A Cold Chain Logistics Fleet Pricing Strategy and Vehicle Scheduling Optimization System and Method Based on Contract and Crowdsourcing Collaboration

Abstract

This invention discloses a pricing strategy and vehicle scheduling optimization system and method for cold chain logistics fleets based on contract and crowdsourcing collaboration. It relates to the field of cold chain logistics system optimization and capacity allocation technology. The system includes: a transportation demand forecasting module for identifying daily order quantity, transportation distance distribution, order type structure, factory departure time distribution, and transit time fluctuation parameters; a franchise vehicle capacity procurement price optimization module for outputting the optimal pricing scheme and freight rate standard for franchise transportation services; a potential franchise vehicle franchise decision-making behavior simulation module for determining the company's daily effective franchise fleet size under a specific pricing scheme; a daytime vehicle dynamic scheduling module for dynamically deciding on the collaborative scheduling of franchise vehicles and crowdsourced vehicles; and a feedback mechanism module for outputting a pricing strategy and daytime vehicle dynamic scheduling scheme that meets the optimal balance between full-cycle transportation demand and cost. This invention can significantly improve transportation service efficiency and cost control capabilities.

Description

Technical Field

[0001] This invention relates to the field of cold chain logistics system optimization and capacity allocation technology, and in particular to a cold chain logistics fleet pricing strategy and vehicle scheduling optimization system and method based on contract and crowdsourcing collaboration. Background Technology

[0002] In the cold chain logistics transportation industry, companies primarily transport fresh food goods, requiring extremely high levels of timeliness, stability, and responsiveness. Currently, most cold chain logistics companies adopt a light-asset operation model, and to rapidly expand their business scale, they commonly employ a "transportation service procurement" approach. This involves signing contracts with potential franchise vehicles, incorporating them into a franchise fleet system to enhance their own transportation capacity. This model effectively reduces fixed asset investment costs and improves operational flexibility. However, with the increasingly significant daily and seasonal fluctuations in the demand for fresh products, relying solely on franchise vehicles for transportation services can no longer meet the complex and ever-changing needs of cold chain logistics.

[0003] In actual operation, if companies rely solely on franchised vehicles to meet peak transportation demands, it can easily lead to an oversized fleet, increased idle rates, and significantly higher operating costs. Introducing crowdsourced vehicles, however, serves as a demand-responsive supplement, demonstrating greater flexibility and economy in meeting sudden and peak-period transportation needs. Therefore, cold chain logistics companies are generally beginning to build a collaborative operation mechanism of "franchised fleets + crowdsourced vehicles." This mechanism requires companies not only to rationally formulate procurement pricing strategies for franchised vehicles' transportation services, balancing transportation costs with the willingness of potential franchisees, but also to dynamically allocate franchised vehicles and responsive crowdsourced vehicles to cope with the combined impact of multiple uncertainties such as factory departure time, transportation time, and vehicle response time.

[0004] The existing technology still has significant shortcomings in terms of transportation capacity procurement pricing and vehicle scheduling: (1) There is a lack of pricing strategy optimization models for the collaborative transportation capacity structure of "franchise + crowdsourcing". The pricing of franchise transportation is rough and it is difficult to take into account both the incentive of franchise vehicles and cost control; (2) The pricing of crowdsourcing vehicle service procurement lacks refined decision-making and cannot take into account the uncertainty of its response time and the cost of delay; (3) The uncertainty of intraday transportation demand and the uncertainty of order delivery time coexist, and the existing scheduling methods are difficult to achieve dynamic and efficient collaboration between franchise vehicles and crowdsourcing vehicles; (4) There is a lack of unified optimization mechanism for the allocation of transportation capacity during the off-season and unreasonable pricing strategies can easily lead to problems such as the loss of franchise vehicles in the off-season and insufficient transportation capacity in the peak season.

[0005] Therefore, there is an urgent need for a contract-based and crowdsourcing-integrated cold chain logistics fleet pricing strategy and intraday dynamic scheduling system and method that comprehensively considers the characteristics of crowdsourcing response, the seasonality of cold chain transportation, and demand volatility. This system should be able to predict transportation demand based on actual waybill data, optimize franchise pricing strategies for peak and off-peak seasons, simulate the franchise decision-making behavior of potential franchise vehicles, and collaboratively schedule franchise and crowdsourced vehicles under multi-source uncertainty, thereby reducing the overall total transportation cost for cold chain logistics companies and improving service efficiency and responsiveness. Summary of the Invention

[0006] To address the technical problems existing in the prior art, this invention proposes a cold chain logistics fleet pricing strategy and vehicle scheduling optimization system and method based on contract and crowdsourcing collaboration. This overcomes the limitations of the prior art in terms of extensive capacity procurement, single pricing strategy, delayed scheduling response, and insufficient ability to cope with multi-source uncertainties, thereby improving the comprehensive operational capabilities of the cold chain logistics system in terms of transportation efficiency, cost control, and service stability.

[0007] On the one hand, to achieve the above objectives, this invention provides a cold chain logistics fleet pricing strategy and vehicle scheduling optimization system based on contract and crowdsourcing collaboration, comprising:

[0008] The transportation demand forecasting module is used to identify daily order quantity, transportation distance distribution, order type structure, factory departure time distribution, and transit time fluctuation parameters based on historical waybill data.

[0009] Franchise vehicle capacity procurement price optimization module: Used to build a franchise vehicle capacity procurement pricing optimization model with the goal of minimizing the total annual transportation cost, and output the optimal pricing scheme and freight rate standard for franchise transportation services;

[0010] The potential franchise vehicle franchise decision-making behavior simulation module is used to calculate the probability of potential franchise vehicles choosing to franchise under different pricing scenarios by using a Logit-based simulation model of potential franchise vehicle decision-making behavior and SP survey data showing preferences. It also determines the company's daily effective franchise fleet size under a specific pricing scheme by combining the supply scale of potential franchise vehicles in the target area.

[0011] Daytime vehicle dynamic scheduling module: It is used to combine the uncertainty of order delivery time and the response time distribution of crowdsourced vehicles, and adopt the Monte Carlo simulation method to dynamically decide on the collaborative scheduling of franchised vehicles and crowdsourced vehicles.

[0012] Feedback Mechanism Module: Used to build a feedback iteration mechanism, which optimizes pricing parameters and scheduling schemes through iterative linkage, forming a closed-loop feedback, and outputs a pricing strategy and daytime vehicle dynamic scheduling scheme that meets the transportation needs of the entire cycle and optimizes costs.

[0013] On the other hand, to achieve the above objectives, the present invention also provides a cold chain logistics fleet pricing strategy and vehicle scheduling optimization method based on contract and crowdsourcing collaboration, including:

[0014] Based on historical waybill data, predict the distribution parameters of transportation demand;

[0015] Based on the aforementioned transportation demand distribution parameters, an optimization model for the procurement and pricing of franchised vehicle capacity is constructed with the objective of minimizing the total annual transportation cost, and the optimal pricing scheme and freight rate standard for franchised transportation services are output.

[0016] By using a simulation model of potential franchise vehicle decision-making behavior based on the Logit model and SP survey data on revealed preferences, the probability of potential franchise vehicles choosing to join under different pricing scenarios is calculated. Combined with the supply scale of potential franchise vehicles in the target area, the daily effective franchise fleet size of the company under a specific pricing scheme is determined.

[0017] Combining the uncertainty of order delivery time and the distribution of crowdsourced vehicle response time, the Monte Carlo simulation method is used to dynamically decide on the collaborative scheduling of franchised vehicles and crowdsourced vehicles.

[0018] A feedback and iteration mechanism is constructed to optimize pricing parameters and scheduling schemes through iterative linkage, forming a closed-loop feedback and outputting a pricing strategy and daytime vehicle dynamic scheduling scheme that meets the transportation needs of the entire cycle and optimizes costs.

[0019] Compared with the prior art, the present invention has the following advantages and technical effects:

[0020] (1) In terms of system design concept, this invention reflects the dual focus of cold chain logistics service systems on the scientific nature of capacity allocation and cost control. Traditional transportation service procurement relies heavily on fixed pricing mechanisms and experience-based vehicle dispatching, which is difficult to adapt to fluctuations in demand and dynamic changes in vehicle franchise behavior. This invention, for the first time, integrates three key mechanisms: differentiated pricing strategy, modeling of potential franchise vehicle franchise behavior, and dynamic scheduling of order shipments. Through a systematic and modular modeling approach, it achieves coordinated optimization of the entire process of cold chain logistics from "price guidance to behavior response to vehicle dispatching," effectively improving the supply and demand matching efficiency and service stability of the transportation system.

[0021] (2) This invention delves into the structural characteristics of cold chain orders in terms of delivery time, delivery period, and customer type. For the first time, it constructs a three-level order classification system of "factory-distributor-e-commerce transit warehouse," quantifying the heterogeneous differences of different types of orders in terms of delivery time window, transportation distance distribution, and response time. Based on this, this invention uses historical data at the order level to identify customers' sensitivity to the uncertainty of factory departure time and constructs a factory departure time distribution model based on customer type. This provides prior probability support for the dynamic scheduling process, improving the accuracy of scheduling prediction and the foresight of resource allocation from the source.

[0022] (3) This invention introduces a “pricing-franchising-scheduling closed-loop optimization” mechanism based on behavioral feedback, which innovatively incorporates vehicle franchising behavior into the design of transportation service procurement strategy. By constructing a Logit franchising behavior simulation model and an intraday dynamic scheduling model, it depicts the dual impact of pricing strategy on franchise capacity supply and order execution path, and realizes dynamic correction of pricing parameters through feedback mechanism, thereby significantly enhancing the system’s adaptability to multi-source uncertainties (such as order delivery time and crowdsourced vehicle response time), and reflecting the intelligent decision-making characteristics under highly dynamic and heterogeneous demand in cold chain transportation.

