Full-link logistics system

Abstract

The invention discloses a full-link logistics system, which is used for forming a logistics plan from a plurality of suppliers to a production factory. Comprises: a processing device comprising: an order management module configured to generate an order and a supplier supply plan; the goods taking configuration module is configured to distribute circulation goods taking or zero-load goods taking according to the physical form and the historical transportation frequency of the goods; the path planning module is configured to generate a first planned logistics path for transporting the cargoes distributed to take the cargoes from the suppliers to a central cargo picking bin; a second planned logistics path is generated and is used for transporting the goods distributed to be taken circularly from the supplier to a central goods picking bin; generating a multimodal transport planning path which is used for transporting the goods in the central goods package warehouse to a supply management warehouse; and generating a terminal transportation planning path for transporting the goods in the supply management warehouse to a production factory. The method is suitable for a multi-supplier mode of a large-scale production enterprise, and the flexibility and stability of a supply chain are ensured.

Description

Technical Field

[0001] The present invention relates to the field of logistics technology, and particularly to an end-to-end logistics system. Background Art

[0002] In modern supply chain management, multiple suppliers rely on different third-party logistics companies for goods transportation. This decentralized logistics operation mode has led to increased complexity and costs. Since the operation processes and standards of each logistics company are different, suppliers need to coordinate multiple resources, increasing the management difficulty and operation costs. In addition, in the small-batch and high-frequency order mode, suppliers can only passively respond to orders and it is difficult to optimize the transportation efficiency. This mode is prone to low load capacity, that is, the amount of goods transported each time is insufficient, or it is difficult to deliver in a timely manner, affecting customer satisfaction.

[0003] To increase the load factor, suppliers may choose to backlog orders and wait until a certain quantity of goods accumulates before shipping them together. However, although this approach can reduce transportation costs to a certain extent, it will lead to low delivery efficiency and prolong the customer waiting time. In particular, for large manufacturing enterprises, since the production line needs to adjust the production rhythm according to orders in a timely manner, backlogging orders may also affect the production plan of the production plant. If the delivery is not timely, the production plan will be negatively affected, resulting in inventory backlog or production interruption, ultimately affecting the efficient operation of the entire supply chain. Summary of the Invention

[0004] In view of the above problems, the present application designs and provides an end-to-end logistics system for forming a logistics plan from multiple suppliers to a production plant; including: a processing device, which includes: an order management module configured to allow a production plant to generate an order and, after receiving confirmation from a supplier, generate a supplier supply plan based on the order; a pick-up configuration module configured to, based on the supplier supply plan, allocate a preset transportation mode for the goods waiting to be transported according to the physical form and historical transportation frequency of the goods waiting to be transported at the supplier; the preset transportation modes include milk run and less-than-truckload pick-up; a path planning module configured to: generate a first planned logistics path; the first planned logistics path is used to transport the goods waiting to be transported allocated for less-than-truckload pick-up from the supplier to the central cargo collection warehouse; generate a second planned logistics path; the second planned logistics path is used to transport the goods waiting to be transported allocated for milk run from the supplier to the central cargo collection warehouse; generate a multimodal transport planning path; the multimodal transport planning path is used to transport the goods in the central cargo collection warehouse to the supply management warehouse; generate a terminal transportation planning path; the terminal transportation planning path is used to transport the goods in the supply management warehouse to the production plant.

[0005] The full-link logistics system provided by this application is especially applicable to the multi-supplier mode of large-scale manufacturing enterprises, ensuring the flexibility and stability of the supply chain. The full-link logistics system is used to form a logistics planning solution from multiple suppliers to the production plant. Through the intelligently generated logistics planning solution, each link from the place where the goods depart to the production plant is efficiently connected. Supported by big data and with the central cargo collection warehouse as the physical core, the types of goods are reasonably divided to improve the efficiency of the supply chain.

[0006] After reading the specific implementation manners of the present invention in conjunction with the accompanying drawings, other features and advantages of the present invention will become clearer. Brief Description of the Drawings

[0007] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the accompanying drawings required to be used in the embodiments. Obviously, the accompanying drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other accompanying drawings can also be obtained based on these drawings.

[0008] Figure 1 It is a structural schematic block diagram of the full-link logistics system provided by this application;

[0009] Figure 2 It is a structural schematic block diagram of the full-link logistics system provided by this application;

[0010] Figure 3 It is a flowchart of the full-link logistics system provided by this application;

[0011] Figure 4 It is a structural schematic block diagram of the full-link logistics system provided by this application;

[0012] Figure 5 It is a flowchart of the full-link logistics system provided by this application;

[0013] Figure 6 It is a structural schematic block diagram of the full-link logistics system provided by this application;

[0014] Figure 7 It is a flowchart of the full-link logistics system provided by this application;

[0015] Figure 8 It is a structural schematic block diagram of the full-link logistics system provided by this application;

[0016] Figure 9 It is a structural schematic block diagram of the full-link logistics system provided by this application;

[0017] Figure 10 It is a structural schematic block diagram of the full-link logistics system provided by this application;

[0018] Figure 11 It is a schematic block diagram of the path of the full-link logistics system provided by this application;

[0019] In the figure: 1. Full-link logistics system; 10. Processing device; 101. Order management module; 102. Pickup configuration module; 103. Path planning module; 104. Storage module; 105. First site selection module; 106. Prediction module; 107. Second site selection module; 108. Inventory allocation module; 109. Display module. Detailed implementation manners

[0020] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0021] In the description of the present invention, it should be understood that the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. indicate the orientation or positional relationship based on the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore, should not be construed as a limitation to the present invention.

[0022] In the description of the present invention, it should be noted that unless otherwise clearly defined and limited, the terms "installation", "connection", and "electrical connection" should be understood in a broad sense. For example, it can be a fixed electrical connection, a detachable electrical connection, or an integral electrical connection. For those of ordinary skill in the art, the specific meanings of the above terms in the present invention can be understood according to specific circumstances. In the description of the embodiments, specific features, structures, materials, or characteristics can be combined in any one or more embodiments or examples in a suitable manner.

