Supply chain monitoring and optimization method and system

Abstract

The present invention relates to the technical field of supply chain management, and specifically discloses a supply chain monitoring and optimization method and system, wherein the method comprises the steps of: acquiring real-time Internet of Things data of a collection of to-be-shipped packages and a collection of transport vehicles within a warehouse; constructing a dynamic bipartite graph based on the real-time Internet of Things data; defining matching weights of edges connecting vertices based on the degree of matching of spatial indicators and / or temporal indicators; running a dynamic matching algorithm to generate a distribution plan for packages and vehicles based on the matching weights; and converting the distribution plan into scheduling instructions for a warehouse operation management system. The method acquires real-time Internet of Things data, constructs a dynamic bipartite graph, and runs a dynamic matching algorithm to generate a distribution plan for packages and vehicles, thereby achieving dynamic matching between packages and vehicles, effectively improving the distribution efficiency of packages and vehicles, and enhancing vehicle resource utilization, thereby achieving effective coordination between warehouse operations and transportation links, and effectively ensuring on-time delivery of orders.

Description

Technical Field

[0001] The present application relates to the technical field of supply chain management, and more specifically, to a supply chain monitoring and optimization method and system. Background Art

[0002] For large-scale commodity distribution companies, their core business is receiving, warehousing, and distributing massive quantities of goods from numerous suppliers to retailers and end customers. These orders are often high-frequency and small-batch, placing extremely high demands on supply chain responsiveness and flexibility.

[0003] To address these challenges, the company has built a complex supply chain network encompassing multiple regional distribution centers and urban delivery warehouses. To improve operational efficiency and visibility, the company has deployed IoT devices at key supply chain nodes. Warehousing centers utilize sensors to monitor inventory locations, automated equipment, and personnel status in real time. All vehicles in the transportation chain are equipped with onboard IoT terminals, which collect and transmit real-time data such as vehicle location, speed, and cargo load to a central data platform.

[0004] Despite widespread deployment of IoT devices, high-frequency, low-volume ordering models result in inefficient matching of packages and vehicles. Warehouse operations and transportation are closely linked but difficult to coordinate, and local issues can easily ripple through the entire process. Therefore, in complex distribution supply chain scenarios, integrating and leveraging IoT data from warehouse operations and transportation vehicles to achieve real-time perception and coordinated scheduling to improve resource utilization and ensure on-time delivery is a pressing technical challenge.

[0005] There is currently no effective technical solution to the above problems. Summary of the Invention

[0006] The purpose of this application is to provide a supply chain monitoring and optimization method and system to achieve dynamic matching of packages and vehicles, effectively improve the allocation efficiency of packages and vehicles and the utilization rate of vehicle resources, and achieve effective coordination between warehouse operations and transportation links.

[0007] In a first aspect, the present application provides a supply chain monitoring and optimization method, which is applied in a warehouse operation management system, and the method comprises the following steps:

[0008] S1. Obtain real-time IoT data on the collection of packages waiting to be shipped and the collection of transport vehicles within the warehouse;

[0009] S2. Construct a dynamic bipartite graph based on the real-time IoT data, where a vertex set on one side of the dynamic bipartite graph represents the set of packages to be shipped, and a vertex set on the other side of the dynamic bipartite graph represents the set of transport vehicles;

[0010] S3. Defining matching weights of edges connecting vertices corresponding to the set of packages to be shipped and vertices corresponding to the set of transport vehicles based on the matching degree of spatial indicators and / or temporal indicators;

[0011] S4. Running a dynamic matching algorithm to generate a package-to-vehicle allocation plan based on the matching weights, wherein the allocation plan is used to guide warehouse loading operations and vehicle scheduling;

[0012] S5. Convert the allocation plan into a dispatch instruction for the warehouse operation management system.

[0013] The supply chain monitoring and optimization method of this application obtains real-time Internet of Things data, constructs a dynamic bipartite graph and runs a dynamic matching algorithm to generate a package and vehicle allocation plan, realizes dynamic matching of packages and vehicles, can effectively improve the efficiency of package and vehicle allocation, and improve vehicle resource utilization, realizes effective coordination between warehouse operations and transportation links, and can effectively ensure that orders are delivered on time.

[0014] The supply chain monitoring and optimization method described above, wherein the real-time IoT data includes the destination, volume, weight, timeliness requirements of the package, as well as the location, speed, remaining available space, and estimated arrival time of the vehicle.

[0015] These specific data items provide the foundation for constructing a dynamic bipartite graph that reflects the potential matching relationships between packages and vehicles. Based on this data, we can define the matching weights of the edges connecting package vertices and vehicle vertices.