[0023] (4) This invention has good scalability and practical implementation value. The pricing strategy system constructed by this invention is applicable to various transportation service scenarios (such as intercity delivery, urban cold chain short-haul transportation, etc.). The system modules have high configurability and interface compatibility, and can be flexibly combined and deployed for different transportation modes (franchise, crowdsourcing, self-operation, etc.). In addition, the key parameters on which each module depends (such as order type, customer preferences, and potential franchise vehicle cost structure, etc.) can be fitted and updated through localized operational data, which has strong potential for practical application and commercial feasibility. Attached Figure Description

[0024] The accompanying drawings, which form part of this application, are used to provide a further understanding of this application. The illustrative embodiments and descriptions of this application are used to explain this application and do not constitute an undue limitation of this application. In the drawings:

[0025] Figure 1 This is a schematic diagram of a cold chain logistics fleet pricing strategy and vehicle scheduling optimization system based on contract and crowdsourcing collaboration, according to an embodiment of the present invention.

[0026] Figure 2 This is a schematic diagram illustrating the implementation process of the cold chain logistics fleet pricing strategy and vehicle scheduling optimization system based on contract and crowdsourcing collaboration, according to an embodiment of the present invention.

[0027] Figure 3 This is a flowchart illustrating the cold chain transportation process according to an embodiment of the present invention.

[0028] Figure 4 This is a schematic diagram of the pricing strategy according to an embodiment of the present invention. Detailed Implementation

[0029] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0030] It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions, and although a logical order is shown in the flowchart, in some cases the steps shown or described may be executed in a different order than that shown here.

[0031] This embodiment proposes a cold chain logistics fleet pricing strategy and vehicle scheduling optimization system based on contract and crowdsourcing collaboration, such as... Figure 1 ,include:

[0032] The transportation demand forecasting module is used to identify daily order quantity, transportation distance distribution, order type structure, factory departure time distribution, and transit time fluctuation parameters based on historical waybill data.

[0033] Franchise vehicle capacity procurement price optimization module: Used to build a franchise vehicle capacity procurement pricing optimization model with the goal of minimizing the total annual transportation cost, and output the optimal pricing scheme and freight rate standard for franchise transportation services;

[0034] The potential franchise vehicle franchise decision-making behavior simulation module is used to calculate the probability of potential franchise vehicles choosing to franchise under different pricing scenarios by using a Logit-based simulation model of potential franchise vehicle decision-making behavior and displayed preference (SP) survey data. It also combines the supply scale of potential franchise vehicles in the target area to determine the company's daily effective franchise fleet size under a specific pricing scheme.

[0035] Daytime vehicle dynamic scheduling module: It is used to combine the uncertainty of order delivery time and the response time distribution of crowdsourced vehicles, and adopt the Monte Carlo simulation method to dynamically decide on the collaborative scheduling of franchised vehicles and crowdsourced vehicles.

[0036] Feedback Mechanism Module: Used to build a feedback iteration mechanism, which optimizes pricing parameters and scheduling schemes through iterative linkage, forming a closed-loop feedback, and outputs a pricing strategy and daytime vehicle dynamic scheduling scheme that meets the transportation needs of the entire cycle and optimizes costs.

[0037] Furthermore, the transportation demand forecasting module constructs an order classification system to characterize the differences in order timeliness demand and quantify the heterogeneity of timeliness default costs. This order classification system includes distributor orders, e-commerce transit warehouse orders, and factory orders. Distributor orders follow a normal distribution, e-commerce transit warehouse orders follow a Cauchy distribution, and factory orders follow a... T distributed.

[0038] Specifically, the transportation demand forecasting module, based on historical waybill data from cold chain logistics companies, characterizes transportation demand features through multi-source data collection and modeling. First, it collects historical order data covering different months, extracting key fields such as order quantity, transportation distance, order type, factory departure time, and transit time. Second, combining peak and off-peak season division rules and national standards, it identifies the probability distribution characteristics of various transportation demand variables under different operating cycles and determines their corresponding distribution forms and parameters. Finally, it outputs complete demand distribution parameters, including daily order quantity, transportation distance distribution, order type structure, factory departure time distribution, and transit time fluctuations, which drive the input of subsequent pricing optimization and vehicle scheduling modules, achieving high-precision modeling and forecasting support for the cold chain logistics transportation demand side.

[0039] The functions of the transportation demand forecasting module include:

[0040] (1) Transportation process overview:

[0041] Before the transportation demand forecasting module begins data collection and modeling, it is essential to first clarify the entire operational process of cold chain transportation. The cold chain transportation process mainly includes the following three key stages: the preparation time stage, the waiting time for vehicles, and the transit time stage.

[0042] (2) Transportation data collection and preprocessing:

[0043] Historical order data from cold chain logistics companies throughout their complete operational cycle is collected, and key attribute fields for each order are extracted, including order shipment date, transportation distance, order type, factory departure time, arrival time, customer type, and delivery result. Based on the national standard GB / T 36345-2018 "Specifications for Cold Chain Logistics Services for Agricultural Products," the annual operational cycle is divided into two categories: off-season (March-August) and peak season (January-February, September-December) transportation cycles to support the identification and modeling of seasonal fluctuations.

[0044] (3) Identification of daily order quantity distribution:

[0045] Based on the order shipment date, a time series of daily order volume is constructed. Distribution identification and fitting methods are used to identify the probability distribution and corresponding parameters of daily order volume under off-season and peak season.

[0046] (4) Identification of transportation distance distribution:

[0047] Based on the transportation distance of each order, distance intervals are divided, and the proportion of orders in different intervals is statistically analyzed. The probability distribution model of the overall transportation distance is then fitted, and the concentration interval of the transportation radius and the probability density function parameters are extracted.

[0048] (5) Order type and delivery time window identification:

[0049] This embodiment constructs a three-level order classification system based on the differences in how different customers respond to delivery delays. This system is used to characterize the differences in order timeliness requirements and quantify the heterogeneity of timeliness default costs. Specifically, orders are divided into the following three categories: distributor orders, fresh food e-commerce transit warehouse orders, and factory orders.

[0050] ① For e-commerce transit warehouse orders, the delivery time window is mainly concentrated between 00:00 and 04:00 every day; ② Distributor orders show two typical time characteristics depending on the delivery distance: when the order distance is less than or equal to 500 kilometers, the delivery time is mainly concentrated between 21:00 and 24:00 at night to adapt to the closing operation cycle after the store closes; while long-distance delivery orders with a distance of more than 500 kilometers mainly arrive after 05:00 the next morning to ensure the customer's inventory needs before opening; ③ The delivery time window for factory orders is mainly concentrated between 13:00 and 21:00, which is highly consistent with the factory warehousing operation rhythm.

[0051] The module records the timeliness requirements and delay tolerance of each type of order, which serves as the basis for prioritizing scheduling strategies.

[0052] (6) Order delivery time distribution identification:

[0053] This embodiment innovatively proposes the concept of identifying the distribution of production time for different types of orders, in order to more accurately depict the time distribution of intraday demand. Specifically, based on historical production time data, probability distribution models are performed for the production times of different order types to identify their distribution types and parameters, thus depicting the uncertainty process of each order's production time and providing dynamic information flow input for intraday dynamic scheduling.

[0054] Specifically, the order shipment times for distributor type, e-commerce transit warehouse type, and factory type respectively conform to a normal distribution, a Cauchy distribution, and a T-distribution, and their distribution functions are as follows: , , The expression is as follows:

[0055] ;

[0056] ;

[0057] ;

[0058] in, t Indicates the order type is Order delivery time; This indicates all distributor type orders ( The average manufacturing time of ( ); Indicates distributor type orders ( Standard deviation of manufacturing time; This represents the median production time for orders placed through e-commerce transit warehouses. It is a scale parameter; This indicates the parameter to be estimated, which determines the factory type. The shape of the order delivery time distribution curve; H This is the earliest manufacturing time.

[0059] (7) Modeling of uncertainties in transit time:

[0060] Extract transit time data from historical transportation processes and distinguish between three delivery statuses: early arrival, delayed arrival, and on-time arrival. Establish probability distribution models for transportation time errors for each status and assess the range of arrival time fluctuations and risk probabilities caused by the transportation process.

[0061] (8) Identification of the distribution of valid franchised vehicles:

[0062] This embodiment innovatively proposes the concept of identifying the distribution of effective franchised vehicles, that is, modeling the relationship between the distribution of effective franchised vehicle numbers and the actual size of the franchised fleet. It statistically analyzes and fits the daily distribution of actually available franchised vehicles under different seasonal cycles, extracting its probability distribution form and parameters to evaluate the vehicle availability distribution trend under different capacity procurement strategies. Through data fitting, this embodiment finds that the daily distribution of effective franchised vehicle numbers during both peak and off-peak seasons conforms to a Poisson distribution.

[0063] ;

[0064] In the formula, k This indicates the number of vehicles in the daily franchise model; This represents the average number of vehicles in the daily active franchise model within a transportation period. The daily distribution of the number of vehicles in the effective franchise model conforms to the Poisson distribution density function. e It is a constant.

[0065] Furthermore, the franchise vehicle capacity procurement pricing optimization model outputs the optimal pricing scheme and freight rate standard for franchise transportation services by setting a pricing strategy and combining peak and off-peak season differentiation factors.

[0066] The pricing strategy includes the transportation costs of logistics companies calling vehicles to transport orders and the vehicle disposal costs of logistics companies when they switch from peak season to off-season.

[0067] The transportation costs incurred by the logistics company in dispatching vehicles to transport orders include transportation service fees, early arrival waiting costs, and late arrival penalty costs.

