Stochastic Static Wagon-Flow Allocation Model in a Marshalling Station

The arrival wagon-flow in nature in the marshalling station has certain randomness, and can be expressed as a stochastic uncertain system. This paper presents the dependent-chance goal programming model to study the static wagon-flow allocation in the marshalling station. The objective function is the difference between the event probabilities of each departure train with maximum axle-load and exact time, the decision parameter is the wagon flow of the departure trains allocated among all the arrival flows, and the restraint conditions include the arrival flow and the objective restraint. The stochastic simulation method is adopted to deal with the chance function; and the genetic algorithm with strong robustness and global search is adopted to solve the model. Finally, the results of the case study indicate that when the arrival wagon-flow complies with special distribution, the ability to solve the allocation problem of the determined optimal model and algorithm is limited, while stochastic optimal model and the genetic algorithm based on stochastic simulation show better optimal effects.