Research On The Optimization Of Urban Logistics Vehicle Routing Problem Considering The Behavior Of Drivers

With the continuous growth of e-commerce and deepening of urbanization,it is obvious that urban logistics is extremely important for the development of economy.Urban logistics,as a connection between merchants,customers and drivers in urban areas,has become a research hotspot.An efficient distribution solution not only reduces logistics costs,but also helps to improve customer satisfaction.During the realistic logistics activities,numerous complex factors affect the actual delivery ability of drivers.Therefore,the optimization of distribution solutions in combination with realistic scenarios and demands is the key to the research of urban logistics distribution.However,existing studies tend to take external factors such as bad weather,traffic congestion and vehicle breakdown into consideration,but pay less attention to the impact of driver behavior on delivery ability.In order to fill this gap,this paper takes the urban logistics vehicle routing problem as the research object with the perspective of driver behavior,and measures the delivery ability of drivers with delivery time.At the same time,this paper mainly focuses on the construction of the vehicle routing optimization model and the design of the method under the premise that the two internal factors of fatigue behavior and learning behavior have an impact on the delivery ability.Firstly,this paper investigates the urban logistics vehicle routing problem with static customer demand and deterministic delivery time,in which the dynamics of the delivery ability caused by the fatigue behavior of drivers is considered,and the vehicle routing optimization model with the objective of minimizing the total delivery time is developed.The fatigue-based delivery time function is introduced in model to measure the change of the actual delivery ability,and then an improved artificial bee colony algorithm is proposed to solve the above model by combining breadth first search and depth first search.Finally,the effectiveness of the proposed model and algorithm is verified through comparative experiments of three artificial bee colony algorithms and analysis of the results before and after the effect of fatigue behavior.It is found that considering the influence of the fatigue behavior helps to improve the balance of workload and reduce the fatigue level of drivers.Secondly,this paper considers the effect of learning behavior on delivery ability in the study of urban logistics vehicle routing problem with dynamic customer demand and stochastic delivery time and proposes a delivery time function based on learning behavior.The multi-stage mixed integer programming model is established with the objective of minimizing total delivery cost.Considering the complexity and specificity of the research,a hybrid heuristic framework is designed to solve the proposed model,in which an improved adaptive large neighborhood search is developed to search promising solutions for deterministic problem,and Monte Carlo simulation is leveraged to evaluate the solution qualities and facilitate the searching process in stochastic scenario.Meanwhile,the framework of the rolling horizon procedure is used for solving the multi-stage optimization model.Numerical experiments show that the proposed method can tackle the problem with satisfactory performance.In addition,the effect of learning behavior is analyzed in-depth,which reveals that considering the influence of the learning behavior of drivers can improve the timeliness of delivery and the balance of workload and reduce the delivery cost.In general,this paper studies the urban logistics vehicle routing problem considering the behavior of drivers,which provides clear guidance for merchants in urban areas to develop practical and effective logistics solutions,thus promoting the sustainable development of urban logistics.Specifically,the main contributions of this paper are as follows:(1)This paper innovatively incorporates the fatigue behavior of drivers into the research on optimization of urban logistics vehicle routing problem,and depicts the impact of fatigue behavior on the delivery routes and drivers,which further expands the research perspective of urban logistics vehicle routing problem.(2)Different from the previous research on vehicle routing problem in a single scenario of stochastic or dynamic,this paper integrates the urban logistics scenario of dynamic customer demand and stochastic delivery time,and further incorporates the learning behavior of drivers,which develops the research on optimization of urban logistics vehicle routing problem with dynamic and stochastic characteristics.(3)Heuristic rules and strategies related to the research problem innovatively are incorporated into the framework of meta-heuristic algorithm to achieve efficient solution of urban logistics vehicle routing problem,which contributes to the innovation of existing urban logistics vehicle routing optimization methods.