现有的出租车合理规模是根据出租车分担率计算得出的，但分担率在调查过程中受抽样比 例、调查方式等因素的影响而存在较大误差，而且现有的出租车规模计算模型均未考虑非正规出 租车分担了部分客运需求的事实，导致规模计算结果普遍偏低。为减少计算偏差，采用全数调查 的方法，并结合出租车计价器的原始数据，直接测定出租车的日均载客人次、运营次数及单次载 客人数等指标，提出在出租车有效行驶里程计算模型中不再考虑出租车分担率指标；基于非正规 出租车客观存在而引入了非正规出租车承运比例指标，建立基于供需平衡的现时出租车合理保有 量计算修正模型。最后通过实证分析验证了修正后的模型是合理和实用的，有利于抑制非正规出 租车非法营运和城市出租车行业的健康发展。

The existing reasonable taxi scale was calculated according to the taxi share ratio. However, the taxi share ratio typically had large deviation due to the influence of sampling rate and survey method in the survey process. In addition, the fact that the unregistered taxi took some shares of market was not considered in the existing model for calculating the taxi scale. As a result, the calculated taxi scale was lower than that it should be. To reduce the deviation, combined with the original data of taxi fare meter, a full-data survey method was adopted, to investigate the daily passenger capacity, operating frequency, and average passengers on board. An effective taxi mileage calculation model which did not need to con⁃ sider the taxi share rate was proposed. Due to the existence of unregistered taxi, the share of unregistered taxi was used to establish the revised reasonable taxi scale calculation model based on the balance of supply and demand. At last, through a case study, it is demonstrated that the revised model is reasonable and applicable. It can control the illegal operating of unregistered taxi and help the development of ur⁃ ban taxi industry.