[0068] Specifically, the franchise vehicle capacity procurement price optimization module takes the logistics company as the decision-making entity and constructs a franchise vehicle capacity procurement pricing optimization model with the goal of minimizing the total annual transportation cost. In this model, the logistics company sets pricing strategies, including single pricing, base pricing, and extra-mileage pricing, based on transportation demand characteristics. It also incorporates seasonal and peak-season differentiation factors to output the optimal pricing scheme and freight rate standard for franchise transportation services. The pricing results are then passed as input parameters to the potential franchise vehicle franchise decision-making behavior simulation module.

[0069] The functions of the franchise vehicle capacity procurement price optimization module include:

[0070] (1) Pricing strategy proposed:

[0071] This embodiment proposes a multi-strategy differentiated pricing mechanism for cold chain transportation service procurement scenarios, aiming to enhance logistics companies' pricing flexibility and cost control capabilities under changing transportation demand conditions. Combining existing mainstream pricing methods in the freight market with practical operational experience, this embodiment summarizes and designs four categories and six pricing models, forming a multi-dimensional and switchable pricing strategy system.

[0072] The four pricing strategies are: Strategy 1 is the composite pricing strategy (ST strategy), Strategy 2 is the multi-level segmented pricing strategy (Segment strategy), Strategy 3 is the static unified pricing strategy (Fixed strategy), and Strategy 4 is the package price + extra mileage pricing strategy (Hybrid strategy).

[0073] (2) Total cost of logistics company:

[0074] Logistics companies employ a model combining crowdsourcing and franchising, dynamically allocating vehicle resources to meet order transportation demands. During peak transportation periods, due to a significant increase in daily factory transport volume, logistics companies need to recruit a large number of franchised vehicles to cope with the surge in transportation demand. However, when transportation demand transitions from peak to off-peak seasons, demand drops sharply, forcing companies to reduce the scale of franchised vehicles. In this process, layoffs may negatively impact the company's market reputation; therefore, to maintain the company's reputation, severance costs are included in the overall cost consideration. Based on this, the total cost of a logistics company over a complete peak and off-peak transportation cycle mainly consists of the following two parts: the transportation cost of the logistics company deploying vehicles to transport orders and the severance costs incurred when the logistics company transitions from peak to off-peak seasons. The specific calculation formula is as follows:

[0075] ;

[0076] In the formula ， This refers to the sum of transportation costs for logistics companies using franchised and crowdsourced vehicles for orders under a given pricing strategy, freight rate standard, and transportation cycle, and the costs of severance pay during the transition from peak to off-peak seasons; peak season m =1 and off-season m =2, A set representing the period of transportation; O This represents the collection of vehicle types called by the logistics company; This represents the set of days in a transportation cycle for a logistics company. Indicates that the logistics company is m Season q Tianxuan Selection s Pricing strategy call o Model vehicle transportation orders i Transportation costs; Cut one job at the logistics company o =1 mode vehicle severance cost; This refers to the number of franchised vehicles that logistics companies reduce when the peak season turns into the off-season. A collection of orders; This is a set of pricing strategies.

[0077] Logistics companies m Season q Tianxuan Selection s Pricing strategy call o Total cost of the model vehicle It consists of transportation service fees, early arrival waiting costs, and late arrival costs;

[0078] ;

[0079] ;

[0080] In the formula, For the company in m Season q Tianxuan Selection s Pricing strategy call o Transportation service fees for the model vehicles; In order to be in m Season q Tianxuan Selection s Pricing strategy call o Model vehicle transportation orders i The waiting cost of arriving early; In order to be in m Season q Tianxuan Selection s Pricing strategy call o Model vehicle transportation orders i The cost of being late; Peak season m =Number of vehicles joining during period 1; Off-season m =Number of vehicles joining during period 2; Peak season m =1 and off-season m =2 The difference in the number of franchised vehicles during period 2.

[0081] (3) Cold chain logistics order delivery time calculation model:

[0082] The shipping process for an order begins from the factory departure time. The company determines the delivery plan based on the order's waiting time, the type of vehicle used, and the delivery route.

[0083] The final delivery time of an order consists of three parts: the order's manufacturing time (order creation time). Waiting time (i.e., the time from the order's production time to the completion of vehicle dispatch) Travel time (That is, the travel time of the vehicle from the origin to the destination), therefore, the total time for the order to reach the customer can be expressed as the following expression:

[0084] ;

[0085] In the formula, In order to be in m Season q Tianxuan Selection s Pricing strategy call o Model vehicle transportation orders i Final arrival time; In order to be in m Season q Tianxuan Selections Pricing strategy call o Model vehicle transportation orders i The manufacturing date; In order to be in m Season q Tianxuan Selection s Pricing strategy call o Model vehicle transportation orders i Waiting time; In order to be in m Season q Tianxuan Selection s Pricing strategy call o Model vehicle transportation orders i Travel time.

[0086] When the logistics company calls o =1 mode vehicle transportation order i When the logistics company calls, the waiting time is 0; o =2 mode vehicle transportation orders i When the waiting time is specified, it is any possible value within the time interval consisting of the order's factory departure time and the latest time allowed by the factory for the logistics company to dispatch a vehicle.

[0087] (4) Early arrival waiting and late delivery penalties:

[0088] Customers negotiate a mutually agreed-upon delivery window with the factory based on the distance to the shipping location and business hours. If an order transported by a logistics company arrives before the delivery window, the vehicle must wait at the location. To prevent spoilage, the vehicle waits at the location; therefore, the cost of an order arriving early is the cost of the vehicle waiting at the location. The specific calculation formula is as follows:

[0089] ;

[0090] ;

[0091] In the formula, The unit time waiting cost for trucks waiting to unload at the customer's unloading point after the vehicle arrives at the unloading point; For the company in m Season q Tianxuan Selection s Pricing strategy call o Model vehicle transportation orders i The time of arrival at the customer's location; For the company in m Season q Tianxuan Selection s Pricing strategy call o Model vehicle transportation orders i The lower limit of the delivery time window; It is a 0-1 variable used to indicate whether the order delivery time is earlier than the lower limit of the delivery time window (i.e., the earliest arrival time).

[0092] When goods arrive late, different types of customers react differently: When the order type is for an e-commerce transit warehouse, customers often refuse to accept the goods if there is a delay. In this case, the logistics company bears all the losses incurred by the customer due to the delayed delivery. However, when the order type is for a factory, if the order carried by the logistics company's truck is delayed, the truck usually has to wait at the factory until the same pickup time the next day before it can deliver the goods. In this case, the logistics company not only incurs additional waiting costs.

[0093] The late arrival penalty cost for orders from distributors, e-commerce transit warehouses, and factories is:

[0094] ;

[0095] ;

[0096] ;

[0097] In the formula, For orders i The cost of being late; It is a 0-1 variable used to indicate whether the delivery time is later than the upper limit of the delivery time window; For the company in m Season q Tianxuan Selection s Pricing strategy call o Model vehicle transportation orders i The final arrival time; For the company in m Season q Tianxuan Selection s Pricing strategy call o Model vehicle transportation orders i The upper limit of the customer's delivery time window; The cost of vehicles arriving late while waiting to be unloaded at the factory; The late arrival time cost for distributors' secondary customers; The time cost of late orders arriving at e-commerce transit warehouses; Indicates order i The type is a factory type; Indicates order i The type is distributor; Indicates order i The type is an e-commerce transit warehouse; This refers to the unit tonnage loss cost incurred by the logistics company after the e-commerce transit warehouse refuses to accept the goods. All three are 0-1 variables, and when one of them is 1, the other two are 0; Indicates that the logistics company is m Transportation period q Tiancheng Transport's orders i The tonnage.

[0098] Furthermore, the simulation module for potential franchise vehicle decision-making behavior takes independent potential franchise vehicles in the market as the decision-making subjects. Based on the franchise vehicle pricing strategy and its corresponding fare standard, combined with the potential franchise vehicles' expectations of transportation revenue (including transportation income, costs, order level, etc.), a simulation model of potential franchise vehicle decision-making behavior based on the Logit model is constructed. By inputting SP survey data on display preferences, the probability of potential franchise vehicles choosing to join under different pricing scenarios is calculated. Combined with the supply scale of potential franchise vehicles in the target area, the daily effective franchise fleet size of the company under a specific pricing scheme is determined, serving as the key input for the daytime vehicle dynamic scheduling module.

[0099] Specifically, the functions of the simulation module for potential franchise vehicle franchise decision-making behavior include:

[0100] (1) Quantification of vehicle revenue under the franchise model:

[0101] Logistics companies m Season q Tianxuan Selection s Pricing strategy call o= Model 1 vehicle transport order i Total benefits for the round trip By m Season q Tianxuan Selection s Pricing strategy call o= Model 1 vehicle transport order i Total income for the round trip ,exist m Season q Tianxuan Selection s Pricing strategy call o= Model 1 vehicle transport order i Fuel cost for the round trip and in m Season q Tianxuan Selection s Pricing strategy call o= Model 1 vehicle transport order i Time value cost of the round trip It consists of three parts:

[0102] ;

[0103] ;

[0104] ;

[0105] ;

[0106] ;

[0107] In the formula ， In order to be in m Season q Orders per day i The transport distance; In order to be in m Season q Tianxuan Selection s Pricing strategy call o= Model 1 vehicle transport order i Return fare; This is the return trip distance coefficient for vehicles; This is the price coefficient for return freight rates; c This refers to the fuel consumption cost of a truck per unit time. The time value per unit kilometer for vehicles; This refers to the average speed of the truck per hour. For the company in m Season q Tianxuan Selection s Pricing strategy call o Transportation service fee for the vehicle model.