[0023] The terms "first" and "second" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance or implicitly indicating the quantity of the indicated technical features. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features.

[0024] In the description of the present invention, unless otherwise stated, the meaning of "a plurality" is two or more.

[0025] This application designs and provides a full-link logistics system 1, and the full-link logistics system 1 provided by this application is particularly applicable to the multi-supplier mode of large-scale manufacturing enterprises to ensure the flexibility and stability of the supply chain. The full-link logistics system 1 is used to form a logistics planning solution from multiple suppliers to the production plant, enabling efficient connection of each link from the origin of the goods to the production plant through the intelligently generated logistics planning solution. Supported by big data and with the central cargo collection warehouse as the physical core, the goods types are reasonably divided to improve the supply chain efficiency.

[0026] The full-link logistics system 1 includes a processing device 10. In this application, the processing device 10 can be implemented by a server in the logistics management control center or a combination of one or more high-performance computers. The logistics management control center is communicatively connected to the ERP systems of the suppliers, and can also be communicatively connected to transportation companies, vehicles, and intelligent terminals or servers set at the central cargo collection warehouse, the vendor managed inventory (VMI) warehouse, and the production plant to ensure data synchronization of all participating parties.

[0027] The processing device 10 is composed of a processor, a storage unit (including a volatile memory and a non-volatile memory), a display device, an operating device, a communication interface, a driving device, etc., and is connected through a bus. The processor can be a dedicated processor, a central processing unit (CPU), etc. The processor can access the storage unit and execute instructions or application programs stored therein to complete relevant functions. The display device is used to display various information. The operating device is used to receive user operation instructions, interact with the storage unit or storage medium, and process interrupt signals. In one or more embodiments of this application, the storage medium can be a CD-ROM, a floppy disk, an optical magneto-optical disk, etc., a medium that records data through optical, electrical, or magnetic means. The storage medium can also be a semiconductor memory such as a ROM, a flash memory, etc. that stores information through electrical means.

[0028] As Figure 1 shown, the processing device 10 includes an order management module 101, a pick-up configuration module 102, and a path planning module 103. Each of these modules can be implemented by the processor running a program.

[0029] The order management module 101 is configured to allow the production plant to generate an order, and based on the order, generate a supplier supply plan after receiving confirmation from the supplier. The supplier supply plan includes a list of goods waiting for transportation at the supplier.

[0030] The pickup configuration module 102 is configured to allocate a preset transportation mode for the goods waiting for transportation based on the supplier's supply plan, according to the physical form and historical transportation frequency of the goods waiting for transportation at the supplier. The preset transportation modes include milk run and less-than-truckload (LTL) pickup. Milk run refers to picking up goods from multiple suppliers according to a certain order and cycle. In this transportation mode, the pickup activities are regular and carried out along a predetermined route. LTL pickup refers to allocating transportation resources according to the goods supply demand of the supplier.

[0031] The route planning module 103 is configured to generate a first planned logistics route, a second planned logistics route, an intermodal transportation planned route, and a terminal transportation planned route; wherein, the first planned logistics route is used to transport the goods waiting for transportation allocated as LTL pickup from the supplier to the central consolidation warehouse; the second planned logistics route is used to transport the goods waiting for transportation allocated as milk run from the supplier to the central consolidation warehouse; the intermodal transportation planned route is used to transport the goods in the central consolidation warehouse to the supply management warehouse; and the terminal transportation planned route is used to transport the goods in the supply management warehouse to the production factory.

[0032] The full-link logistics system 1 provided by this application integrates the order management, pickup configuration, and route planning module 103, and forms a logistics planning scheme from the supplier to the production factory in an intelligent manner. Among them, the order management module 101 can effectively generate the supplier's supply plan, and the pickup configuration module 102 reasonably allocates the transportation mode based on the goods characteristics and historical transportation data, so as to reasonably allocate between milk run and LTL pickup and improve the transportation efficiency. The route planning module 103 can generate multi-layer logistics routes to ensure the efficient transportation of goods from the supplier directly to the central consolidation warehouse, and then from the central consolidation warehouse to the supply management warehouse and the production factory, simplify the logistics chain, integrate the process of LTL to the consolidation center and then to the city distribution center in the intermediate links into a direct way from the supplier to the central consolidation warehouse, optimize the short-haul transportation into an intermodal transportation planned route, and efficiently connect the production factory and the supply management warehouse where the production factory is located to support the smooth operation of large-scale manufacturing enterprises.

[0033] The route planning module 103 can be configured to execute the multi-source shortest path algorithm to generate the first planned logistics route; specifically, regard each central consolidation warehouse as a starting node, and then run the multi-source shortest path algorithm to calculate the shortest path from each central consolidation warehouse to the supplier.

[0034] The route planning module 103 can also be configured to execute other optimization algorithms to generate the first planned logistics route, such as the Floyd-Warshall algorithm, the greedy algorithm, etc. The route planning module 103 can also communicate with the logistics management system of a third party providing LTL transportation services to obtain a feasible first planned logistics route.

[0035] The path planning module 103 can be configured to execute mixed integer linear programming to generate a second planned logistics path. In the following, a preferred algorithm for generating the second planned logistics path will be introduced in detail.

[0036] The path planning module 103 can be configured to execute mixed integer linear programming, genetic algorithm, ant colony algorithm, etc. to generate an intermodal transport planning path.

[0037] Exemplarily, taking the execution of mixed integer linear programming to generate an intermodal transport planning path as an example, an intermodal transport planning path optimization model is established.

