Express network cargo transfer game decision-making method based on dynamic busy degree

Abstract

The dynamic busy degree-based express delivery network cargo transfer game decision-making method is characterized by comprising the following steps: each logistics collection and distribution center in a region discloses current logistics package accumulation data and average package processing speed data to all express delivery networks; each express branch obtains distance information sent to different logistics distribution centers, and calculates distance satisfaction of the different logistics distribution centers; each express branch calculates an alliance revenue function and an individual revenue function according to the received public data of the logistics distribution center and the obtained distance data; all the express branches form an alliance through a game model, and a cooperation strategy is optimized through Nash equilibrium; and each express network point carries out distributed independent decision making according to the calculated individual income, and local information communication among the express network points is allowed. According to the invention, a distributed algorithm is adopted to endow each express delivery network with the right to autonomously select the logistics distribution center, and the average cargo sending efficiency of all the express delivery networks is significantly improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent logistics, and relates to a game decision-making method for cargo transfer in express delivery outlets based on dynamic busyness, which is applicable to scenarios where multiple express delivery outlets can independently select different logistics distribution centers for cargo transfer. Background Art

[0002] With the booming development of the e-commerce industry, the demand for cargo sending in express delivery outlets has shown an explosive growth. Especially during some important e-commerce shopping festivals (such as before Spring Festival, back-to-school season, Double 11, etc.), consumers place orders intensively under the stimulation of time-limited discounts and full-reduction offers, causing the order volume to soar several times or even dozens of times in a short period. The express delivery outlets serving e-commerce merchants have witnessed a "blowout" growth in the shipment volume. If the goods cannot be sent out in time at this time, it will not only easily lead to a sharp increase in consumer returns, affecting the order turnover of merchants, but may also trigger third-party mechanisms such as platform fines or a decrease in store ratings, causing hidden revenue losses to merchants. Therefore, the efficiency of cargo distribution and the ability to cope with the risk of explosive growth in logistics have become the primary considerations for merchants when choosing express delivery companies. To win the competition, express delivery companies usually use a centralized logistics scheduling system to actively conduct logistics transfer and allocation between express delivery outlets and logistics distribution centers. Although this scheduling method can ensure the overall logistics sending efficiency of the express delivery company, it cannot guarantee the sending efficiency of many small and medium-sized express delivery outlets, and still causes a large amount of cargo accumulation in the outlets. If the express delivery outlets independently select the logistics distribution center for cargo transfer, it requires numerous express delivery outlets to make scientific and reasonable selection decisions based on the busyness of the distribution center to avoid a large number of concentrated transfer demands causing a certain distribution center to fall into an overloaded operation state. Therefore, designing a reasonable and efficient cargo transfer decision-making method for express delivery outlets has become an urgent need for express delivery companies to improve the logistics scheduling system and enhance their business competitive advantages. Currently, there is no relevant decision-making method for the above-mentioned needs, and each logistics company still uses the method of fixed association between express delivery outlets and logistics distribution centers during off-peak hours and centralized scheduling decisions during peak hours. For example, Patent CN202411264586.2 discloses a method for selecting the location of a logistics distribution center based on logistics services and costs. This method only focuses on a single link or a static logistics transfer scenario. After determining the location selection result, the allocation relationship between the express delivery outlet and the logistics distribution center is fixed, without fully considering the impact of the dynamic busyness of the logistics distribution center on the shipping decision. In the case of an increase in the shipment volume, it is very easy to lead to a decrease in the transfer efficiency, resulting in cargo accumulation. Summary of the Invention

[0003] Aiming at the defects in the prior art, the present invention provides a game decision-making method for cargo transfer in express delivery outlets based on dynamic busyness.

