Construction And Empirical Research Of Data-driven Operational Decision Model Based On C2M E-commerce Model

The digital economy has become a new driving force and opportunity for the high-quality development of China’s economy.In 2017,the "Government Work Report" of China paid attention to digital economy and proposed that “We will push forward with the internet Plus action plan and speed up the development of the digital economy.” In 2021,“The Outline of the 14 th Five-Year Plan(2021-2025)for National Economic and Social Development and the Long-Range Objectives Through the Year 2035” clearly proposed developing new advantages of digital economy,making full use of the advantages of massive data and rich application scenarios,promoting deep integration of digital technology and the real economy,empowering transformation and upgrading of traditional industries,spawning new industries,new formats and new models,and strengthening new engines of economic development.The manufacturing industry is the mainstay of the real economy.Promoting digital transformation and upgrading of the manufacturing industry model and corporation form,has become a key engine for leveraging the new advantages of the digital economy and promoting the high-quality development of the Chinese economy.With the continuous development of manufacturing digitalization,the integration of manufacturing and retail has become one of the development trends of manufacturing digitalization,and data resources have gradually become a bridge between production and sales.Under the trend of digitalization,the C2M(Customerto-Manufactory)e-commerce model has become one of the important ways for the manufacturing industry to realize digital transformation.However,in practice,the C2 M ecommerce model manufacturer’s operational decision-making has the problem of insufficient digitalization,which can be summarized in two aspects: First,the digital capability needs to be improved.Second,the development of consumer-related big data resources is insufficient.This research takes the C2 M e-commerce model as the research object and takes heterogeneous data as the research basis.Focusing on the above research question,we carry out innovative research work in the following five aspects:(1)We build a C2 M control tower model to enhance digital capabilities by enhancing the end-to-end visibility of C2 M.According to the supply chain control tower theory,combined with the big data value chain,we build a C2 M control tower model,and propose the definition,characteristics,enabling mechanism and operation mechanism of the C2 M control tower.Based on the model,we propose a theoretical solution for enhancing digital capability using C2 M control towers to enhance end-to-end visibility.(2)We build a style design generative model driven by image data.Based on the rich product modeling data in the C2 M model,using the idea of mixed method research,combining the generative adversarial networks,convolutional neural networks,self-attention mechanism and designer design experience,we build a product style design generative model driven by image data,which can automatically generate product style design schemes based on image data of hot-selling products.The model can exploit the value of image data and empower design of C2 M model manufacturers with solving the problems of strong subjective factors and high design cost in traditional designer schemes.(3)We build a sales forecasting model driven by time series data.Based on the rich product sales data in the C2 M model,using the idea of transfer learning,combining particle swarm optimization and long and long short-term memory,we build a sales forecasting model driven by time series data,which can not only effectively forecast the future sales of existing products,but also can predict the future sales of new products.The model can exploit the value of time series data and empower production of C2 M model manufacturers with solving the problems of difficulty and low accuracy in forecasting the future sales of new products.(4)We build a customer satisfaction analysis model driven by text data.Based on the rich customer review data in the C2 M model,integrating multiple text mining methods,we build a customer satisfaction analysis model driven by text data,which can effectively mine the customer satisfaction and product demand trends contained in the reviews.The model can exploit the value of text data and empower marketing of C2 M model manufacturers with solving the problem of lack of characteristic analysis in customer satisfaction analysis.(5)We propose a digital operation decision-making practice path for manufacturers in the C2 M model.Based on the C2 M control tower model,we propose a digital operation decisionmaking practice path for C2 M mode enterprises,namely building a C2 M control tower model,integrating and processing heterogeneous data related to customers,carrying out demand analysis and building data-driven operational decision model.Taking clothing industry in the C2 M model as an example,we verify the effectiveness of the three data-driven operational decision model and practical path through empirical research based on three types of heterogeneous data from real clothing e-commerce platforms.This research provides a new theoretical path for improving the digital capabilities of manufactures in the C2 M e-commerce model,provides a new method support for exploiting big data resources in the C2 M e-commerce model,and provides a good practice path for improving the digital capabilities of manufacturers’ operational decision-making in the C2 M ecommerce model.