[0108] (2) Quantification of vehicle revenue in the crowdsourcing model:

[0109] Logistics companies m Season q Tianxuan Selection s Pricing strategy call o= 2-mode vehicle transportation order i Total benefits for the round trip Depend on o =2 mode vehicle q Tianxuan Selection s Pricing strategy for carrier orders i Total income for the round trip ,exist m Season q Tianxuan Selection s Pricing strategy call o= 2-mode vehicle transportation order i Fuel cost for the round trip and in m Season q Tianxuan Selections Pricing strategy call o= 2-mode vehicle transportation order i Round trip time value cost It consists of three parts;

[0110] ;

[0111] ;

[0112] ;

[0113] In the formula, In order to be in m Season q Tianxuan Selection s Pricing strategy call o= 2-mode vehicle transportation order i Fuel costs for the entire journey; In order to be in m Season q Tianxuan Selection s Pricing strategy call o= 2-mode vehicle transportation order i The total time value cost; In order to be in m Season q Orders per day i The transport distance; for o= 2-mode vehicle transportation order i The total revenue is the fee paid by the logistics company to the crowdsourced vehicles. This represents the unit time value of crowdsourced vehicles.

[0114] (3) Modeling the selection behavior of potential franchise vehicles:

[0115] This embodiment adopts Logit The model analyzes the selection behavior of potential franchise vehicles under different transportation scenarios. Specifically, the model determines the probability of a potential franchise vehicle choosing a franchise company based on the utility it receives in both franchise transportation orders and independent transportation orders in the market.

[0116] ;

[0117] ;

[0118] ;

[0119] ;

[0120] In the formula, Potential franchise vehicles are interested in becomingo= The proportion of vehicles joining the franchise under the Model 1; In order to be in m Season q Tianxuan Selection s Vehicle franchise transportation orders under pricing strategy i The utility; In order to be in m Season q Tianxuan Selection s Vehicle self-operated orders under pricing strategy i The utility; , , All are parameters to be estimated; In order to be in m Season q Tianxuan Selection s Pricing strategy call o= Model 1 vehicle transport order i Total revenue for the entire round trip; In order to be in m Season q Orders per day i The transport distance; In order to be in m Season q Orders per day i Order level; In order to be in m Season q Tianxuan Selection s Pricing strategy call o= 2-mode vehicle transportation order i Total revenue for the entire round trip; The scale of the franchised vehicles; n This represents the total number of potential franchise vehicles within a given area.

[0121] Furthermore, the daytime vehicle dynamic scheduling module enables a dynamic scheduling mechanism for scenarios where the number of orders exceeds the number of available affiliated vehicles for the day. Given that the logistics company knows the customer type, transportation distance, and arrival time window of each order in the daily order set, and considering the uncertainty of order departure and the response time distribution of crowdsourced vehicles, a Monte Carlo simulation method is used to make real-time decisions on whether to use affiliated or crowdsourced vehicles for order execution. This module, from an opportunity cost perspective, constructs a dynamic decision-making model at the order departure time, quantifying the opportunity cost of using affiliated vehicles for the current order and the cost difference caused by the reduction of affiliated vehicles for future orders. By comparing the two, the optimal dispatch strategy is derived, achieving real-time collaborative allocation of affiliated and crowdsourced vehicles during the daytime scheduling process. In other words, it determines in real-time whether the transportation task for each order should be executed by affiliated or crowdsourced vehicles, achieving an optimal balance between scheduling efficiency and transportation costs.

[0122] Specifically, the functions of the daytime vehicle dynamic scheduling module include:

[0123] (1) Orders with confirmed delivery time ID Vehicle dispatching:

[0124] Order ID This refers to an order with a fixed production time at a specific point in time. i Indicates excluding orders ID Foreign order collection I This embodiment actually aims to determine an order that was determined at the current manufacturing time. ID Matching vehicles with franchise or crowdsourcing models:

[0125] ;

[0126] ;

[0127] In the formula, Indicates in m Season q Tian Logistics Company selects s Pricing strategy calls O-mode vehicle transportation orders ID Opportunity cost; These are 0-1 variables used to represent the current order being shipped. ID Whether to use franchised vehicles; Indicates in m Season q Tian Logistics Company does not use the o=1 mode for vehicle transportation orders. ID Opportunity cost; exist m Season q Tian Logistics Company selects s Pricing strategy calls o=1 mode vehicle transportation orders ID Opportunity cost.

[0128] Order ID Do not call o =1 mode vehicle transportation order ID opportunity cost That is, the current orders leaving the factory. ID use o =2 and o =1 mode vehicle transportation cost difference :

[0129] ;

[0130] In the formula, For orders ID use o =1 mode vehicle transportation cost; For order ID o =1 mode vehicle transportation cost.

[0131] use o =1 mode vehicle transportation order ID opportunity cost That is, orders ID use o =1 and o When using the =2 mode vehicle for transportation, the logistics company will handle the remaining orders (orders) on the same day. ID Cost differences for all orders that have been prepared and shipped afterward :

[0132] ;

[0133] ;

[0134] ;

[0135] ;

[0136] ;

[0137] ;

[0138] In the formula, Indicates in m Season q Tian Logistics Company selects s Pricing strategy call o= Model 1 vehicles transport current factory orders. ID After this order, the logistics company called... o =1 mode vehicles and a certain number of o =2 mode vehicles will transport all remaining orders that have not yet left the factory. i Total cost; Indicates in m Season q Tian Logistics Company selects s Pricing strategy call o= Mode 2 vehicles transport current factory orders. ID After this order, the logistics company called... o =1 mode vehicles and a certain number of o =2 mode vehicle transport of remaining orders i Total cost; It is a 0-1 variable used to determine the order at the current moment. i Has the transportation requirement been fulfilled, i.e., if This indicates the current order. iFor shipments that have not yet been completed, the corresponding cost records are the shipping costs for the remaining orders; This refers to the order's production date; The manufacturing time is Order call o =1 mode vehicle transportation costs; The manufacturing time is Order call o =2 mode vehicle transportation costs; and This indicates different manufacturing times for the orders. The corresponding transportation costs for the two transportation modes and With manufacturing time Corresponding probability The product of these values ​​is used to weight the costs at different departure times; This indicates the current order quantity. ID The manufacturing date is used to determine the lower limit of points, i.e., for order calculation. ID The remaining shipping costs for orders after they have left the factory; and The variables are 0 and 1, respectively, used to determine the remaining orders. i Whether the transportation is completed by a franchise vehicle depends on the current order. ID There are two types of vehicles: franchised vehicles and crowdsourced vehicles; Let be the order delivery time distribution function, and the remaining number of franchised vehicles corresponding to the two scenarios are respectively and .

[0139] Furthermore, the feedback mechanism module constructs a feedback iteration mechanism based on the information flow between the franchise vehicle capacity procurement price optimization module, the potential franchise vehicle franchise behavior simulation module, and the daytime vehicle dynamic scheduling module. In the intraday operation simulation, the estimated franchise vehicle scale is corrected in reverse based on the scheduling results, and pricing parameters are further adjusted, thereby achieving closed-loop optimization between pricing, franchise decisions, and vehicle scheduling. Ultimately, it outputs a pricing strategy and a daytime vehicle dynamic scheduling scheme that meet the full-cycle transportation needs and achieve optimal cost.

[0140] Specifically, the core function of the feedback mechanism module is to dynamically correct the original pricing parameters and iteratively optimize the scheduling scheme based on the pricing strategy output by the franchise vehicle capacity procurement price optimization module, combined with the franchise fleet size calculated by the potential franchise vehicle franchise behavior simulation module and the actual operating cost information fed back by the daytime vehicle dynamic scheduling module, ultimately forming a closed-loop feedback system that couples "pricing-response-execution-correction". Its specific implementation steps are as follows:

[0141] Step 1: Initial generation of pricing strategy;

[0142] Step 2: Quantify the scale of the franchise;

[0143] Step 3: Cost Differences Before Scheduling Evaluate;

[0144] Step 4: Identify remaining capacity status;

[0145] Step 5: Cost Differences in Subsequent Orders ( )estimate;

[0146] Step 6: Execute the optimal scheduling decision.

[0147] The daytime vehicle dynamic scheduling module can not only be used as an independent module for dynamic order allocation decisions during operation, but also feed back the daily simulated transportation cost results to the franchise vehicle capacity procurement price optimization module as an important input variable for pricing strategy updates. This enables dynamic optimization and systematic iterative closed-loop control of pricing strategies, ensuring that transportation costs are minimized while guaranteeing service capacity.

[0148] Through iterative and interconnected calculations using the aforementioned feedback mechanism, this embodiment can achieve a closed-loop information system between pricing of franchised vehicles, simulation of franchised vehicle behavior, and daily scheduling strategies. This enhances the system's adaptability to complex and uncertain environments and significantly reduces transportation costs for cold chain logistics companies during seasonal transitions and daily fluctuations.

[0149] This embodiment also provides a cold chain logistics fleet pricing strategy and vehicle scheduling optimization method based on contract and crowdsourcing collaboration, including:

[0150] Based on historical waybill data, predict the distribution parameters of transportation demand;

[0151] Based on the aforementioned transportation demand distribution parameters, an optimization model for the procurement and pricing of franchised vehicle capacity is constructed with the objective of minimizing the total annual transportation cost, and the optimal pricing scheme and freight rate standard for franchised transportation services are output.

[0152] By using a simulation model of potential franchise vehicle decision-making behavior based on the Logit model and SP survey data on revealed preferences, the probability of potential franchise vehicles choosing to join under different pricing scenarios is calculated. Combined with the supply scale of potential franchise vehicles in the target area, the daily effective franchise fleet size of the company under a specific pricing scheme is determined.

