Home

Welcome to the two-sided-matching-method wiki! Abstract: Aiming at various kinds of passenger no-show behaviors that are common in online ride-hailing services, a personalized driver and passenger bilateral matching decision analysis method considering passenger no-show behaviors is proposed. Firstly, a passenger choice model is established to estimate the probability of passenger no-show, aiming at the frequent cancellation behavior of passengers in the process of calling a car reservation on the ride-hailing platform. Secondly, an attribute association matrix based on language scale is established to calculate the weight of each attribute of drivers in the matching process. Further, considering the psychological expectation and perception of passengers, the comprehensive prospect value of passengers' satisfaction with drivers is calculated according to the prospect theory. On this basis, the multi-objective optimization model of driver and passenger matching service is constructed, and the result of driver and passenger matching is obtained by solving the model. Finally, the effectiveness and feasibility of the proposed method are further verified by a matching example of platform passenger booking.

摘要:针对网约车服务中普遍存在的各种乘客爽约行为，提出了一种考虑乘客爽约行为的个性化司机与乘客双边匹配决策分析方法。首先，针对乘客在网约车平台呼叫预约车辆的过程中频繁存在的取消订单行为，建立了乘客选择模型以测算乘客爽约概率:其次，建立了基于语言标度的属性关联矩阵，计算司机在网约车匹配过程中的各个属性权重:进一步的。考虑乘客的心理期望与感知，依据前景理论计算出乘客对司机满意度的综合前景值:在此基础上，构建了司机和乘客匹配服务的多目标优化模型,并通过模型求解获得了司机和乘客匹配的结果;最后，通过平台乘客预约网约车的匹配实例，进一步验证了提出方法的有效性和可行性。