[0004] A method for game decision-making on goods transfer in express delivery outlets based on dynamic busyness includes the following steps:

[0005] Step S1: Each logistics distribution center within the region discloses the current logistics parcel accumulation data and the average parcel processing speed data to all express delivery outlets;

[0006] Step S2: Each express delivery outlet obtains the distance information to different logistics distribution centers and calculates the distance satisfaction degree for different logistics distribution centers;

[0007] Step S3: Each express delivery outlet calculates the coalition revenue function and the individual revenue function according to the public data of the logistics distribution center received in Step S1 and the distance data obtained in Step S2;

[0008] Step S4: Each express delivery outlet forms a coalition through a game model and optimizes the cooperation strategy through Nash equilibrium;

[0009] Step S5: Each express delivery outlet makes a distributed independent decision according to the individual revenue calculated in Step S3. To promote decision consistency, local information communication is allowed among express delivery outlets to coordinate the shipping choices.

[0010] Further, in Step S1, the current logistics parcel accumulation data and the average parcel processing speed data are collected in real time through the internal logistics management system of the logistics distribution center.

[0011] Further, the parcel accumulation data is the total weight of the parcels to be processed.

[0012] Further, the average parcel processing speed data is obtained by statistically calculating the weight of the parcels processed within a unit time and is quantitatively represented in kilograms per hour.

[0013] Further, the distance satisfaction degree in Step S2 is specifically:

[0014] where d ij is the distance between the obtained express delivery outlet and the logistics distribution center, max(d ij ) is the maximum distance, and the distance satisfaction degree is the normalization processing of the distance.

[0015] Further, the coalition revenue function in Step S3 is specifically:

[0016] where t j is the time for each logistics distribution center to process parcels, obtained by dividing the total parcel weight by the average parcel processing speed. Among them, the total parcel weight is obtained by adding the current logistics parcel accumulation amount obtained in Step S1 and the parcel amount of the express delivery outlets about to ship to this logistics distribution center.

[0017] The individual revenue function is specifically as follows:

[0018] Furthermore, the game model in step S4 is specifically where: P represents the set of players, corresponding to the set of all express delivery outlets in the region; S represents the set of strategies, corresponding to the set of alliances (i.e., logistics distribution centers) that each express delivery outlet can choose to join; U represents the set of individual revenues of each express delivery outlet; and E represents the set of alliance revenues.

[0019] The purpose of forming an alliance is to achieve a stable allocation structure A = {A1, A2, …, A j , …, A n}, where A j represents the set of express delivery outlets that choose logistics distribution center c j .

[0020] Define the allocation structure that reaches the Nash equilibrium as A * , that is, the choice of each express delivery outlet cannot obtain a higher individual revenue through unilateral change given the choices of other express delivery outlets. Specifically:

[0021]

[0022] where U i is the individual revenue of the i-th express delivery outlet, S i is the alliance that the i-th express delivery outlet joins, is the alliance that the i-th express delivery outlet joins when reaching the Nash equilibrium, is the alliance that other express delivery outlets choose to join except the i-th express delivery outlet when reaching the Nash equilibrium.

[0023] Furthermore, the specific steps of the distributed independent decision-making in step S5 include the following steps:

[0024] S51: Initialize the allocation, satisfaction value, and iteration times of all express delivery outlets.

[0025] S52: Each express delivery outlet independently chooses to join the alliance that maximizes its individual utility.

[0026] S53: If the choice of the current express delivery outlet changes once, then change the allocation of this express delivery outlet and increase the iteration times by one.

[0027] S54: Set the satisfaction value of the current express delivery outlet to 1.

[0028] S55: Determine whether the allocation results of all express delivery outlets are the same. If they are the same, end the process; if not, traverse all express delivery outlets, change the allocation of each express delivery outlet to that of the express delivery outlet with the maximum number of iterations within the communication range, and set the satisfaction value of this express delivery outlet to 0.

[0029] S56: Determine whether the satisfaction values of all express delivery outlets are 1. If so, end the process; if not, re - execute the steps of S52 - S55.

