With the rise of e-commerce logistics networks, cargo volume prediction in sorting centers has become increasingly important and a key research topic. The aim of this study is to predict the daily cargo volume and daily cargo volume of 57 sorting centers in the future, in order to optimize the human resource allocation of logistics network sorting centers. LSTM model, ARIMA model, and integer programming model based on genetic algorithm were established according to the different problems. Regarding problem one: In the task of predicting daily and hourly cargo volume for the next 30 days, the ARIMA model is used. The ARIMA model has shown advantages in short-term forecasting and has been applied to predict the hourly cargo volume of each sorting center within 30 days. By combining these two models, the final prediction data can be obtained. Regarding problem two: Draw a directed graph to visually display the data and provide visual support for subsequent analysis. Based on the average transportation volume of sorting centers and changes in transportation routes over the past 90 days, use a mathematical model to calculate the rate of change in cargo volume after changing transportation routes. For sorting centers that have not been affected, sum the rate of change and calculate the average. Finally, use mathematical models to calculate the revised predicted daily and hourly cargo volumes. Regarding problem three: constructing an integer linear programming model based on genetic algorithm, whose decision variables involve the attendance of formal and temporary workers in different shifts of each sorting center. In the model construction, strict constraints were set to ensure the completion of cargo volume processing, prioritize the use of formal workers, and maintain a balance of actual hourly labor efficiency. Use a linear programming solver to accurately calculate the optimal attendance allocation for each shift in each sorting center. Regarding problem four: Taking SC1 sorting center as an example, and based on the prediction results of problem two, a shift attendance plan model for formal and temporary workers was constructed. On this basis, this study aims to minimize the total number of formal and temporary workers as the objective function, and considers the constraint conditions that the attendance rate of each formal worker does not exceed 85% and the continuous attendance days do not exceed 7 days. I wrote code to solve the model and successfully obtained an optimized scheduling plan. **************** ACKNOWLEDGEMENTS**************** This work is supported by ministry of education industry-university cooperative education project (Grant No.: 231106441092432), the research and practice of integrating "curriculum thought and politics" into the whole process of graduation design of Mechanical engineering major: (Grant. No.: 30120300100-23-yb-jgkt03), research on the integration mechanism of "course-training-competition-creation-production" for innovation and entrepreneurship of mechanical engineering majors in applied local universities (Grant. No.: CXKT202405), Mechanical manufacturing equipment design school-level "gold class" construction project (Grant. No.: 30120324001).
Read full abstract