[0038] The objective of the intermodal transport planning path model can be to minimize the total transportation cost, and thus the objective function is:

[0039]

[0040] The constraint conditions of the intermodal transport planning path optimization model include:

[0041] To ensure that the demand of each supply management warehouse is met

[0042] To limit the transportation capacity of the central cargo collection warehouse;

[0043] Connect the binary decision variable and the quantity variable to ensure that there is only a cargo transportation quantity when a certain path is selected; that is, if the path (x ijk = 1) is selected, the corresponding cargo quantity can be transported; if a certain path (x ijk = 0) is not selected, the transportation quantity is 0;

[0044] The decision variable can only be 0 or 1;

[0045] In the intermodal transport planning path optimization model:

[0046] C i is the i-th central cargo collection warehouse, i ∈ I, and I is the set of central cargo collection warehouses; S j is the j-th supply management warehouse, j ∈ J, and J is the set of supply management warehouses; M k is the k-th transportation mode (such as road, waterway, railway, etc.), k ∈ K, and K is the set of transportation modes; c ijk is the transportation cost of using the transportation mode M k , from the central cargo collection warehouse C i to the supply management warehouse S j ; x ijkis a binary decision variable indicating whether to choose to use transportation mode M k from the central cargo collection warehouse C i to the supply management warehouse S j ; y ijk indicates using transportation mode M k and the quantity of goods transported from the central cargo collection warehouse C i to the supply management warehouse S j is a non - negative real number; Q i is the maximum transportation capacity of the central cargo collection warehouse; D i is the demand of the supply management warehouse S j ; M is a sufficiently large constant for constraint relations. The maximum transportation capacity of the central cargo collection warehouse refers to the quantity of goods that the central cargo collection warehouse can handle and transport within a certain period (such as daily, weekly, or monthly), expressed in unit quantity (such as the number of pieces, tons, or volume of goods), depending on factors such as the scale, facilities, equipment, and operation process of the warehouse.

[0047] Suppose there are 3 central cargo collection warehouses, C1, C2, and C3, 2 supply management warehouses, S1 and S2, and 2 transportation modes, M1 (road) and M2 (waterway). Through historical data, the transportation cost matrix from each central cargo collection warehouse to the supply management warehouse and the maximum transportation capacity of each central cargo collection warehouse can be further obtained. The demand of each supply management warehouse can also be further obtained. Substituting the above - mentioned data into the objective function and constraint relations, using a linear programming solver can solve the above - mentioned model to obtain the optimal transportation path and transportation quantity, and the path planning module 103 can generate an intermodal transportation planning path.

[0048] The path planning module 103 can be configured to execute a mixed - integer linear programming or a greedy algorithm to generate an end - to - end transportation planning path. Taking the greedy algorithm as an example, the goal is to transport goods from the supply management warehouse S j to the production factory F and reduce the transportation cost. The path planning module 103 pre - stores the transportation cost c j from each supply management warehouse S j to transport goods to the production factory and the maximum transportation capacity Q j , as well as the total quantity of goods required by the production factory D. Sort the supply management warehouses according to the transportation cost, traverse the sorted list of supply management warehouses (from low to high), and for each supply management warehouse S j , calculate the transportable quantity y j ; the transportable quantity y j is the minimum value of the maximum transportation capacity q j of the supply management warehouse S j and the total quantity of goods required by the current production factory D; when generating the transportable quantity y jAfter that, utilize the transportable quantity y j Update the demand D of the production plant (subtract the transportable quantity y from the demand D of the production plant j ), if the demand of the production plant has been met, end the algorithm and generate the terminal transport planning path.

[0049] As Figure 2 shown, the processing device 10 further includes a storage module 104, and the storage module 104 stores the historical transport frequency of goods and a pre-configured transport frequency threshold.

[0050] The pickup configuration module 102 is configured to execute multiple steps as shown in the figure to allocate a preset transport mode for the goods waiting for transport.

[0051] Step S101: Determine whether the historical transport frequency of the goods waiting for transport at the supplier exceeds the pre-configured transport frequency threshold.

[0052] Step S102: If so, allocate a milk run mode for the goods waiting for transport.

[0053] Step S103: If not, further determine whether the goods waiting for transport belong to the pre-configured list of bulk goods.

[0054] Step S104: If it belongs to the pre-configured list of bulk goods, allocate a milk run mode for the goods waiting for transport.

[0055] Step S105: If it does not belong to the pre-configured list of bulk goods, allocate a less-than-truckload pickup mode for the goods waiting for transport.

[0056] The list of bulk goods includes but is not limited to compressors, motors, electric heating elements, main control boards, heat exchangers, housing components, pipes, power adapters, sensors, switches, relays.

[0057] As shown in the figure, the processing device 10 further includes: a first site selection module 105, which is configured to generate alternative central cargo collection warehouses based on the distribution of suppliers; the central cargo collection warehouse is selected from the alternative central cargo collection warehouses.

[0058] The first site selection module 105 is configured to execute the following steps as shown in the figure to generate alternative central cargo collection warehouses.

[0059] Step S201: Obtain the geographical location points of the suppliers.

[0060] Exemplarily, it can be represented by longitude and latitude.

[0061] Step S202: Through the K-means clustering algorithm, randomly select K from the obtained geographical location points of the suppliers as the initial clustering centers.

[0062] In one or more embodiments of the present application, the stability of clustering can be improved by performing multiple random selections and comparing the results.

[0063] Step S203: Calculate the distance between each supplier's geographical location point and the current cluster center.

[0064] Since the supplier's geographic location point is represented by longitude and latitude coordinates, the longitude and latitude distance calculation formula is used to calculate the distance between suppliers A and B.

[0065]

[0066] Among them, latA and lngA are the geographical locations of supplier A, and latB and lngB are the geographical locations of supplier B. The spherical distance is calculated by the longitude and latitude calculation formula to obtain the distance in kilometers.

[0067] Step S204: For each supplier, determine the closest cluster center.

[0068] Step S205: assign the corresponding suppliers to the nearest cluster center to form K clusters.

[0069] Step S206: Calculate the average value of all supplier geographic location points in each cluster as a new cluster center.

[0070] Step S207: Repeat the above steps until the cluster center meets the preset stability condition or reaches the preset maximum number of iterations.

[0071] The preset stability conditions can be:

[0072] |F1-F2|≤ε

[0073] F1 and F2 are the measurement function values of two iterations, and ε is a very small number used to determine whether the change in the cluster center is small; if the change is less than this threshold, the algorithm stops.

[0074] Step S208: Calculate the cohesion, separation and silhouette coefficient.

