The Conservative Value Method MATLAB code Distribution_Planning_Lot_sizing_Decomposition.m is used for the Lagrangian Relaxation Method and decomposition algorithms applied to high-speed railway (CSHR) catering services. These programs are coded based on the following works: the time-varying demand and pedestrian congestion-based high-speed railway catering distribution planning problem and the batch-based model and decomposition algorithm developed by the Beijing Jiaotong University Research Team under the guidance of Professor Nie Lei. All these codes were written by Dr. Wu Xin. For any inquiries, please contact him. Your feedback is important to us, and the code will continue to be updated and improved in the future. The code includes three main parts:

  1. Main Program: main.m is the key component that initiates all related algorithms. The CPLEX solver used in the file can decompose the mixed-integer programming model into submodels. Therefore, the program will work only if the CPLEX interface is correctly installed in the MATLAB environment.

  2. Convex Program: A program that solves a series of single-variable convex maximization submodels. The submodels can be solved using the fmincon function in MATLAB's optimization toolbox.

  3. HCEA Functions: All files prefixed with HCEA_ embed the Convex Group Method (Frank Wolfe algorithm) as part of the Hybrid Cross-Entropy Algorithm (HCEA). The implementation of HCEA can be used to compare with the proposed decomposition method. Default settings are provided for various configurations.