感知乘客心理的出租车动态合乘优化方法

为研究考虑乘客感知的动态合乘问题，本文提出一种改进的算法框架。基于可行出行对概念，构建乘客满意度最大、出行时间最少的多目标线性规划问题，将合乘问题转化为车辆和乘客间的线性分配问题，并采用基于精英策略的人工蜂群算法(Elitism based Multi-Objective Artificial Bee Colony，EMOABC)求解。根据海口市出租车订单数据建立算例，实验结果表明，该算法框架能够实时提供优质动态合乘方案。相比单纯优化出行效率，考虑乘客心理的合乘策略，相对提高12%的乘客满意度，服务率等方面也有较好表现。

Dynamic Shared Taxi Optimization Method Considering Passengers Perceptions

This paper proposes an improved algorithm framework to study the dynamic ride- sharing service optimization problem considering passengers' perceptions of service quality. The problem is modeled as a linear assignment problem between vehicles and passengers based on the concept of feasible trip pairs, which is formulated as a multi-objective linear programming model, with the objectives of maximizing passengers' satisfaction and minimizing their total travel time. An elitism- based multi- objective artificial bee colony (EMOABC) algorithm is developed to solve the model. A case study on the taxi order service in Haikou, China is conducted. The computation results indicate that the proposed framework could provide a high-quality scheme in real time. Compared with only optimizing trip efficiency, the ride-sharing strategy with perceiving passenger psychology can improve passenger satisfaction by 12%. The service rate, as well as other indicators, is also at a high level.