[0016] In the supply chain monitoring and optimization method, step S1 includes:

[0017] S11, scanning the label on the package to collect the unique identification code of each package in the set of packages to be shipped, and using the GPS positioning module to collect the location information of each vehicle in the set of transport vehicles;

[0018] S12. Upload the collected unique identification codes and vehicle location information to a cloud data center for cleaning, conversion, and integration to form a set of packages to be shipped that includes the destination, volume, weight, and timeliness requirements of the packages, and a set of transport vehicles that includes the vehicle location, speed, remaining available space, and estimated arrival time.

[0019] In this example, through cleaning, conversion, and integration in a cloud data center, the raw identification and location data is enriched and structured, forming a collection of packages to be shipped and a collection of transport vehicles that incorporate all necessary attributes. This provides an accurate, complete, and real-time data foundation for the subsequent construction of a dynamic bipartite graph and matching calculations, improving the effectiveness of the entire supply chain monitoring and optimization approach.

[0020] The supply chain monitoring and optimization method described above includes one or more of the following: the degree of match between the package's destination and the vehicle's scheduled route, the degree of match between the package's volume and the remaining available space on the vehicle, and the distance between the vehicle's current location and the warehouse where the package is located. The temporal indicator includes the degree of match between the package order's timeliness requirements and the vehicle's estimated delivery time.

[0021] In the supply chain monitoring and optimization method, step S3 includes:

[0022] S31. Obtain the destination coordinates of each package in the set of packages to be shipped, obtain multiple route coordinate points on a predetermined route of each vehicle in the set of transport vehicles, and calculate the minimum geographical distance between each of the destination coordinates and the multiple route coordinate points of each vehicle to obtain a distance set.

[0023] S32. Calculate the degree of spatial matching between the destination of each package and the vehicle's scheduled route using a Gaussian function based on the distance set;

[0024] S33. Obtain the latest delivery time for each package in the set of packages to be shipped, calculate the estimated delivery time for each package by each vehicle in the set of transport vehicles, and calculate the time margin between the latest delivery time and the estimated delivery time to obtain a time margin set.

[0025] S34. Calculate the time matching degree between each package and each vehicle using a sigmoid function according to the time margin set;

[0026] S35. Calculate the matching weight between each package and each vehicle by weighted fusion based on the spatial matching degree and the temporal matching degree.

[0027] In the supply chain monitoring and optimization method, step S35 includes:

[0028] S351: Based on a preset weight coefficient of the spatial matching degree and a preset weight coefficient of the temporal matching degree, a weighted average method is used to fuse the spatial matching degree and the temporal matching degree to obtain an initial matching weight;

[0029] S352: Acquire a vehicle type preference parameter, where the vehicle type preference parameter indicates the priority of different vehicle types;

[0030] S353. Adjust the initial matching weight according to the vehicle type preference parameter to obtain the final matching weight.

[0031] In the supply chain monitoring and optimization method, step S4 includes:

[0032] S41. Obtain warehouse loading resource restriction information from the warehouse operation management system;

[0033] S42. Acquire vehicle route information from the vehicle dispatching system, where the vehicle route information includes multiple unloading points of the vehicle and a time window requirement for each unloading point;

[0034] S43. Using the warehouse loading resource restriction information and the vehicle driving route information as constraints and combining them with the matching weights, a dynamic matching algorithm is used to generate an initial allocation plan for each package.

[0035] S44. Based on the initial allocation plan, evaluate the degree of matching between the arrival time of each vehicle at each unloading point and the time window requirement. If any unloading point does not meet the time window requirement, adjust the matching relationship between packages and vehicles in the initial allocation plan. Reallocate the packages that do not meet the time window requirement to other vehicles that meet the time window requirement, and generate an adjusted allocation plan.

[0036] S45. Evaluate the warehouse loading resource utilization rate based on the adjusted allocation plan. If the warehouse loading resource utilization rate is lower than a preset threshold, adjust the matching relationship between packages and vehicles in the adjusted allocation plan, and reallocate packages allocated to vehicles with low resource utilization rates to vehicles with high resource utilization rates, thereby generating a final allocation plan.

[0037] In the supply chain monitoring and optimization method, wherein the warehouse loading resource restriction information includes the number of loading platforms and the number of loading personnel, step S43 includes:

[0038] S431. Allocate multiple alternative vehicles for each package based on the matching weights, sort the multiple alternative vehicles based on the matching weights, and generate a set of alternative transportation solutions based on the sorted alternative vehicles.

[0039] S432. For each alternative transportation plan in the set of alternative transportation plans, based on the number of loading platforms and the number of loading personnel, the loading efficiency of the alternative transportation plan in the warehouse is evaluated using a pre-built loading efficiency evaluation model to obtain a loading efficiency evaluation result.