[0030] Compared with the prior art, the present invention has the following beneficial effects:

[0031] (1) The present invention can make real - time and dynamic delivery decisions according to the busy degree of logistics distribution centers. Compared with the traditional delivery method relying on experience, this method can scientifically analyze the dynamic situation of each logistics distribution center, effectively avoid problems such as package backlog and delivery delay caused by blind decision - making, and improve the overall operation efficiency of the logistics system.

[0032] (2) By endowing each express delivery outlet with the right to independently select a logistics distribution center, the present invention enables it to make transfer decisions based on its own interests and the dynamic busy degree of the logistics distribution center. While ensuring the full utilization of the transfer capacity of the logistics distribution center, it significantly improves the average cargo sending efficiency of all express delivery outlets.

[0033] (3) The present invention adopts a distributed algorithm. Compared with the scheduling algorithm relying on global information and centralized decision - making, in the application scenario with a large number of express delivery outlets, it can significantly shorten the calculation time and has stronger dynamic adaptability and flexibility. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 is the implementation flowchart of a game - theoretic decision - making method for cargo transfer at express delivery outlets proposed by the present invention;

[0035] Figure 2 is the schematic diagram of the spatial positions and respective package weights of logistics distribution centers and express delivery outlets in the numerical example;

[0036] Figure 3 is the model schematic diagram of a game - theoretic decision - making method for cargo transfer at express delivery outlets proposed by the present invention;

[0037] Figure 4 is the final allocation result diagram of the game - theoretic decision - making for cargo transfer at express delivery outlets in the numerical example. DETAILED DESCRIPTION OF THE INVENTION

[0038] The method of the present invention will be described in detail below in conjunction with specific embodiments.

[0039] Embodiment 1

[0040] AsFigure 1 As shown in Figure 1 , a game decision-making method for cargo transfer in an express delivery network point includes the following steps:

[0041] Step S1: Each logistics distribution center within the region publicly discloses the current logistics parcel accumulation data and the average parcel processing speed data to all express delivery network points.

[0042] Step S2: Each express delivery network point obtains the distance information to different logistics distribution centers and calculates the distance satisfaction degree for different logistics distribution centers.

[0043] Step S3: Each express delivery network point combines the data publicly disclosed by the logistics distribution centers received in Step S1 and the distance data obtained in Step S2 to calculate the coalition revenue function and the individual revenue function.

[0044] Step S4: Each express delivery network point forms a coalition through a game model and optimizes the cooperation strategy through Nash equilibrium.

[0045] Step S5: Each express delivery network point makes distributed independent decisions based on the individual revenue calculated in Step S3. To promote decision consistency, local information communication is allowed among express delivery network points to coordinate the shipping selection.

[0046] Specifically, in Step S1, the current logistics parcel accumulation data and the average parcel processing speed data are collected in real time through the internal logistics management system of the logistics distribution center.

[0047] Suppose the current parcel accumulations of 5 logistics distribution centers within the region are 1000, 2000, 3000, 3000, and 5000 respectively, with the unit of kg.

[0048] The parcel accumulation data is the total weight of the parcels to be processed.

[0049] Suppose the weights of the parcels to be shipped and processed by 40 express delivery network points within the region are 300, 200, 200, 300, 100, 200, 300, 200, 100, 300, 200, 100, 300, 200, 100, 100, 500, 400, 200, 200, 300, 400, 200, 200, 100, 200, 300, 300, 100, 200, 200, 300, 400, 100, 300, 300, 300, 200, 300, 100 respectively, with the unit of kg.

[0050] The average parcel processing speed data is obtained by counting the weight of the parcels processed within a unit time and is quantitatively represented in kg / hour.

[0051] Suppose the average parcel processing speeds of the 5 logistics distribution centers in the area are 350, 550, 450, 600, and 550 respectively, with the unit of kg / h.

[0052] The distance satisfaction degree in step S2 is specifically:

[0053] where d ij is the distance between the obtained express delivery outlets and the logistics distribution centers, max(d ij ) is the maximum distance, and the distance satisfaction degree is the normalization processing of the distance.

