A scale optimization method for one-way vehicle sharing system based on queuing theory

Abstract

The invention provides a method for optimizing the scale of a one-way vehicle sharing system based on queuing theory, and relates to the technical field of planning and design of a one-way vehicle sharing system. The present invention uses the mixed BCMP queuing network model to describe the vehicle circulation in the one-way sharing system, clearly introduces the reservation strategy, road congestion and limited parking spaces and other practical factors, and clarifies the system scale and vehicle availability V i and the parking occupancy rate U i The asymptotic relation of , and incorporate it into the joint optimization framework. Then, according to the principle of profit maximization, the joint optimization problem of system scale is proposed to determine the optimal fleet size and parking capacity that meet the given performance index. The present invention explicitly introduces road congestion by mapping the remaining capacity of the physical road to the number of servers in the queuing network, and places constraints on parking lot capacity by modeling the parking lot model as an interdependent open network.

Description

technical field [0001] The invention relates to the technical field of planning and design of a one-way vehicle sharing system, in particular to a method for optimizing the scale of a one-way vehicle sharing system based on queuing theory. Background technique [0002] The vehicle sharing industry has developed rapidly and has received more and more attention in recent years. Common research methods include algorithms based on linear programming and network flow, algorithms based on space-time models, and queuing theory. The modeling and analysis of the system using queuing theory methods is a key research line. The queuing network allows the calculation of key performance indicators, such as the availability of vehicles at the station and the waiting time of customers. These performance indicators are very important for the measurement of system service quality. [0003] Based on queuing theory, Kochel et al. established a general vehicle sharing system model based on queui...

AI Technical Summary

Problems solved by technology

Although Hu et al. paid attention to the above problems and introduced road congestion and parking capacity into the model, this work pre-sets the road capacity and sets up a buffer zone at the parking station to fix passengers without parking spaces in order to prevent customer loss in the closed network. Therefore, the queuing network constructed by it cannot accurately describe the real system. In addition, the relationship between the system scale and the parking occupancy rate is not clearly given.

Images