[0153] Combining the uncertainty of order delivery time and the distribution of crowdsourced vehicle response time, the Monte Carlo simulation method is used to dynamically decide on the collaborative scheduling of franchised vehicles and crowdsourced vehicles.

[0154] A feedback and iteration mechanism is constructed to optimize pricing parameters and scheduling schemes through iterative linkage, forming a closed-loop feedback and outputting a pricing strategy and daytime vehicle dynamic scheduling scheme that meets the transportation needs of the entire cycle and optimizes costs.

[0155] Furthermore, the pricing schemes include: a composite pricing strategy, a multi-tiered segmented pricing strategy, a static unified pricing strategy, and a package price plus extra-mileage pricing strategy.

[0156] To more clearly illustrate the technical solution of the present invention, specific embodiments are provided below for description:

[0157] A cold chain logistics fleet pricing strategy and vehicle scheduling optimization system based on contract and crowdsourcing collaboration, such as Figure 2 The specific implementation methods include:

[0158] Transportation demand forecasting module:

[0159] The transportation demand forecasting module in this embodiment, based on historical waybill data from cold chain logistics companies, systematically characterizes the temporal and spatial features of transportation demand through multi-dimensional information collection and statistical modeling methods, providing data-driven support for subsequent optimization of franchise capacity procurement pricing and daytime vehicle scheduling modules.

[0160] The transportation demand forecasting module includes the following functional steps:

[0161] (1) Transportation process overview:

[0162] Before the transportation demand forecasting module begins data collection and modeling, it is essential to first clarify the entire operational flow of the cold chain transportation task. The cold chain transportation process mainly includes the following three key stages: the preparation time stage, the vehicle waiting stage, and the transit time stage. The specific flow is as follows: Figure 3 As shown.

[0163] (2) Transportation data collection and preprocessing:

[0164] Historical order data from cold chain logistics companies throughout their complete operational cycle is collected, and key attribute fields for each order are extracted, including order shipment date, transportation distance, order type, factory departure time, arrival time, customer type, and delivery result. Based on the national standard GB / T 36345-2018 "Specifications for Cold Chain Logistics Services for Agricultural Products," the annual operational cycle is divided into two categories: off-season (March-August) and peak season (January-February, September-December) transportation cycles to support the identification and modeling of seasonal fluctuations.

[0165] (3) Identification of daily order quantity distribution:

[0166] A daily order volume time series is constructed based on the order shipment date. Distribution identification and fitting methods are used to identify the probability distribution and corresponding parameters of daily order volume during peak and off-peak seasons. This process automatically determines the distribution type (e.g., normal distribution, ...). t Distributions and their characteristic parameters such as expected value and variance provide quantitative inputs for simulating order fluctuations during peak and off-peak seasons.

[0167] ;

[0168] ;

[0169] In the formula, and These represent the daily number of transportation demand orders during peak and off-peak seasons, respectively. This indicates the shape of the daily order quantity distribution curve during peak season; parameters This represents the average number of daily orders during the off-season. This indicates the degree of dispersion in the daily order quantity values ​​during the off-season.

[0170] (4) Identification of transportation distance distribution:

[0171] Based on the transportation distance of each order, distance intervals are divided, and the proportion of orders within different intervals is statistically analyzed. Furthermore, a probability distribution model of the overall transportation distance is fitted, and the concentration interval of the transportation radius and the probability density function parameters are extracted. The formula below represents the distribution of transportation distances and provides a spatial distribution basis for vehicle operating cost estimation and scheduling optimization:

[0172] ;

[0173] in, d Indicates the order's shipping distance; , , Both represent parameters of a Gaussian distribution.

[0174] (5) Order type and delivery time window identification:

[0175] This embodiment constructs a three-level order classification system based on the differences in how different customers respond to delivery delays. This system is used to characterize the differences in order timeliness requirements and quantify the heterogeneity of timeliness default costs. Specifically, orders are divided into the following three categories: distributor orders, fresh food e-commerce transit warehouse orders, and factory orders.

[0176] ① For e-commerce transit warehouse orders, the delivery time window is mainly concentrated between 00:00 and 04:00 every day, which is highly concentrated and has prominent time sensitivity characteristics;

[0177] ② Distributor orders exhibit two typical time characteristics depending on the delivery distance: When the order distance is less than or equal to 500 kilometers, the delivery time is mainly concentrated between 21:00 and 24:00 at night to accommodate the closing operation cycle after the store closes; while long-distance delivery orders with a distance of more than 500 kilometers mainly arrive after 05:00 the next morning to ensure the customer's inventory needs before opening.

[0178] ③ The receiving time window for factory orders is mainly concentrated between 13:00 and 21:00, which is highly consistent with the rhythm of factory warehousing operations - finished products are mainly shipped out in the morning, and raw materials are arranged to be put into storage in the afternoon.

[0179] (6) Order delivery time distribution identification:

[0180] Based on historical production time data, probability distribution models are built for production times of different order types to identify their distribution types and parameters, characterizing the uncertainty process of order production time and providing dynamic information flow input for intraday dynamic scheduling. Specifically, the production times of orders from distributors, e-commerce transit warehouses, and factories conform to normal, Cauchy, and [other distributions, respectively]. t distributed.

[0181] (7) Modeling of uncertainties in transit time:

[0182] We extract transit time data from historical transportation processes and distinguish between three delivery statuses: early arrival, delayed arrival, and on-time arrival. We then establish probability distribution models for each type of transit time error to assess the range of arrival time fluctuations and the probability of risk caused by the transportation process. This indicates the uncertainty of transit time. Specifically, the error in transit time follows a Rayleigh distribution.

[0183] ;

[0184] In the formula, Indicates order i The value of the time in transit; The scale parameter represents the distribution curve of the uncertainty in order transit time, which determines the width of the curve distribution.

[0185] (8) Identification of the distribution of valid franchised vehicles:

[0186] The distribution of the actual number of franchised vehicles available daily under different seasonal cycles was statistically analyzed and fitted, and its probability distribution form and parameters were extracted to evaluate the distribution trend of vehicle availability under different capacity procurement strategies. Through data fitting, this embodiment found that the distribution of the number of effective franchised vehicles daily during both peak and off-peak transportation periods conforms to a Poisson distribution.

[0187] ;

[0188] in, k This indicates the number of vehicles in the daily franchise model; This represents the average number of vehicles in the daily active franchise model within a transportation period.

[0189] Ultimately, the transportation demand forecasting module outputs a series of high-dimensional probability distribution parameters, including daily order quantity distribution, transportation distance distribution, order type structure, order delivery time distribution, in-transit time disturbance model, and daily effective franchise vehicle quantity distribution. These parameters drive the downstream franchise vehicle pricing model and daytime vehicle scheduling module, enabling high-precision demand forecasting and full-process data support for the cold chain transportation system.

[0190] Franchise Vehicle Capacity Procurement Price Optimization Module: This module, with the logistics company as the decision-making entity, constructs a pricing optimization model for franchise vehicle capacity procurement, aiming to minimize the total annual transportation cost. In this model, the logistics company sets pricing strategies based on transportation demand characteristics, including a single price, a base price, and additional mileage pricing, and incorporates seasonal differences to output the optimal pricing scheme and freight rate standard for franchise transportation services. This pricing result is then passed as input parameters to the potential franchise vehicle franchise decision-making behavior simulation module.

[0191] This embodiment divides the decision-making of two different stakeholders into two levels. In the franchise vehicle capacity procurement price optimization module, the decision-maker (logistics company) decides the pricing strategy for franchise vehicles, and the corresponding optimization model is the model of the franchise vehicle capacity procurement price optimization module.

[0192] (1) Propose a pricing strategy:

[0193] This embodiment proposes a multi-strategy differentiated pricing mechanism for cold chain transportation service procurement scenarios, aiming to enhance logistics companies' pricing flexibility and cost control capabilities under varying transportation demand conditions. Combining existing mainstream pricing methods in the freight market with practical operational experience, this embodiment summarizes and designs four main categories and six pricing models, forming a multi-dimensional, switchable pricing strategy system, the structure of which is as follows: Figure 4 As shown.

[0194] The four pricing strategies include: composite pricing strategy (ST strategy), multi-level segmented pricing strategy (Segment strategy), static unified pricing strategy (Fixed strategy), and package price + extra mileage pricing strategy (Hybrid strategy).

[0195] Specifically as follows:

[0196] Category 1: Compound pricing strategies ( ST Strategy):

[0197] The first pricing strategy adopts a "base price + extra kilometer price" structure model, distinguishing between two implementation schemes: a low base price with a high rate per unit kilometer, and a high base price with a low rate per unit kilometer. The pricing structure is set for different order distances. Based on a fixed unit price p 1 Billing. When the order delivery distance... The portion exceeding this will be charged at an additional unit price. p 2 Billing. The formula for the transportation service fee for vehicles under this strategy in the franchise model is as follows:

[0198] ;

[0199] in, Indicates the period m、 No. q The day, orders for vehicle transportation under the franchise model were processed. i Transportation service fees; This is the first transport distance threshold.

[0200] The second type: Multi-level segmented pricing strategy (Segment strategy):

[0201] This strategy is based on the base freight rate. p 3 Based on this, for those exceeding different transport distance thresholds (e.g.: , For each order, a differentiated progressive additional coefficient is set (e.g.: , This implements a cost control mechanism that achieves non-linear growth in transport distance across segments. Its transport service fee function is as follows:

[0202] ;

[0203] In the formula, , These are the second and third transport distance thresholds, respectively.

[0204] The third type: Static uniform pricing strategy (Fixed strategy):

[0205] The third strategy uses a fixed unit price structure, does not differentiate between different shipping distances, and applies a uniform price to all orders. This strategy sets a minimum price limit. With the highest price limit The transportation service fees are expressed as follows:

[0206] ;

[0207] .