[0054] The schematic diagram of the spatial positions of the logistics distribution centers and express delivery outlets and their respective parcel weights in the assumed 20km×20km area is as Figure 2 shown.

[0055] The specific form of the alliance revenue function in step S3 is:

[0056] where t j is the time for each logistics distribution center to process parcels, obtained by dividing the total parcel weight by the average parcel processing speed. Among them, the total parcel weight is obtained by adding the current logistics parcel accumulation amount obtained in step S1 and the parcel amounts of the express delivery outlets to be shipped to this logistics distribution center.

[0057] The specific form of the individual revenue function is:

[0058] The specific form of the game model in step S4 is where: P represents the set of players, corresponding to the set of all express delivery outlets in the area; S represents the set of strategies, corresponding to the set of alliances (i.e., logistics distribution centers) that each express delivery outlet can choose to join; U represents the set of individual revenues of each express delivery outlet; E represents the set of alliance revenues.

[0059] The purpose of forming an alliance is to achieve a stable allocation structure A = {A1, A2, …, A j , …, A n}, where A j represents the set of express delivery outlets that choose the logistics distribution center c j .

[0060] Define the allocation structure that reaches the Nash equilibrium as A * , that is, the choice of each express delivery outlet cannot obtain a higher individual revenue through unilateral change given the choices of other express delivery outlets. Specifically:

[0061]

[0062] where U iis the individual income of the i-th express delivery outlet, S i is the alliance that the i-th express delivery outlet joins, is the alliance that the i-th express delivery outlet joins when reaching Nash equilibrium, is the alliance that other express delivery outlets except the i-th express delivery outlet choose to join when reaching Nash equilibrium.

[0063] The specific steps of the distributed independent decision-making in step S5 include the following steps:

[0064] S51. Initialize the distribution, satisfaction value, and iteration times of all express delivery outlets.

[0065] S52. Each express delivery outlet independently chooses to join the alliance that maximizes its individual utility.

[0066] S53. If the choice of the current express delivery outlet changes once, then change the distribution of this express delivery outlet and increase the iteration times by one.

[0067] S54. Set the satisfaction value of the current express delivery outlet to 1.

[0068] S55. Judge whether the distribution results of all express delivery outlets are the same. If they are the same, end. If not, traverse all express delivery outlets, change the distribution of each express delivery outlet to the distribution of the express delivery outlet with the largest iteration times within the communication range, and set the satisfaction value of this express delivery outlet to 0.

[0069] S55. Judge whether the satisfaction values of all express delivery outlets are 1. If so, end. If not, re-execute steps S52 - S55.

[0070] The schematic diagram of the logistics delivery game decision-making based on dynamic busyness is as Figure 3 shown. This figure shows the communication and logistics relationships between three express delivery outlets (DS1, DS2, DS3) and two logistics distribution centers (DC1, DC2). The logistics distribution centers are divided into two categories: high load (DC) and low load (DC) according to the load situation, and information sharing is achieved by publicly releasing the load data. The express delivery outlets make their respective delivery choices through local communication to achieve efficient distribution of logistics.

[0071] The final distribution result of this example is as Figure 4 shown. This distribution can take into account both the busyness of the logistics distribution center and the distance between the express delivery outlet and the logistics distribution center to obtain a higher utility for the express delivery outlet.

Claims

1. A game decision-making method for cargo transfer in express delivery outlets based on dynamic busyness, characterized in that It includes the following steps: Step S1: Each logistics distribution center within the region discloses the current logistics package backlog data and the average package processing speed data to all express delivery outlets. Step S2: Each express delivery outlet obtains the distance information to different logistics distribution centers and calculates the distance satisfaction degree for different logistics distribution centers. Step S3: Each express delivery outlet calculates the alliance revenue function and the individual revenue function based on the public data of the logistics distribution center received in Step S1 and the distance data obtained in Step S2. Step S4: Each express delivery outlet forms an alliance through a game model and optimizes the cooperation strategy through Nash equilibrium. Step S5: Each express delivery outlet makes a distributed independent decision based on the individual revenue calculated in Step S3. To promote decision consistency, local information communication is allowed among express delivery outlets to coordinate the shipping choices.

