wall Automobile began to decline from 2017.In 2017,the sales of Great Wall Automobile decreased by nearly 4,500 compared to 2016.2018 Compared with 2017,the annual sales volume has decreased by nearly 20,000 units.Under the situation of fierce competition in the domestic automobile industry,research on the sales volume of Great Wall Automobile can provide a basis for the company to formulate a reasonable production capacity plan and marketing strategy.While changing the current downward trend in sales volume,reducing inventory costs caused by excess capacity And ultimately increase the company's earnings.Through combing the literature,this paper selects 9 factors that may affect car sales,and explores the correlation and relevance of each influencing factor with Great Wall Motor's lagging by one period.Then,using the combined data of Great Wall Motor's sales volume lagging by one period and current variable indicators,linear regression prediction models such as ordinary linear regression,Lasso regression,and ridge regression were constructed,respectively,and BP neural network with many applications in the field of sales volume prediction.Forecast models,and compare and analyze the prediction results of Great Wall Motor's sales volume with different models.Finally,based on the established model,the data of various indicators in 2018 are used to predict the approximate sales volume of Great Wall Motor in the coming year.The results show that: research and development costs,sales costs,management costs,GDP,per capita disposable income,county and township road mileage,urbanization rate,total oil consumption and urban population are highly correlated with the sales volume of Great Wall Motor,which lags behind the first phase.Among the four prediction models of the sales volume of Great Wall Automobile,the BP neural network model has the smallest prediction error and the most accurate prediction result.Using the BP neural network model,it can be predicted that the sales volume of Great Wall Automobile in the coming year will be approximately 1.06994 million.Based on this,it is recommended that Great Wall Automobile should appropriately increase the investment in research and development costs,management costs and advertising costs,improve product quality,service level and visibility,and make reasonable arrangements for production and sales plans with reference to the forecast results of sales in the next year,and formulate a marketing strategy that matches the current Chinese automotive market environment.