[0075] Cohesion refers to the average distance from a sample point to all other points in the same cluster, reflecting the closeness of the data point within the cluster to which it belongs. Separation refers to the minimum value of the average distance from a data point to all other points in all other clusters, reflecting the degree of separation between the sample point and other clusters; the silhouette coefficient combines cohesion and separation. The calculation of cohesion, separation and silhouette coefficient all adopts mature algorithms in the prior art, which will not be repeated here.

[0076] Step S209: Calculate the sum of squared errors.

[0077] The sum of squared errors is used to measure the distance between sample points and their corresponding distance centers.

[0078] Step S210: Based on the silhouette coefficient and the sum of squared errors, select the optimal K value.

[0079] Among them, a silhouette coefficient close to 1 indicates that the sample points are very close within their clusters and are well separated from other clusters; close to 0 indicates that the sample points are located on the boundary of two clusters; close to -1 indicates that the sample points may be wrongly assigned to the wrong cluster; the smaller the sum of squared errors, the more obvious the clustering effect, the smaller the distance between the sample points and their clustering centers, and the stronger the cohesion of the clustering.

[0080] Step S211: Use the clustering center corresponding to the optimal K value as the alternative center receiving warehouse.

[0081] As shown in the figure, in some embodiments of the present application, the processing device 10 further includes: a prediction module 106, which is configured to: based on the historical demand data of the suppliers, generate the predicted demand of the suppliers through a pre-configured prediction model; based on the total predicted demand of the suppliers corresponding to the alternative center receiving warehouse, generate the predicted capacity and predicted demand of the alternative center receiving warehouse.

[0082] More specifically, when generating the predicted demand of the suppliers through a pre-configured prediction model based on the historical demand data of the suppliers, the prediction module 106 performs the following steps as shown in the figure.

[0083] Step S301: Obtain the historical demand data of the suppliers.

[0084] The historical demand data of the suppliers includes the historical supply volume of the suppliers.

[0085] The historical demand data of the suppliers may also include parameters such as the production capacity and delivery timeliness of the suppliers.

[0086] The historical demand data of the suppliers can be stored in the form of supplier feature vectors.

[0087] Step S302: Obtain the geographical location points of the suppliers.

[0088] The geographical location of the suppliers includes the longitude and latitude of the suppliers.

[0089] Step S303: Through the K-means clustering algorithm, perform K-means clustering on the suppliers based on the geographical location.

[0090] When using the weighted Euclidean distance in K-means clustering, different weights are assigned to each feature (including geographical location, historical supply volume, production capacity, delivery timeliness, etc.) so that features with larger weights (such as historical supply volume) have a greater impact on clustering, and features with smaller weights have a smaller impact on clustering.

[0091] The K-means clustering algorithm has been described in detail above and will not be elaborated here.

[0092] Step S304: Based on the administrative division to which the supplier belongs, divide the historical demand data of the supplier to obtain the administrative division historical demand volume data.

[0093] The administrative division can be the municipal administrative division. The administrative division historical demand volume data includes at least the historical supply volume of all suppliers in the city.

[0094] In some embodiments of the present application, urban economic parameters such as the economic growth rate, seasonal demand, consumer preferences, raw material prices, and labor costs of each city are collected in the full-link logistics system 1. Using the urban economic parameters and the administrative division historical demand volume, features at the administrative division level can be generated and stored in the form of administrative division feature vectors.

[0095] In some embodiments of the present application, the supplier feature vector and the administrative division feature vector can be further fused. When clustering, corresponding weights are also assigned to the features at the administrative division level, so as to efficiently divide the suppliers with similar features at the administrative division level (such as representing urban economic features) and supplier features into the same cluster.

[0096] Step S305: Use the grey prediction model to make a prediction based on the administrative division historical demand volume data to obtain the grey prediction model prediction output.

[0097] For example, use the grey model (GM(1,1)) for prediction. For each city, the grey model can be used to predict the historical supply volume. The grey model predicts the demand volume in an adaptive manner and is especially suitable for small sample data and situations with large uncertainties.

[0098] Step S306: Integrate the sample center point of the K-means clustering and the grey prediction model prediction output with the historical demand volume data of the supplier to obtain an extended sample point.

[0099] Step S307: Input the extended sample point into the BP neural network for training.

[0100] Step S308: Pre-configure the trained BP neural network as a prediction model.

[0101] Step S309: Use the prediction model to make a prediction to obtain the predicted demand quantity of the suppliers.

[0102] The training process of the BP neural network can adopt the mature methods in the prior art, which is not the focus of protection of this application and will not be elaborated here.

[0103] Exemplarily, the required warehouse area (capacity) can be calculated by setting the storage capacity per unit area according to the demand quantity predicted by the prediction model. The safety stock and buffer area, etc. can also be considered simultaneously to calculate the warehouse area (capacity).

[0104] The prediction module 106 first performs clustering through the historical data and geographical locations of the suppliers, then combines the supplementation of data by the grey model, and finally sends it into the BP neural network for the final prediction. In this way, the prediction ability of the BP neural network is enhanced, especially when the data is scarce or volatile. By fusing the economic characteristics of the city, not only can the demand quantity of each supplier be predicted more accurately, but also the prediction results of the clustering and the grey model can be made more representative and accurate.

[0105] In some embodiments of this application, the processing device 10 further includes:

[0106] A second site selection module 107, which is configured to select the central cargo collection warehouse from the alternative central cargo collection warehouses by using a preset selection optimization model, and minimize the total cost and maximize the supplier satisfaction when meeting the preset constraint conditions:

[0107] The selection optimization model is:

[0108]

[0109] Among them, F1 and F2 are objective functions. F1 aims to minimize the total cost, and F2 aims to maximize the supplier satisfaction.