[0208] Category 4: Package Price + Extra Mileage Pricing Strategy (Hybrid Strategy)

[0209] The fourth strategy is suitable for short-distance order scenarios. Logistics companies divide the order distance into four package price ranges (e.g., ...). , , , Each range has a corresponding fixed price ( h 1 、h 2 、h 3 、h 4 If the order's shipping distance exceeds the longest package price range. Then the unit price is used. Billing. The transportation service fee function is as follows:

[0210] ;

[0211] In the formula, , , , These are the fourth, fifth, sixth, and seventh transport distance thresholds, respectively. h 1 、h 2 、h 3 、h 4 , h 5 These are the shipping costs corresponding to different shipping cost ranges.

[0212] (2) Total cost of logistics company:

[0213] Logistics companies employ a model combining crowdsourcing and franchising, dynamically allocating vehicle resources to meet order transportation demands. During peak transportation periods, due to a significant increase in daily factory transport volume, logistics companies need to recruit a large number of franchised vehicles to cope with the surge in transportation demand. However, when transportation demand transitions from peak to off-peak seasons, demand drops sharply, forcing companies to reduce the scale of franchised vehicles. In this process, layoffs may negatively impact the company's market reputation; therefore, to maintain the company's reputation, severance costs are included in the overall cost consideration. Based on this, the total cost of a logistics company over a complete peak and off-peak transportation cycle mainly consists of the following two parts: the transportation cost of the logistics company deploying vehicles to transport orders and the severance costs incurred when the logistics company transitions from peak to off-peak seasons. The specific calculation formula is as follows:

[0214] ;

[0215] ;

[0216] ;

[0217] In the formula ， This refers to the sum of transportation costs for logistics companies using franchised and crowdsourced vehicles for orders under a pricing strategy, freight rate standard, and transportation cycle, and the costs of severance pay during the transition from peak to off-peak seasons; peak season ( m =1) and off-season ( m =2), A set representing the period of transportation; O This represents the collection of vehicle types called by the logistics company; This represents the set of days in a transportation cycle for a logistics company. Indicates that the logistics company is m Season q Tianxuan Selection s Pricing strategy call o Model vehicle transportation orders i Transportation costs; Cut one job at the logistics company o =1 mode vehicle severance cost; This refers to the number of franchised vehicles that logistics companies reduce when the peak season turns into the off-season.

[0218] (3) Costs incurred by logistics companies in dispatching vehicles to transport orders:

[0219] Logistics companies m Season q Tianxuan Selection s Pricing strategy call o Total cost of the model vehicle Transportation service fee Early arrival waiting costs Late arrival penalty cost composition;

[0220] ;

[0221] ;

[0222] In the formula, For the company in m Season q Tianxuan Selection s Pricing strategy call o Transportation service fees for the model vehicles; In order to be in m Season q Tianxuan Selection sPricing strategy call o Model vehicle transportation orders i The waiting cost of arriving early; In order to be in m Season q Tianxuan Selection s Pricing strategy call o Model vehicle transportation orders i The cost of being late; Peak season ( m =1) Number of franchised vehicles during the period; Off-season m =2) Number of vehicles joining the franchise during the period; Peak season ( m =1) and off-season ( m =2) The difference in the number of franchised vehicles during the period.

[0223] (4) Costs incurred by the company in using crowdsourced vehicles to transport orders:

[0224] The company m Season q Tianxuan Selection s Pricing strategy call o =2 mode vehicle transportation orders i Transportation service fee :

[0225] ;

[0226] in, In order to be in m Season q Tianxuan Selection s Pricing strategy call o= 2-mode vehicle transportation order i Pricing; In order to be in m Season q Orders per day i The distance of transport.

[0227] (5) Cold chain logistics order delivery time calculation model:

[0228] The order's shipping process begins from the factory departure time. The company determines the delivery plan based on the order's waiting time, vehicle type, and delivery route. The final arrival time of the order consists of three parts: the order's factory departure time (order creation time). Waiting time (i.e., the time from the order's production time to the completion of vehicle dispatch) Travel time (i.e., the time it takes for a vehicle to travel from its origin to its destination) 。 Therefore, the total time for an order to reach the customer can be expressed as the following expression:

[0229] ;

[0230] In the formula, In order to be in m Season q Tianxuan Selection s Pricing strategy call o Model vehicle transportation orders i Final arrival time; In order to be in m Season q Tianxuan Selection s Pricing strategy call o Model vehicle transportation orders i The manufacturing date; In order to be in m Season q Tianxuan Selection s Pricing strategy call o Model vehicle transportation orders i Waiting time; In order to be in m Season q Tianxuan Selection s Pricing strategy call o Model vehicle transportation orders i Transit time. When the logistics company calls... o =1 mode vehicle transportation order i At that time, the waiting time is 0.

[0231] When the logistics company calls o =2 mode vehicle transportation orders i When the waiting time is specified, it is any possible value within the time interval consisting of the order's factory departure time and the latest time allowed by the factory for the logistics company to dispatch a vehicle.

[0232] (6) Early arrival waiting and late delivery penalties:

[0233] Customers negotiate a mutually agreed-upon delivery window with the factory based on the distance to the shipping location and business hours. If an order transported by a logistics company arrives before the delivery window, the vehicle must wait at the location. To prevent spoilage, the vehicle waits at the location; therefore, the cost of an order arriving early is the cost of the vehicle waiting at the location. The specific calculation formula is as follows:

[0234] ;

[0235] ;

[0236] In the formula, The unit time waiting cost for trucks waiting to unload at the customer's unloading point after the vehicle arrives at the unloading point; For the company in m Season q Tianxuan Selection s Pricing strategy call o Model vehicle transportation orders i The time of arrival at the customer's location; For the company in m Season q Tianxuan Selection s Pricing strategy call o Model vehicle transportation orders i The lower limit of the delivery time window; It is a 0-1 variable used to indicate whether the order delivery time is earlier than the lower limit of the delivery time window (i.e., the earliest arrival time).

[0237] When goods arrive late, different types of customers react differently: some choose to continue receiving them, some refuse to accept them, and others request delivery within the next specified time window. The penalty costs borne by the logistics company vary accordingly. Specifically, when the order type is for distributors, their downstream customers are usually loyal and will continue to accept the goods upon arrival, but as the delay lengthens, these customers may gradually churn. When the order type is for e-commerce transit warehouses, customers often refuse to accept the goods if they arrive late. In this case, the logistics company bears all the losses incurred by customers due to unsold goods caused by the logistics company's late arrival. When the order type is for factories, if the order carried by the logistics company's truck arrives late, the truck usually has to wait at the factory until the same delivery time the next day before it can deliver the goods. In this case, the logistics company not only incurs additional waiting costs.

[0238] The simulation module for potential franchise vehicle decision-making behavior takes independent potential franchise vehicles in the market as the decision-making subjects. Based on the franchise vehicle pricing strategy and its corresponding fare standard, combined with the potential franchise vehicles' expected transportation revenue (including transportation income, costs, order level, etc.), a simulation model of the potential franchise vehicle decision-making behavior based on the Logit model is constructed. By inputting SP survey data on display preferences, the probability of potential franchise vehicles choosing to join under different pricing scenarios is calculated. Combined with the supply scale of potential franchise vehicles in the target area, the daily effective franchise fleet size of the company under a specific pricing scheme is determined, serving as the key input for the daytime vehicle dynamic scheduling module.

[0239] In the simulation module for potential franchise vehicle decision-making behavior, the decision-maker (potential franchise vehicle) decides whether to join the company, and the corresponding optimization model is the model of the simulation module for potential franchise vehicle decision-making behavior.

[0240] (1) Quantification of vehicle revenue under the franchise model:

[0241] Logistics companies m Season q Tianxuan Selection s Pricing strategy call o= Model 1 vehicle transport order i Total benefits for the round trip By m Season q Tianxuan Selection s Pricing strategy call o= Model 1 vehicle transport order i Total income for the round trip ,exist m Season q Tianxuan Selection s Pricing strategy call o= Model 1 vehicle transport order i Fuel cost for the round trip and in m Season q Tianxuan Selection s Pricing strategy call o= Model 1 vehicle transport order i Time value cost of the round trip It consists of three parts:

[0242] ;

[0243] ;

[0244] ;

[0245] ;

[0246] ;

[0247] In the formula ， In order to be in m Season q Orders per day i The transport distance; In order to be in m Season q Tianxuan Selection s Pricing strategy call o= Model 1 vehicle transport order i Return fare; This is the return journey distance coefficient; This is the price coefficient for return freight rates; c This refers to the fuel consumption cost of a truck per unit time. The time value per unit kilometer for vehicles; This refers to the average speed of the truck per hour. For the company in m Season q Tianxuan Selection s Pricing strategy call o Transportation of modular vehicles.

[0248] (2) Quantification of vehicle revenue in the crowdsourcing model:

[0249] Logistics companies m Season q Tianxuan Selection s Pricing strategy call o= 2-mode vehicle transportation order i Total benefits for the round trip Depend on o =2 mode vehicle q Tianxuan Selection s Pricing strategy for carrier orders i Total income for the round trip ,exist m Season q Tianxuan Selection s Pricing strategy call o= 2-mode vehicle transportation order i Fuel cost for the round trip and in m Season q Tianxuan Selection s Pricing strategy call o= 2-mode vehicle transportation order i Round trip time value cost It consists of three parts;

[0250] ;

[0251] ;

[0252] ;

[0253] In the formula, In order to be in m Season q Tianxuan Selection s Pricing strategy call o= 2-mode vehicle transportation order i Fuel costs for the entire journey; In order to be in m Season q Tianxuan Selection s Pricing strategy call o= 2-mode vehicle transportation order i The total time value cost; In order to be in m Season qOrders per day i The distance of transport.

