Two-Dimensional Bin Packing Problem With Strongly Heterogeneous Cargo Conflicts

The logistics operation practice of cargo consolidation and mixed loading of multiple customer orders increases the complexity and variability of the classic packing problem.Implementing different processing requirements based on different attributes of the goods is the root of the problem,such as physical attributes(such as volatile odor,(Fragile,etc.),safety(battery,etc.),hygiene(medicine)and other special handling requirements,making the goods in the same loading unit mixed loading conflict conflict,that is,two pieces of goods in the cargo set are not allowed at the same time Loaded into the same loading unit for storage and transportation,cargo conflicts become one of the high-frequency hotspot decision-making problems in the current mixed flow loading and transportation situations,which leads to the problem of packing with cargo conflicts.The dissertation studies the packing problem with strong heterogeneous cargo conflict relationship,and establishes a two-dimensional packing integer programming model.This model belongs to the discrete combination optimization model,and optimizes the conflict relationship with the least use of container resources as the optimization goal.,Positional relationship,load balance,external size relationship and other important constraints.Furthermore,in order to solve the two-dimensional packing problem applicable to strong heterogeneous relations,a hybrid chaotic particle swarm optimization algorithm is designed,which breaks the limitation of the standard particle swarm algorithm in the solution process and overcomes its shortcoming of being easily trapped into local extreme values.The main improvements include: First,the speed and position of particles are updated using inertial weights that are adaptive to the number of iterations to slow down the search speed in later iterations and locate the global optimal solution more accurately.Second,in the particle search phase,a genetic algorithm is introduced to generate a neighborhood solution,and the elite twopoint crossover and mutation operations are used to update the particles in the group to improve the search ability of the neighborhood solution.Thirdly,using the chaotic nature of ergodicity and pseudo-randomness in chaos theory,based on the chaotic search optimization idea of Logistic map,a dynamic optimization strategy is designed to improve the search accuracy of particles.Finally,in order to verify the effectiveness of the model and its algorithm performance,based on the two angles of small,medium and large-scale cargo quantity N and the conflict sparsity ? of different data dimensions,the paper designed 27 sets of numerical experiments to analyze the packing area in depth.The impact of changes in utilization rate and conflict sparsity on the number of cases and the effect of packing;the performance improvement of the improved standard particle swarm algorithm is analyzed from the aspects of the convergence speed and convergence accuracy of the solution.Numerical experimental results show that the model and algorithm proposed in this paper are effective.The packing integer programming model with strong heterogeneous cargo conflict relationship is a useful extension of the classic packing problem;the improved hybrid chaos particle swarm algorithm is superior to the standard particle swarm algorithm in terms of packing effect and algorithm convergence performance.