[0110] The constraint conditions of the selection optimization model include:

[0111] To limit that at least one central cargo collection warehouse should be built;

[0112] The goods from the alternative central cargo collection warehouses to the central cargo collection warehouse cannot exceed the capacity of the central cargo collection warehouse;

[0113] The transportation volume of each alternative central cargo collection warehouse should meet the demand of the central cargo collection warehouse;

[0114] x ij ≥0, non - negative constraint, indicating that the transportation volume from the alternative central cargo collection warehouse to the central cargo collection warehouse is non - negative;

[0115] yi ∈ {0, 1}, 0-1 constraint, a value of 1 indicates that the i-th alternative central cargo collection warehouse is selected, otherwise it is 0

[0116]

[0117] In the selection optimization model: i is the i-th alternative central cargo collection warehouse, i = {0, 1, 2, …, m}; j is the j-th central cargo collection warehouse, j = {0, 1, 2, …, n}; c ij is the transportation cost from the i-th alternative central cargo collection warehouse to the j-th central cargo collection warehouse; x ij is the transportation volume from the i-th alternative central cargo collection warehouse to the j-th central cargo collection warehouse; D ij is the distance from the i-th alternative central cargo collection warehouse to the j-th central cargo collection warehouse; V ij is the vehicle speed from the i-th alternative central cargo collection warehouse to the j-th central cargo collection warehouse; V j is the capacity of the j-th central cargo collection warehouse; d j is the demand of the j-th central cargo collection warehouse; F j is the fixed cost of the j-th central cargo collection warehouse; T j is the operating cost of the j-th central cargo collection warehouse; y i is the decision variable of the selection optimization model; y i = 1 indicates the establishment of the i-th central cargo collection warehouse; k is the supplier number, k = {0, 1, 2, …, q}; L k represents the longest waiting time that supplier k is particularly satisfied with; U k represents the shortest waiting time that supplier k is extremely dissatisfied with; t jk represents the service time; λ t(jk) represents the time satisfaction function of supplier k for the central cargo collection warehouse j.

[0118] The genetic-ant colony algorithm is used to solve the problem to select multiple central cargo collection warehouses from the alternative central cargo collection warehouses.

[0119] Among them, the transportation cost can be obtained by analyzing past transportation records to obtain transportation cost data between different locations. The fixed cost and operating cost can be obtained from the financial statements of existing warehouses. The capacity and demand of the central cargo collection warehouse can be generated as a set value based on the output of the prediction module 106 and integrating the opinions of industry experts. The longest waiting time, shortest waiting time, delivery speed, etc. can all be obtained through market research and stored in the storage module 104 in the form of preset values.

[0120] In some embodiments of the present application, the path planning module 103 is configured to generate a second planned logistics path based on a preset path planning model.

[0121] The path planning model is:

[0122]

[0123] In the path planning model, Z is the objective function of the path planning model.

[0124] In the objective function of the above path planning model, the first part is the transportation cost of all paths, the second part is the penalty cost for exceeding the time window, and the third part is the vehicle activation cost.

[0125] The constraint conditions of the path planning model include:

[0126] The first group: path constraints:

[0127]

[0128] Indicates that the transport vehicle departs from the distribution center and finally returns to the central collection warehouse, ensuring that the vehicle departs from and returns to the distribution center.

[0129]

[0130] Indicates that each supplier is only assigned one pick-up path, ensuring that each supplier is only assigned one pick-up path. And i≠0, ensuring that all suppliers are visited.

[0131] And i≠0, indicating that the number of vehicles visiting the supplier is the same as the number of vehicles leaving. The transport vehicle leaves after completing the pick-up, ensuring equal in-and-out visits.

[0132] Eliminates the sub-circuit constraint.

[0133] The second group: vehicle constraints

[0134] Indicates that the total pick-up volume on a path is within the volume range of the transport vehicle, ensuring that the pick-up volume on the path does not exceed the vehicle's loading capacity, otherwise the pick-up cannot be completed.

[0135] Indicates the number of vehicles required during the cyclic pick-up process, calculates the number of vehicles required, and sets the number of vehicles not to exceed 10.

[0136] The third group: time constraints

[0137] Indicates that the time of cyclic pick-up is continuous, and there is no time interval between adjacent pick-up cycles, ensuring time continuity and avoiding time intervals.

[0138] The time of arrival is calculated as the time when the transportation vehicle arrives at supplier j is equal to the departure time from supplier i plus the driving time on the route between supplier i and supplier j.

[0139] It is to ensure the time window constraint, indicating that the vehicle cannot pick up goods from supplier i to supplier j before the specified time, ensuring the order of picking up goods.

[0140] The time window constraint ensures the calculation of the time of arrival.

[0141] 0 ≤ t i ≤ LT i - ET i , indicating that the transportation vehicle should complete the goods pickup within the specified time window, and the pickup time cannot exceed the length of the time window for accepting the pickup, ensuring that the pickup is within the time window.

[0142] The fourth group: inventory constraint

[0143] It indicates that the total quantity of goods picked up in the entire cycle cannot exceed the maximum inventory level of the central cargo collection warehouse.

[0144] In the path planning model, A is the set of suppliers, A = {i | i = 0, 1, 2,..., n}, where i = 0 represents the central cargo collection warehouse; H is the set of vehicles providing services, H = {1, 2,..., k}; L is the set of paths, L = {l | l = 0, 1, 2,..., m}; d ij is the distance from supplier i to supplier j; V i is the total supply volume of the i-th supplier on the same day, expressed in volume m 3 ; V max is the maximum loading capacity of the pickup vehicle; t arr,i is the time when the pickup vehicle arrives at supplier i; t lea,i is the time when the pickup vehicle leaves supplier i; t ij is the driving time of the pickup vehicle from supplier i to supplier j; t i is the pickup time of the pickup vehicle at supplier i; v is the average driving speed of the pickup vehicle; c is the unit transportation cost of the pickup vehicle; c0 is the one-time startup cost of the pickup vehicle; Q max is the maximum limit of the inventory; ET i is the earliest time when supplier i can accept the pickup; LT i is the latest time when supplier i can accept the pickup; is a decision variable, representing whether there is a path to complete from supplier i to supplier j, 1 for yes and 0 for no; is a decision variable, representing whether supplier i is picked up by path l, 1 for yes and 0 for no; q iis the quantity taken from supplier i in a single pick-up; f l is the pick-up frequency of route l; M is a sufficiently large constant.