2. The method for making a game decision on the goods transfer of an express delivery outlet based on dynamic busyness according to claim 1, wherein In Step S1, the current logistics package backlog data and the average package processing speed data are collected in real time through the internal logistics management system of the logistics distribution center.

3. A method for game decision-making on the transfer of express delivery network goods based on dynamic busyness degree according to claim 1 or 2, characterized in that The package backlog data is the total weight of the packages to be processed.

4. A method for game decision-making on goods transfer in an express delivery outlet based on dynamic busyness degree according to claim 3, characterized in that The average package processing speed data is obtained by counting the weight of the packages processed within a unit time and is quantitatively represented in kilograms per hour.

5. A method for game decision-making of goods transfer in an express delivery outlet based on dynamic busyness, as claimed in claim 4, wherein The distance satisfaction degree in the step S2 is specifically as follows: Among them, d ij is the distance between the obtained express delivery outlet and the logistics distribution center, and max(d ij ) is the maximum distance, and the distance satisfaction degree is the normalization processing of the distance.

6. A method for game decision-making of goods transfer in an express delivery outlet based on dynamic busyness according to claim 4 or 5, characterized in that The specific alliance revenue function in step S3 is as follows: where t j is the time for each logistics distribution center to process packages, which is obtained by dividing the total package weight by the average package processing speed. The total package weight is obtained by adding the current accumulated quantity of logistics packages obtained in step S1 to the quantity of packages at the express delivery outlets that will be shipped to this logistics distribution center; The specific individual income function is as follows:

7. A method for game decision-making on goods transfer in an express delivery outlet based on dynamic busyness degree according to claim 6, characterized in that, In step S4, the specific game model is as follows Where: P represents the set of players, corresponding to the set of all express delivery outlets in the region; S represents the set of strategies, corresponding to the set of coalitions that each express delivery outlet can choose to join; U represents the set of individual benefits of each express delivery outlet; E represents the set of coalition benefits. The purpose of forming an alliance is to achieve a stable allocation structure A = {A1, A2, …, A j , …, A n}, where A j represents the set of express delivery outlets that select the logistics distribution center c j ; Define the allocation structure that reaches the Nash equilibrium as A * , that is, the choice of each express delivery outlet cannot obtain a higher individual benefit through unilateral change given the choices of other express delivery outlets. Specifically: Among them, U i is the individual revenue of the i-th express delivery outlet, S i is the alliance that the i-th express delivery outlet joins, is the alliance that the i-th express delivery outlet joins when reaching Nash equilibrium, is the alliance that other express delivery outlets choose to join except the i-th express delivery outlet when reaching Nash equilibrium.

8. A method for game decision-making on goods transfer in an express delivery outlet based on dynamic busyness, as claimed in claim 7, wherein The specific steps of the distributed independent decision in Step S5 include the following steps: S51: Initialize the allocation, satisfaction value, and iteration times of all express delivery outlets. S52: Each express delivery outlet independently selects the alliance that maximizes its individual utility to join. S53: If the choice of the current express delivery outlet changes once, then change the allocation of this express delivery outlet and increase the iteration times by one. S54: Set the satisfaction value of the current express delivery outlet to 1. S55: Determine whether the allocation results of all express delivery outlets are the same. If they are the same, end; if not, traverse all express delivery outlets, change the allocation of each express delivery outlet to the allocation of the express delivery outlet with the largest iteration times within the communication range, and set the satisfaction value of this express delivery outlet to 0. S56: Determine whether the satisfaction values of all express delivery outlets are 1. If so, end; if not, re-execute the steps of S52 - S55.

Images