[0254] (3) Modeling the selection behavior of potential franchise vehicles:

[0255] This embodiment adopts Logit The model analyzes the selection behavior of potential franchise vehicles under different transportation scenarios. Specifically, the model determines the probability of a potential franchise vehicle choosing a franchise company based on the utility it receives in both franchise transportation orders and independent transportation orders in the market.

[0256] ;

[0257] ;

[0258] ;

[0259] ;

[0260] In the formula, Potential franchise vehicles are interested in becoming o= The proportion of vehicles joining the franchise under the Model 1; In order to be in m Season q Tianxuan Selection s Vehicle franchise transportation orders under pricing strategy i The utility; In order to be in m Season q Tianxuan Selection s Vehicle self-operated orders under pricing strategy i The utility; , , These are the parameters to be estimated; In order to be in m Season q Tianxuan Selection s Pricing strategy call o= Model 1 vehicle transport order i Total revenue for the entire round trip; In order to be in m Season q Orders per day i The transport distance; In order to be in m Season q Orders per day i Order level; In order to be in m Season q Tianxuan Selection s Pricing strategy call o= 2-mode vehicle transportation order iTotal revenue for the entire round trip; The scale of the franchised vehicles; n This represents the total number of potential franchise vehicles within a given area.

[0261] The daytime vehicle dynamic scheduling module addresses scenarios where the number of orders exceeds the number of available affiliated vehicles for the day. It employs a dynamic scheduling mechanism, assuming the logistics company has prior knowledge of each order's customer type, transport distance, and arrival time window. Combining the uncertainty of order departure and the response time distribution of crowdsourced vehicles, it uses Monte Carlo simulation to make real-time decisions on whether to use affiliated or crowdsourced vehicles for order execution. From an opportunity cost perspective, this module constructs a dynamic decision-making model at the order departure time, quantifying the opportunity cost of using affiliated vehicles for the current order and the cost difference caused by a reduction in affiliated vehicles for future orders. By comparing these two factors, it derives the optimal dispatch strategy, enabling real-time collaborative allocation of affiliated and crowdsourced vehicles during the daytime scheduling process. In other words, it determines in real-time whether each order's transportation task should be performed by an affiliated or crowdsourced vehicle, achieving an optimal balance between scheduling efficiency and transportation costs.

[0262] (1) Orders with confirmed delivery time ID Vehicle dispatching:

[0263] Order ID This refers to an order with a fixed production time at a specific point in time. i Indicates excluding orders ID Foreign order collection I A specific order. What this invention actually aims to determine is an order that was determined at the current manufacturing time. ID Matched with vehicles in the franchise model or crowdsourcing model.

[0264] ;

[0265] ;

[0266] in, Indicates in m Season q Tian Logistics Company selects s Pricing strategy call o Model vehicle transportation orders ID Opportunity cost; Indicates in m Season q Tian Logistics Company selects s Pricing strategy utilizes crowdsourcing model ( o =2) Vehicle transport orders ID With the call to franchise model ( o =1) Vehicle, for the current order ID The cost difference; Indicates in m Season q Tian Logistics Company selects s Pricing strategy utilizes franchise model for vehicle transportation orders ID Compared to using crowdsourced vehicles, for all subsequent orders on the same day (i.e., orders...) ID The opportunity cost difference for all orders that are subsequently prepared and shipped.

[0267] Order ID Do not call o =1 mode vehicle transportation opportunity cost Orders ID use o =2 and o =1 mode vehicle transportation cost differences :

[0268] ;

[0269] In the formula, For orders ID use o =1 mode vehicle transportation cost; For orders ID use o =1 mode vehicle transportation cost;

[0270] Order ID Call o =1 mode vehicle opportunity cost That is, orders ID use o =1 and o When using a vehicle operating under the =2 mode, the cost difference for the logistics company handling the remaining orders on the same day is calculated. :

[0271] ;

[0272] ;

[0273] ;

[0274] ;

[0275] ;

[0276] ;

[0277] ;

[0278] In the formula, Indicates in m Seasonq Tian Logistics Company selects s Pricing strategy call o= Model 1 vehicle transport order ID Afterwards, the logistics company called upon... o =1 mode vehicles and a certain number of o =2 mode vehicle transport of remaining orders i Opportunity cost; Indicates in m Season q Tian Logistics Company selects s Pricing strategy call o= Model 1 vehicle transport order ID Afterwards, the logistics company called upon... o =1 mode vehicles and a certain number of o =2 mode vehicle transport of remaining orders i The cost; Indicates in m Season q Tian Logistics Company selects s Pricing strategy call o= 2-mode vehicle transportation order ID Afterwards, the logistics company called upon... o =1 mode vehicles and a certain number of o =2 mode vehicle transport of remaining orders i The cost; In order to be in m Season q Heavenly Call o= Model 1 vehicle transport order i The manufacturing date; Indicates in m Season q Heavenly Call o Model vehicle cost carrier order i Possible manufacturing time t The probability density distribution function; The manufacturing time is Order call o =1 mode vehicle transportation costs; The manufacturing time is Order call o =2 mode vehicle transportation costs; and This indicates different manufacturing times for the orders. The corresponding transportation costs for the two transportation modes and With manufacturing time Corresponding probability The product of these values ​​is used to weight the costs at different departure times; This indicates the current order quantity. ID The manufacturing date is used to determine the lower limit of points, i.e., for order calculation. ID The remaining shipping costs for orders after they have left the factory; k This indicates the number of vehicles that are currently active in the franchise area. X This indicates the number of franchise vehicles remaining on the day during routine scheduling. and The variables are 0 and 1, respectively, used to determine the remaining orders. i Whether the transportation is completed by a franchise vehicle depends on the current order. ID There are two types of vehicles: franchised vehicles and crowdsourced vehicles.

[0279] The feedback mechanism module constructs a feedback iteration mechanism based on the information flow between the franchise vehicle capacity procurement price optimization module, the potential franchise vehicle franchise behavior simulation module, and the daytime vehicle dynamic scheduling module. In the daytime operation simulation, the estimated franchise vehicle scale is corrected in reverse based on the scheduling results, and pricing parameters are further adjusted. This achieves closed-loop optimization between pricing, franchise decisions, and vehicle scheduling, ultimately outputting a pricing strategy and daytime vehicle dynamic scheduling scheme that meets the full-cycle transportation needs and optimizes costs.

[0280] In this embodiment, the feedback mechanism module is used to realize the linkage and closed-loop iteration between the optimization of the pricing strategy for franchised vehicles, the prediction of the franchised vehicle behavior, and the daytime dynamic scheduling strategy. Given that the integrated pricing-decision-scheduling system constructed in this embodiment involves multiple modules, and the overall solution belongs to the NP-hard problem type, this embodiment adopts a cyclical iterative strategy between multiple modules to obtain an approximate optimal solution.

[0281] The core function of the feedback mechanism module is to dynamically correct the original pricing parameters and iteratively optimize the scheduling scheme based on the pricing strategy output by the franchise vehicle capacity procurement price optimization module, combined with the franchise fleet size calculated by the potential franchise vehicle behavior simulation module and the actual operating cost information fed back by the daytime vehicle dynamic scheduling module. This ultimately forms a closed-loop feedback system that couples "pricing-response-execution-correction". The specific implementation steps are as follows:

[0282] Step 1: Initial generation of pricing strategy:

[0283] The franchise vehicle capacity procurement price optimization module sets the base freight rate and differentiated surcharge rate for franchise vehicles based on the transportation cycle and service range, generating a set of alternative pricing strategies.

[0284] Step 2: Quantifying Franchise Scale

[0285] The pricing strategy set is input into the potential franchise vehicle franchise behavior simulation module. Based on the Logit model and the response parameters fitted by SP survey data, the probability of potential franchise vehicles joining for each pricing strategy is calculated, and the franchise fleet size under different strategies is determined accordingly.

[0286] Step 3: Cost variance assessment before scheduling:

[0287] In the daytime operation simulation, orders were calculated separately at the time of order shipment. ID The difference between the franchise and crowdsourced transportation costs for the current order is calculated based on the real-time transportation costs of both franchised and crowdsourced vehicles. .

[0288] Step 4: Remaining Capacity Status Identification:

[0289] Based on the current scheduling node, determine the remaining number of valid franchise vehicles under both the franchise mode and the crowdsourcing mode for the current order, and record it as follows: and .

[0290] Step 5: Estimation of Cost Differences for Subsequent Orders:

[0291] The set of unscheduled orders following the current order will be processed separately in... and The problem is transformed into a static 0-1 integer programming problem under the configuration of franchised transportation capacity. The optimal franchised vehicle scheduling scheme is solved, and the difference in the total transportation cost of subsequent orders under the two scenarios is calculated. .

[0292] Step 6: Execute the optimal scheduling decision:

[0293] If the overall opportunity cost of the current order is greater under the franchise vehicle scheme than under the crowdsourcing vehicle scheme, then the following conditions are met. If the order fails to meet the minimum requirement, it will be transported by affiliated vehicles; otherwise, it will be transported by crowdsourced vehicles. This principle is applied to each order individually, generating the optimal daily order-vehicle matching scheme and corresponding daily transportation cost. The daytime vehicle dynamic scheduling module can not only be used as an independent module for dynamic order allocation decisions during operation, but also feed back the daily simulated transportation cost results to the affiliated vehicle capacity procurement price optimization module. This serves as an important input variable for updating the pricing strategy, thereby achieving dynamic optimization and systematic iterative closed-loop control of the pricing strategy, ensuring the minimization of transportation costs while guaranteeing service capacity.