[0145] The genetic algorithm is used to solve the route planning model to generate the second planned logistics route.

[0146] Data such as the supply volume of suppliers, time windows, the number, speed, loading capacity, transportation cost, start-up cost, distance between suppliers, transportation time, maximum inventory level, and the name, type, quantity, and packaging size of items in the route planning model can be obtained through telecommunication or based on market research and stored in the form of constants.

[0147] Furthermore, as Figure 9 shown, the processing device 10 further includes:

[0148] An inventory allocation module 108 configured to generate the allocation quantity from the central cargo collection warehouse to the target supply management warehouse based on the single-period inventory allocation model;

[0149] The single-period inventory allocation model is:

[0150]

[0151] Among them, the first part is the transportation cost, that is, the total transportation cost from the central cargo collection warehouse i to the supply management warehouse j; the second part is the storage cost, that is, the storage cost of the goods in the supply management warehouse j multiplied by the inventory quantity.

[0152] The constraint conditions of the single-period inventory allocation model include:

[0153] Ensure that the storage volume of each supply management warehouse j does not exceed its capacity.

[0154] Ensure that each supply management warehouse can meet the demand for at least one kind of goods.

[0155] Calculate the safety inventory to meet the service level.

[0156] Ensure that the inventory quantity can meet the demand and safety inventory.

[0157] Set the initial inventory quantity.

[0158] To ensure the relationship between the goods allocation quantity and the decision variables.

[0159] Decision variable x pijis a binary variable indicating whether the goods are allocated from the central cargo collection warehouse i to the supply management warehouse j.

[0160] In the single-period inventory allocation model:

[0161] P is the set of goods, P = {1, 2, …, p}; I is the set of central cargo collection warehouses, I = {1, 2, …, i}; J is the set of supply management warehouses, J = {1, 2, …, j}; T is the decision-making period; l is the service level; W j is the storage capacity of the supply management warehouse j; c ij is the unit transportation cost from the central cargo collection warehouse i to the supply management warehouse j; h pj is the unit storage cost of the goods p in the supply management warehouse j; β p is the unit shortage cost of the goods p; ESC pj is the shortage quantity of the goods p in the supply management warehouse j; d pj is the average daily demand of the goods p in the supply management warehouse j; I pj is the inventory level of the goods p in the supply management warehouse j; v pj is the initial inventory level of the goods p in the supply management warehouse j; SS pj is the safety inventory of the goods p in the supply management warehouse j; F pj is the normal distribution function of the goods p in the supply management warehouse j; K is a very large positive integer; x pij is a decision variable, a 0-1 variable, indicating whether the goods p are allocated from the central cargo collection warehouse i to the supply management warehouse j; y pij is a decision variable, the allocation quantity of the goods p from the central cargo collection warehouse i to the supply management warehouse j;

[0162] The artificial fish swarm algorithm is used to solve the inventory allocation model to generate the allocation quantity from the central cargo collection warehouse to the target supply management warehouse.

[0163] Furthermore, the inventory allocation module 108 is configured to generate the allocation quantity from the central cargo collection warehouse to the target supply management warehouse based on the multi-period inventory allocation model;

[0164] The multi-period inventory allocation model is:

[0165]

[0166] Among them, the first part is the transportation cost, that is, the total transportation cost from the central cargo collection warehouse i to the supply management warehouse j within the decision-making period t; the second part is the storage cost, that is, the storage cost of the goods in the supply management warehouse j within the decision-making period t multiplied by the inventory level.

[0167] The constraint conditions of the multi-period inventory allocation model include:

[0168] Ensure that the inventory level of each supply management warehouse j in each decision-making cycle t does not exceed its capacity.

[0169] Ensure that each supply management warehouse j can meet the demand for at least one item in each decision-making cycle t.

[0170] Calculate the safety stock level in decision-making cycle t to meet the service level. Ensure that the inventory level in decision-making cycle t can meet the demand and safety stock.

[0171] Update the inventory level in each decision-making cycle t.

[0172] Set the initial inventory level in decision-making cycle t.

[0173] Ensure the relationship between the allocation quantity of items in decision-making cycle t and the decision variables.

[0174] Decision variable x t pij is a binary variable indicating whether an item is allocated from central collection warehouse i to supply management warehouse j.

[0175] In the multi-period inventory allocation model: P is the set of items, P = {1, 2, …, p}; I is the set of central collection warehouses, I = {1, 2, …, i}; J is the set of supply management warehouses, J = {1, 2, …, j}; T is the set of decision-making cycles, T = {1, 2, …, t}; l is the service level; W j is the storage capacity of supply management warehouse j; c ij is the unit transportation cost from central collection warehouse i to supply management warehouse j; h pj is the unit storage cost of item p in supply management warehouse j; β p is the unit shortage cost of item p; ESC pj is the shortage quantity of item p in supply management warehouse j; d t pj is the average daily demand of item p in supply management warehouse j; I t pj is the inventory level of item p in supply management warehouse j; v t pj is the initial inventory level of item p in supply management warehouse j; SS t pj is the safety stock level of item p in supply management warehouse j; F pj is the normal distribution function of item p in supply management warehouse j; K is a very large positive integer; xt pij is a decision variable, a 0-1 variable, indicating whether the goods p are allocated from the central cargo collection warehouse i to the supply management warehouse j; y t pij is a decision variable, representing the quantity of goods p allocated from the central cargo collection warehouse i to the supply management warehouse j.

[0176] The artificial fish swarm algorithm is used to solve the inventory allocation model to generate the allocation quantity from the central cargo collection warehouse to the target supply management warehouse.

[0177] In the multi-period inventory allocation model, the transportation cost, storage cost, shortage cost, demand mean, safety inventory, and service level (expressed as a probability) can be obtained based on market analysis and customer surveys, and can also be predicted by an artificial intelligence model. The initial inventory and inventory can be obtained through communication from the management system of the supply management warehouse. The storage capacity of the supply management warehouse is a fixed value and can be pre-stored in the storage module 104 for ready access.

