“人货合乘”模式下考虑碳排放成本的网约车路径优化

Ride-hailing Vehicle Routing Optimization Considering Carbon Emissions Costunder "Cargo-passenger" Joint Transportation Model

随着社会行业数量增加、业务种类丰富,许多行业都需要通过交通运输来为客户提供特定的服务,不时会产生诸如道路拥堵的交通问题以及诸如二氧化碳气体排放的环境问题。为在合理运用道路交通资源的同时减少碳排放量,在共享经济的基础上将网约车行业与物流业相联系,使网约车协助快递公司配送货物,形成一种“人货合乘”的配送模式。基于此,考虑碳排放成本,构建以网约车利润最大、乘客与司机的等待时间总和最小的双目标优化模型,以NSGA-II算法为求解框架,采用启发式方法构造初始解,并设计了保序交叉算子和双车辆交叉算子,同时通过融合、替换、移动等变异算子对解产生扰动。最后结果验证了“人货合乘”配送模式在缓解交通问题和环境问题上的有效性。

With the rapid development of Internet technology,the ride-hailing service has become in⁃creasingly popular.However,a significant issue arises:Ride-hailing vehicles often run unloaded,leading to a waste of road resources and further exacerbating traffic congestion.Additionally,the logistics industry faces the challenge of increasing distribution pressure over the years.The growing number of vehicles contributes to higher carbon emissions,posing a significant threat to the environment.Hence,this paper explores the po⁃tential for combining ride-hailing with the logistics industry and proposes the“cargo-passenger”joint trans⁃portation mode for the purpose to alleviate the empty runs of the ride-hailing vehicles,ease the distribution pressure on the logistics companies,and lower carbon emissions by the transportation sector.Under this mode,the ride-hailing vehicle has the option to pick up several minor cargoes from a distribution center de⁃pending on whether the passengers allow the additional stops while ensuring that both the passengers'board⁃ing time window and the deadline for receiving the cargoes are met during the trip.To optimize this process,a dual-objective programming model is designed to maximize the total profit for all ride-hailing vehicles and to minimize the total waiting time for both the passengers and the ride-hail⁃ing vehicles.To achieve these objectives,the NSGA-II algorithm is adopted as the framework for solving the problem.First,a heuristic initial solution scheme is designed,and individuals from the initial population are selected as parents according to a binary tournament.Second,the order-preserving crossover operator and two-vehicle crossover operator are designed in the crossover stage;and mutation operators for cargo ex⁃change,for moving cargo before the pick-up point,for moving the cargo between the pick-up and the drop-off points,and for vehicle fusion are designed in the mutation stage.Third,the crossed and mutated populations are sorted to form new progeny.Finally,the final solution is obtained through iteration.By analyzing the optimization effect of the final solution on the decision objectives in the case of a relevant numerical example,it can be found that compared with the initial solution,the final solution can bring high⁃er vehicle profits given almost the same passenger waiting time,and can significantly avoid the no-load situa⁃tion of the vehicle,thereby greatly improving the operating efficiency of the vehicles.In addition,compared with the traditional distribution mode,the"cargo-passenger"joint transportation mode considering carbon emission cost can also reduce carbon emissions and realize the optimal allocation of road resources.To sum up,the"cargo-passenger"joint transportation mode has broad application prospects,which not only helps improve the profitability of the ride-hailing industry,but also plays an important role in alleviating urban traffic pressure and promoting the low-carbon development of the logistics industry.