改进H-GASA算法求解网约车拼车服务问题  

Improved H-GASA Algorithm for Solving the Carpooling Service Problem of Online Taxi

作　　者：王海晓[1] 冯星霖 郭敏 WANG Haixiao;FENG Xinglin;GUO Min(College of Energy and Transportation Engineering,Inner Mongolia Agricultural University,Hohhot O10018,China)

机构地区：[1]内蒙古农业大学能源与交通工程学院,呼和浩特010018

出　　处：《内蒙古农业大学学报(自然科学版)》2025年第2期53-62,共10页Journal of Inner Mongolia Agricultural University(Natural Science Edition)

基　　金：内蒙古自治区自然科学基金项目(2024LHMS05024);内蒙古高等学校科学研究项目(NJZY23112)。

摘　　要：目前网约车拼车服务存在响应不及时、乘客舒适度低等现象,亟需对拼车路线及求解算法进行优化。本文首先考虑路网条件及时间窗阈值影响,以网约车运行成本与乘客出行成本最小化为目标,构建基于双向线路的网约车拼车优化模型("one-to-many"online car-hailing carpooling model under multiple constraints)。其次以遗传算法为基础,结合模拟退火温度调控机制,改进适应度评价和接受准则,提出混合遗传-模拟退火算法(hybrid genetic-simulated annealing algorithm,H-GASA)。最后以呼和浩特东站及其周围交通网络为例进行实例验证。实验结果表明,与其他算法相比,H-GASA算法在多种时间窗下均能有效降低乘客出行时间和车辆运营成本。此外,H-GASA算法得到的网约车拼车服务问题求解方案更优,收敛曲线更平缓,效率更高,验证了H-GASA在克服遗传算法过快收敛问题上的有效性。At present,online ride-hailing carpooling services face challenges such as untimely response and low passenger comfort.Therefore,optimizing carpooling routes and solution algorithms has become urgent.This paper first considered the influence of road network conditions and time window thresholds,aiming to minimize both the operating cost of car-hailing services and the travel cost of passengers.Based on these factors,an optimized car-hailing carpooling model was constructed under multiple constraints(referred to as the"One-to-many"online car-hailing carpooling model under multiple constraints).Secondly,based on the genetic algorithm,and combined with the simulated annealing temperature regulation mechanism,the fitness evaluation and acceptance criteria were improved,leading to the development of a hybrid genetic-simulated annealing algorithm(H-GASA).Taking Hohhot East Railway Station and its surrounding transportation network as an example,this study conducted empirical validation.The experimental re-sults showed that,compared with other algorithms,the H-GASA algorithm effectively reduced passenger travel time and vehicle op-erating costs under various time windows.In addition,the solutions obtained by the H-GASA algorithm for the online car-hailing car-pooling problem exhibited superior performance,characterized by a smoother convergence curve and higher efficiency.This validat-ed the effectiveness of H-GASA in overcoming the problem of premature convergence in the genetic algorithm.

关 键 词：线路规划 网约车拼车 H-GASA算法 时间窗阈值 

分 类 号：U491[交通运输工程—交通运输规划与管理]