[0178] Furthermore, as Figure 10 shown, the processing device 10 further includes: a display module 109. The display module 109 is configured to display a human-computer interaction interface, a first planned logistics path, a second planned logistics path, a multimodal transportation planned path, and / or a terminal transportation planned path; the human-computer interaction interface is also used to allow the production factory to generate orders.

[0179] The above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, for those of ordinary skill in the art, it is still possible to modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions required to be protected by the present invention.

Claims

1. A full-link logistics system for forming a logistics plan from multiple suppliers to a production factory; It is characterized in that: Comprising: A processing device, which includes: An order management module configured to allow a production factory to generate an order and, after receiving confirmation from a supplier, generate a supplier supply plan based on the order; A pick-up configuration module configured to, based on the supplier supply plan, allocate a preset transportation mode for the goods waiting for transportation according to the physical form and historical transportation frequency of the goods waiting for transportation at the supplier; the preset transportation modes include milk run and less-than-truckload pick-up; A path planning module configured to: Generate a first planned logistics path; the first planned logistics path is used to transport the goods waiting for transportation allocated for less-than-truckload pick-up from the supplier to the central cargo collection warehouse; Generate a second planned logistics path; the second planned logistics path is used to transport the goods waiting for transportation allocated for milk run from the supplier to the central cargo collection warehouse; Generate an intermodal transportation planned path; the intermodal transportation planned path is used to transport the goods in the central cargo collection warehouse to the supply management warehouse; Generate a terminal transportation planned path; the terminal transportation planned path is used to transport the goods in the supply management warehouse to the production factory.

2. The full-link logistics system according to claim 1, wherein: The processing device further includes: A storage module in which the historical transportation frequency of goods and a pre-configured transportation frequency threshold are stored; The pick-up configuration module is configured to: Judge whether the historical transportation frequency of the goods waiting for transportation at the supplier exceeds the pre-configured transportation frequency threshold; If so, allocate a milk run mode for the goods waiting for transportation; If not, further judge whether the goods waiting for transportation belong to a pre-configured list of bulk goods; If they belong to the pre-configured list of bulk goods, allocate a milk run mode for the goods waiting for transportation; If they do not belong to the pre-configured list of bulk goods, allocate a less-than-truckload pick-up mode for the goods waiting for transportation.

3. The full-link logistics system according to claim 2, wherein: The processing device further includes: A first site selection module configured to generate alternative central cargo collection warehouses based on the distribution of suppliers; the central cargo collection warehouse is selected from the alternative central cargo collection warehouses; The first site selection module is configured to perform the following steps to generate alternative central cargo collection warehouses: Obtain the geographical location points of suppliers; Randomly select K of the obtained geographical location points of suppliers as the initial cluster centers through the K-means clustering algorithm; Calculate the distance between each geographical location point of the supplier and the current cluster center; For each supplier, determine the nearest cluster center; Allocate the corresponding supplier to the nearest cluster center to form multiple clusters; Calculate the average value of all geographical location points of suppliers in each cluster as the new cluster center; Repeat the above steps until the cluster centers meet the preset stability conditions or reach the preset maximum number of iterations; Calculate the cohesion, separation degree and silhouette coefficient; Calculate the sum of squared errors; Based on the silhouette coefficient and the sum of squared errors, select the optimal K value; Take the cluster center corresponding to the optimal K value as the alternative central cargo collection warehouse.

4. The full-link logistics system according to claim 3, characterized in that: The processing device further includes: A prediction module configured to: Based on the historical demand data of the suppliers, generate the predicted demand of the suppliers through a pre-configured prediction model; Based on the total predicted demand of the suppliers corresponding to the alternative central cargo collection warehouse, generate the predicted capacity and predicted demand of the alternative central cargo collection warehouse.

5. The full-link logistics system according to claim 4, characterized in that: When the prediction module generates the predicted demand of the suppliers based on the historical demand data of the suppliers through a pre-configured prediction model, the following steps are executed: Obtain the historical demand data of the suppliers; Obtain the geographical location points of the suppliers; Based on the geographical locations, perform K-means clustering on the suppliers through the K-means clustering algorithm; Based on the administrative divisions to which the suppliers belong, divide the historical demand data of the suppliers to obtain the administrative division historical demand data; Use the grey prediction model to perform prediction based on the administrative division historical demand data to obtain the prediction output of the grey prediction model; Integrate the sample center points of the K-means clustering and the prediction output of the grey prediction model with the historical demand data of the suppliers to obtain the augmented sample points; Input the augmented sample points into a BP neural network for training; Use the trained BP neural network as the pre-configured prediction model; Use the prediction model to perform prediction to obtain the predicted demand of the suppliers.

6. The full-link logistics system according to claim 5, characterized in that: The processing device further includes: A second site selection module configured to select the central cargo collection warehouse from the alternative central cargo collection warehouses using a preset selection optimization model, minimizing the total cost and maximizing the supplier satisfaction when meeting the preset constraint conditions: The selection optimization model is: where F1 and F2 are objective functions; The constraint conditions of the selection optimization model include: x ij ≥0 y i ∈{0,1} In the selection optimization model: i is the i-th alternative central cargo collection warehouse, i' = {0, 1, 2, …, m'}; j is the j-th central cargo collection warehouse, j' = {0, 1, 2, …, n'}; c ij is the transportation cost from the i-th alternative central cargo collection warehouse to the j-th central cargo collection warehouse; x ij is the transportation volume from the i-th alternative central cargo collection warehouse to the j-th central cargo collection warehouse; D ij is the distance from the i-th alternative central cargo collection warehouse to the j-th central cargo collection warehouse; V ij is the vehicle speed from the i-th alternative central cargo collection warehouse to the j-th central cargo collection warehouse; V j is the capacity of the j-th central cargo collection warehouse; d i is the demand of the j-th central cargo collection warehouse; F j is the fixed cost of the j-th central cargo collection warehouse; T j is the operating cost of the j-th central cargo collection warehouse; y i is the decision variable of the selection optimization model; y i = 1 indicates that the i-th central cargo collection warehouse is established; k is the supplier number, k = {0, 1, 2, …, q}; L k represents the longest waiting time that supplier k is particularly satisfied with; U k represents the shortest waiting time that supplier k is extremely dissatisfied with; t jk represents the service time; λ t(jk) represents the time satisfaction function of supplier k for the central cargo collection warehouse j; Solve using the genetic-ant colony algorithm to select multiple central cargo collection warehouses from the alternative central cargo collection warehouses.