[0294] Through iterative and interconnected calculations via a feedback mechanism, this embodiment can achieve a closed-loop information system between pricing of franchised vehicles, simulation of franchised vehicle behavior, and daily scheduling strategies. This enhances the system's adaptability to complex and uncertain environments and significantly reduces transportation costs for cold chain logistics companies during seasonal transitions and daily fluctuations.

[0295] The above are merely preferred embodiments of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A cold-chain logistics fleet pricing strategy and vehicle scheduling optimization system based on contract and crowdsourcing collaboration, characterized in that, Comprise: Transportation demand prediction module: for identifying daily order quantity, transportation distance distribution, order type structure, departure time distribution and in-transit time fluctuation parameters based on historical shipment data; Franchise vehicle capacity procurement price optimization module: for constructing a franchise vehicle capacity procurement pricing optimization model with the objective of minimizing total annual transportation cost, outputting the optimal pricing scheme and transportation price standard of franchise transportation services; Potential franchise vehicle joining decision behavior simulation module: for calculating the probability of potential franchise vehicles choosing to join under different pricing scenarios through a potential franchise vehicle decision behavior simulation model based on Logit model and revealed preference SP survey data, and determining the daily effective franchise fleet size under a specific pricing scheme by combining the potential franchise vehicle supply size in the target area, serving as a key input for the daily vehicle dynamic scheduling module; Daily vehicle dynamic scheduling module: for dynamically deciding the collaborative scheduling of franchise vehicles and crowd-sourced vehicles by adopting Monte Carlo simulation method in combination with order departure time uncertainty and crowd-sourced vehicle response time distribution; Wherein, the dynamic decision of the collaborative scheduling of franchise vehicles and crowd-sourced vehicles comprises: ； ； wherein denotes the use of m seasonal q the day the logistics company chooses to use s the pricing strategy o = 1 mode vehicle to carry the order ID the opportunity cost of is a 0-1 variable indicating whether the current factory order ID uses a franchise vehicle; denotes the use of m seasonal q the day the logistics company chooses to use s the pricing strategy o = 1 mode vehicle to carry the order ID the opportunity cost of denotes the use of m seasonal q the day the logistics company chooses to use s the pricing strategy o = 1 mode vehicle to carry the order ID the opportunity cost of Feedback mechanism module: for constructing a feedback iteration mechanism to iteratively optimize pricing parameters and scheduling schemes, forming a closed-loop feedback, and outputting a pricing strategy and daily vehicle dynamic scheduling scheme that meet the transportation demand and cost optimization in the whole cycle.

2. The system according to claim 1, wherein, The transportation demand prediction module depicts the difference in time efficiency demand of orders and quantifies the heterogeneity of time efficiency default cost by constructing an order classification system, wherein the order classification system comprises distributor orders, e-commerce transfer warehouse orders and factory orders, the distributor orders obey a normal distribution, the e-commerce transfer warehouse orders obey a Cauchy distribution, and the factory orders obey a T distribution. 3.The cold-chain logistics fleet pricing strategy and vehicle dispatching optimization system based on crowdsourcing and contract synergy according to claim 1, wherein, The franchise vehicle capacity procurement pricing optimization model outputs the optimal pricing scheme and transportation price standard of franchise transportation services by setting a pricing strategy and combining differentiated factors in off-peak and peak seasons; Wherein, the pricing strategy includes the transportation cost of orders carried by vehicles called by the logistics company and the dismissal cost of the logistics company when switching from the transportation peak season to the transportation off-peak season; The transportation cost of orders carried by vehicles called by the logistics company includes transportation service fee, early arrival waiting cost and late arrival penalty cost.

4. The system according to claim 3, wherein, The late arrival penalty cost is calculated differently according to order type: ； ； ； In the formula, For orders i The cost of being late; It is a 0-1 variable used to indicate whether the delivery time is later than the upper limit of the delivery time window; For the company in m Season q Tianxuan Selection s Pricing strategy call o Model vehicle transportation orders i The final arrival time; For the company in m Season q Tianxuan Selection s Pricing strategy call o Model vehicle transportation orders i The upper limit of the customer's delivery time window; The cost of vehicles arriving late while waiting to be unloaded at the factory; The late arrival time cost for distributors' secondary customers; The time cost of late orders arriving at e-commerce transit warehouses; Indicates order i The type is a factory type; Indicates order i The type is distributor; Indicates order i The type is an e-commerce transit warehouse; This refers to the unit tonnage loss cost incurred by the logistics company after the e-commerce transit warehouse refuses to accept the goods. Indicates that the logistics company is m Transportation period q Tiancheng Transport's orders i The tonnage. 5.The cold-chain logistics fleet pricing strategy and vehicle dispatching optimization system based on crowdsourcing and contract synergy according to claim 1, The probability of potential franchise vehicles choosing to join under different pricing scenarios is calculated as follows: ； ； ； ； In the formula, Potential franchise vehicles are interested in becoming o= The proportion of vehicles joining the franchise under the Model 1; In order to be in m Season q Tianxuan Selection s Vehicle franchise transportation orders under pricing strategy i The utility; In order to be in m Season q Tianxuan Selection s Vehicle self-operated orders under pricing strategy i The utility; , , All are parameters to be estimated; In order to be in m Season q Tianxuan Selection s Pricing strategy call o= Model 1 vehicle transport order i Total revenue for the entire round trip; In order to be in m Season q Orders per day i The transport distance; In order to be in m Season q Orders per day i Order level; In order to be in m Season q Tianxuan Selection s Pricing strategy call o= 2-mode vehicle transportation order i Total revenue for the entire round trip; The scale of the franchised vehicles; n This represents the total number of potential franchise vehicles within a given area. 6.The cold-chain logistics fleet pricing strategy and vehicle dispatch optimization system based on crowdsourcing and contract synergy according to claim 1, wherein, The calculation is in m Quartile q The opportunity cost to the o = 1 mode vehicle carrier orders ID The opportunity cost is: ； wherein for the order ID using o transportation costs for the =1 mode vehicle; for the order ID using o transportation costs for the =2 mode vehicle; The calculation is in m Quartile q The opportunity cost o = 1 mode vehicle carrier orders ID The opportunity cost is: ； ； ； ； In the formula, is the order ID the probability corresponding to the time of the factory is the distribution density function of the order delivery time; represents the delivery time of the order m season q The logistics company chooses s Pricing strategy call o= 1 mode vehicle to carry orders i After the order, the logistics company calls o =1 mode vehicle and a number o =2 mode vehicle to carry all the remaining orders i The total cost is; represents the delivery time of the order m season q The logistics company chooses s Pricing strategy call o= 2 mode vehicle to carry orders i After the order, the logistics company calls o =1 mode vehicle and a number o =2 mode vehicle to carry the remaining orders i The total transportation cost is; is a 0-1 variable, used to determine whether the current time order i has completed the transportation demand; is the order delivery time distribution function; is the delivery time of the order is the order with The transportation cost of calling o =1 mode vehicle to carry the corresponding transportation cost; is the order with The transportation cost of calling o =1 mode vehicle to carry the corresponding transportation cost; represents the delivery time of the order ID currently delivered; and are both 0-1 variables, used to decide whether the remaining orders i are completed by the franchise vehicle, respectively corresponding to the current order ID by the franchise vehicle and the crowd-sourcing vehicle. 7.The contract and crowdsourcing collaborative cold chain logistics fleet pricing strategy and vehicle scheduling optimization system of claim 1, wherein, The closed-loop feedback is formed by: According to the scheduling results, the franchise vehicle size estimation is corrected in reverse, the corrected parameters are input into the franchise vehicle capacity procurement pricing optimization model to generate a new pricing strategy, and the iteration process is repeated until the total transportation cost converges to the optimal value.

8. A cold-chain logistics fleet pricing strategy and vehicle scheduling optimization method based on contract and crowdsourcing collaboration, characterized in that, Comprise: Based on historical shipment data, predict transportation demand distribution parameters; Based on the transportation demand distribution parameters, construct a franchise vehicle capacity procurement pricing optimization model with the objective of minimizing total annual transportation cost, outputting the optimal pricing scheme and transportation price standard of franchise transportation services; Through a potential franchise vehicle decision behavior simulation model based on Logit model and revealed preference SP survey data, calculate the probability of potential franchise vehicles choosing to join under different pricing scenarios, and determine the daily effective franchise fleet size under a specific pricing scheme by combining the potential franchise vehicle supply size in the target area, serving as a key input for the daily vehicle dynamic scheduling module; In combination with the order delivery time uncertainty and the crowd-sourcing vehicle response time distribution, a Monte Carlo simulation method is used to dynamically decide the cooperative scheduling of the franchise vehicles and the crowd-sourcing vehicles. The dynamic decision of the cooperative scheduling of the franchise vehicles and the crowd-sourcing vehicles includes: ； ； wherein denotes the number of days m quarter q the number of days s the number of days o the number of days ID the opportunity cost of a mode vehicle shipment order is a 0-1 variable indicating whether the current shipment order is shipped by a franchise vehicle ID is a 0-1 variable indicating whether the current shipment order is shipped by a franchise vehicle denotes the number of days m quarter q the number of days s the number of days o the opportunity cost of a mode vehicle shipment order ID denotes the number of days quarter m the number of days q the number of days s the number of days o the opportunity cost of a mode vehicle shipment order ID denotes the number of days A feedback iteration mechanism is constructed to iteratively optimize the pricing parameters and the scheduling scheme, form a closed-loop feedback, and output a pricing strategy and a daily vehicle dynamic scheduling scheme that meet the whole-cycle transportation demand and cost optimization.

9. The method of claim 8, wherein, The pricing scheme includes a composite pricing strategy, a multi-stage segmented pricing strategy, a static uniform pricing strategy, and a package price + extra-kilometer pricing strategy.

Images