[0040] S433. For each alternative transportation plan, based on the multiple unloading points of the vehicle and the time window requirements of each unloading point, the time window satisfaction degree of the alternative transportation plan during the transportation process is evaluated based on a pre-established time window satisfaction degree evaluation model to obtain a time window satisfaction degree evaluation result;

[0041] S435. Based on the loading efficiency evaluation results and the time window satisfaction evaluation results, a weighted average method is used to calculate the comprehensive score of each alternative transportation plan, and the alternative transportation plan with the highest comprehensive score is used as the initial allocation plan for the corresponding package.

[0042] In the supply chain monitoring and optimization method, step S5 includes:

[0043] S51, parsing the allocation plan, extracting the unique identification code of the package to be loaded and the corresponding vehicle identification information;

[0044] S52. Obtaining physical attribute information of the package from a warehouse operation management system based on the unique identification code of the package;

[0045] S53. Acquire vehicle loading layout information from a warehouse operation management system based on the vehicle identification information;

[0046] S54: Generate a loading sequence instruction based on the physical attribute information of the package and the loading layout information of the vehicle;

[0047] S55: Convert the loading sequence instruction into a scheduling instruction executable by the warehouse operation management system.

[0048] In a second aspect, the present application further provides a supply chain monitoring and optimization system for use in a warehouse operation management system, the system comprising:

[0049] The data acquisition module is used to obtain real-time IoT data on the collection of packages to be shipped and the collection of transport vehicles in the warehouse;

[0050] A bipartite graph construction module, configured to construct a dynamic bipartite graph based on the real-time IoT data, wherein a vertex set on one side of the dynamic bipartite graph represents the set of packages to be shipped, and a vertex set on the other side of the dynamic bipartite graph represents the set of transport vehicles;

[0051] a weight definition module, configured to define a matching weight of an edge connecting a vertex corresponding to the set of packages to be shipped and a vertex corresponding to the set of transport vehicles based on a matching degree of a spatial index and / or a temporal index;

[0052] A matching algorithm module, configured to run a dynamic matching algorithm and generate a distribution plan for packages and vehicles based on the matching weights, wherein the distribution plan is used to guide warehouse loading operations and vehicle scheduling;

[0053] The instruction sending module is used to convert the allocation plan into a scheduling instruction of the warehouse operation management system.

[0054] The supply chain monitoring and optimization system of this application realizes the dynamic matching of packages and vehicles, which can effectively improve the allocation efficiency of packages and vehicles, and improve the utilization rate of vehicle resources, realize the effective coordination of warehouse operations and transportation links, and effectively ensure the timely delivery of orders.

[0055] From the above, it can be seen that the present application provides a supply chain monitoring and optimization method and system. Among them, the supply chain monitoring and optimization method of the present application obtains real-time Internet of Things data, constructs a dynamic bipartite graph and runs a dynamic matching algorithm to generate a package and vehicle allocation plan, realize dynamic matching of packages and vehicles, and can effectively improve the allocation efficiency of packages and vehicles, and improve vehicle resource utilization, realize effective coordination between warehouse operations and transportation links, and effectively ensure that orders are delivered on time. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 Flowchart of the supply chain monitoring and optimization method provided in an embodiment of the present application.

[0057] Figure 2 Schematic diagram of the dynamic bipartite graph constructed in step S2.

[0058] Figure 3 Schematic diagram of the modular structure of the supply chain monitoring and optimization system provided in an embodiment of the present application.

[0059] Reference numerals: 201, data acquisition module; 202, bipartite graph construction module; 203, weight definition module; 204, matching algorithm module; 205, instruction sending module. DETAILED DESCRIPTION

[0060] The technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, rather than all of the embodiments. The components of the embodiments of the present application generally described and shown in the drawings here can be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the application for protection, but merely represents the selected embodiments of the present application. Based on the embodiments of the present application, all other embodiments obtained by those skilled in the art without making creative work fall within the scope of protection of the present application.

[0061] It should be noted that similar reference numerals and letters represent similar items in the following drawings. Therefore, once an item is defined in one drawing, it does not need to be further defined or explained in subsequent drawings. At the same time, in the description of this application, the terms "first", "second", etc. are only used to distinguish the description and should not be understood as indicating or implying relative importance.

[0062] First, please refer to Figure 1-Figure 2 Some embodiments of the present application provide a supply chain monitoring and optimization method, which is applied in a warehouse operation management system. The method includes the following steps:

[0063] S1. Obtain real-time IoT data on the collection of packages waiting to be shipped and the collection of transport vehicles within the warehouse;

[0064] S2. Construct a dynamic bipartite graph based on real-time IoT data. The vertex set on one side of the dynamic bipartite graph represents the set of packages to be shipped, and the vertex set on the other side represents the set of transport vehicles.