7. The full-link logistics system according to claim 6, characterized in that: The path planning module is configured to generate the second planned logistics path based on a preset path planning model; the path planning model is: where Z is the objective function: The constraint conditions of the path planning model include: 0 ≤ t i ≤ LT i - ET i In the path planning model, A is the set of suppliers, A = {i | i = 0, 1, 2, …, n}, where i = 0 represents the central cargo collection warehouse; H is the set of vehicles providing services, H = {1, 2, …, k}; L is the set of paths, L = {l | l = 0, 1, 2, …, m}; d ij is the distance from supplier i to supplier j; V i is the total supply volume of the i-th supplier on the day, expressed in volume m 3 ; V max is the maximum loading capacity of the pick-up vehicle; t arr,i is the time when the pick-up vehicle arrives at supplier i; t lea,i is the time when the pick-up vehicle leaves supplier i; t ij is the driving time of the pick-up vehicle from supplier i to supplier j; t i is the pick-up time of the pick-up vehicle at supplier i; v is the average driving speed of the pick-up vehicle; c is the unit transportation cost of the pick-up vehicle; c0 is the one-time start-up cost of the pick-up vehicle; Q max is the maximum limit of the inventory; ET i is the earliest time when supplier i can accept pick-up; LT i is the latest time when supplier i can accept pick-up; is a decision variable, representing whether there is a path completed from supplier i to supplier j. If yes, it is 1; if no, it is 0; is a decision variable, representing whether supplier i is picked up by path l. If yes, it is 1; if no, it is 0; q i is the single pick-up volume from supplier i; f l is the pick-up frequency of path l; M is the total number of arranged paths; Solve the path planning model using the genetic algorithm to generate the second planned logistics path.

8. The full-link logistics system according to claim 7, characterized in that: The processing device further includes: An inventory allocation module configured to generate the allocation quantity from the central cargo collection warehouse to the target supply management warehouse based on a single-period inventory allocation model; The single-period inventory allocation model is: The constraint condition of the inventory allocation model is: In the single-period inventory allocation model: Let \(P\) be the set of goods, \(P = \{1, 2, \ldots, p\}\); \(I\) be the set of central cargo collection warehouses, \(I = \{1, 2, \ldots, i\}\); \(J\) be the set of supply management warehouses, \(J = \{1, 2, \ldots, j\}\); \(T\) be the decision-making period; \(l\) be the service level; \(W\) j is the storage capacity of supply management warehouse \(j\); \(c\) ij is the unit transportation cost from central cargo collection warehouse \(i\) to supply management warehouse \(j\); \(h\) pj is the unit storage cost of goods \(p\) in supply management warehouse \(j\); \(\beta\) p is the unit shortage cost of goods \(p\); \(ESC\) pj is the shortage quantity of goods \(p\) in supply management warehouse \(j\); \(d\) pj is the daily demand mean of goods \(p\) in supply management warehouse \(j\); \(I\) pj is the inventory level of goods \(p\) in supply management warehouse \(j\); \(v\) pj is the initial inventory level of goods \(p\) in supply management warehouse \(j\); \(SS\) pj is the safety stock of goods \(p\) in supply management warehouse \(j\); \(F\) pj is the normal distribution function of goods \(p\) in supply management warehouse \(j\); \(K\) is a very large positive integer; \(x\) pij is a decision variable, a 0-1 variable, indicating whether goods \(p\) are allocated from central cargo collection warehouse \(i\) to supply management warehouse \(j\); \(y\) pij is a decision variable, the allocation quantity of goods \(p\) from central cargo collection warehouse \(i\) to supply management warehouse \(j\); Solve the inventory allocation model using the artificial fish swarm algorithm to generate the allocation quantity from the central cargo collection warehouse to the target supply management warehouse.

9. The full-link logistics system according to claim 8, characterized in that: The processing device further includes: An inventory allocation module configured to generate the allocation quantity from the central cargo collection warehouse to the target supply management warehouse based on a multi-period inventory allocation model; The multi-period inventory allocation model is: The constraint conditions of the multi-period inventory allocation model are: In the multi-period inventory allocation model: P is the set of goods, P = {1, 2, …, p}; I is the set of central cargo collection warehouses, I = {1, 2, …, i}; J is the set of supply management warehouses, J = {1, 2, …, j}; T is the set of decision-making periods, T = {1, 2, …, t}; l is the service level; W j is the storage capacity of supply management warehouse j; c ij is the unit transportation cost from central cargo collection warehouse i to supply management warehouse j; g pj is the unit storage cost of goods p in supply management warehouse j; β p is the unit shortage cost of goods p; ESC pj is the shortage quantity of goods p in supply management warehouse j; d t pj is the daily demand mean of goods p in supply management warehouse j; I t pj is the inventory level of goods p in supply management warehouse j; v t pj is the initial inventory level of goods p in supply management warehouse j; SS t pj is the safety stock of goods p in supply management warehouse j; F pj is the normal distribution function of goods p in supply management warehouse j; K is a very large positive integer; x t pij is a decision variable, a 0-1 variable, indicating whether goods p are allocated from central cargo collection warehouse i to supply management warehouse j; y t pij is a decision variable, the allocation quantity of goods p from central cargo collection warehouse i to supply management warehouse j; An artificial fish swarm algorithm is used to solve the inventory allocation model to generate the allocation quantity from the central cargo collection warehouse to the target supply management warehouse.

10. The full-link logistics system according to claim 8, characterized in that: The processing device further includes: A display module configured to display a human-computer interaction interface, a first planned logistics path, a second planned logistics path, an intermodal transportation planned path, and / or a terminal transportation planned path; the human-computer interaction interface is used to allow a production factory to generate an order.

Images