机场出租车载客优化问题研究
Research on Optimization of Airport Taxi Carrying Passengers

针对机场出租车司机等待载客、放空返回的选择问题，研究了一种提高出租车司机运载乘客效率与收益的选择策略。根据模型的合理性和相关因素的依赖性，建立了基于多属性的出租车司机接客决策树模型，并通过Python的爬虫技术获得交通信息数据集，在基于大数据的程序批量预处理后，协同其他不确定属性验证模型的可行性。建立了数学模型，试验测试计算结果表明：当计算得平均到达率λ = 5.98、单个上车点的平均服务率μ = 2.21、乘客等待的时间费用w = 0.001时，模型可靠性最高，且上车点数量m等于5时乘客时间成本、出租车时间成本与效益值达到最优，实现排队服务系统成本和费用最小。
Aiming at the choice problem of airport taxi drivers waiting for passengers and returning empty, a selection strategy to improve the efficiency and income of taxi drivers to carry passengers is stud-ied. According to the rationality of the model and the dependence of related factors, a multi-attri- bute-based taxi driver pick-up decision tree model is established, and a traffic information dataset is obtained through Python crawler technology. After batch preprocessing of programs based on big data, the feasibility of other uncertain attribute verification models is coordinated. A mathematical model is established, and the experimental test calculation results show that when the calculated average arrival rate λ = 5.98, the average service rate μ = 2.21 of a single boarding point, and the passenger waiting time cost w = 0.001, the reliability of the model is the highest, and when the number of boarding points m is equal to 5, the passenger time cost, taxi time cost and benefit values are optimal, and the cost and cost of the queuing service system are minimized.