[0065] S3. Defining matching weights of edges connecting vertices corresponding to the set of packages to be shipped and vertices corresponding to the set of transport vehicles based on the matching degree of the spatial index and / or the temporal index;

[0066] S4. Run a dynamic matching algorithm to generate a package-to-vehicle allocation plan based on the matching weights. The allocation plan is used to guide warehouse loading operations and vehicle scheduling.

[0067] S5. Convert the allocation plan into a scheduling instruction for the warehouse operation management system.

[0068] Specifically, the supply chain monitoring and optimization method of the embodiment of the present application solves the problems of efficient matching of packages and vehicles and coordination of warehouse loading and vehicle scheduling by integrating and utilizing real-time Internet of Things data.

[0069] More specifically, step S1 acquires real-time IoT data about packages to be shipped and transport vehicles, providing information about the system's current status and supporting subsequent dynamic decision-making. Real-time IoT data can be collected by sensors, vehicle-mounted terminals, and other devices and transmitted over the network to a data processing platform.

[0070] More specifically, step S2 constructs a dynamic bipartite graph based on real-time data, abstracts the matching relationship between packages and vehicles into a graph structure, and provides a model for algorithm processing. Figure 2 As shown, the vertices on one side of the dynamic bipartite graph represent packages, and the vertices on the other side represent vehicles. There is an undetermined edge between the package vertex and the vehicle vertex. Steps S3 and S4 are equivalent to determining the edge connecting the package vertex and the vehicle vertex to indicate that the package connected by the edge is transported by the corresponding vehicle. Figure 2 The edge connecting one of the packages (Package 1) is shown in .

[0071] More specifically, step S3 calculates the degree of match between the package and the vehicle based on spatial and temporal metrics. It also defines matching weights for the edges connecting package vertices and vehicle vertices, quantifying the degree of fit and providing evaluation criteria for the matching algorithm. Spatial metrics can include location, volume, and other factors, while temporal metrics can include the degree of match between timeliness requirements and estimated delivery times.

[0072] More specifically, step S4 runs a dynamic matching algorithm, using the matching weights calculated in step S3 to determine a specific allocation plan for packages and vehicles. This plan guides warehouses in loading packages and vehicles in transport scheduling. The dynamic matching algorithm can adjust matching results based on real-time data changes.

[0073] More specifically, step S5 converts the allocation plan generated in step S4 into dispatch instructions that can be recognized and executed by the warehouse operation management system, thus implementing the optimization plan in actual operations. The dispatch instructions may include information such as the loading order of packages and the designated vehicles.

[0074] More specifically, the supply chain monitoring and optimization method of the present embodiment solves the technical problem of efficiently matching packages and vehicles using real-time data and coordinating warehouse loading and vehicle scheduling. Its working principle is as follows:

[0075] First, in step S1, real-time IoT data on packages waiting to be shipped and transport vehicles in the warehouse is acquired. This data reflects the current status of packages (e.g., picked, waiting to be loaded) and the current status of vehicles (e.g., location, available capacity). This provides the system with the real-time information it needs to make decisions.

[0076] Next, in step S2, a dynamic bipartite graph is constructed based on the acquired real-time data. Nodes on one side of the graph represent packages to be shipped, and nodes on the other side represent transport vehicles. This graph structure intuitively represents the potential transport relationship between packages and vehicles, laying the foundation for subsequent matching calculations.

[0077] Then, in step S3, the degree of match between each package to be shipped and each transport vehicle is calculated based on predetermined spatial and temporal metrics. This match is then defined as the weight of the corresponding edge in the bipartite graph. For example, if the package's destination is close to the vehicle's route, or if the package's timeliness requirement aligns with the vehicle's estimated delivery time, the corresponding match weight is higher. This quantifies the quality of the match between the package and the vehicle.

[0078] Next, in step S4, a dynamic matching algorithm is run. This algorithm uses the matching weights calculated in step S3 as input, considers any constraints (e.g., vehicle capacity), and calculates an optimal or near-optimal plan for allocating packages to vehicles. This plan specifies which packages should be transported by which vehicle, as well as the possible loading order. This generates a plan to guide actual operations.

[0079] Finally, in step S5, the allocation plan generated in step S4 is converted into dispatch instructions that the warehouse operation management system can understand and execute. These instructions are sent to the warehouse operation management system to guide warehouse personnel or automated equipment in loading packages. They are also synchronized with the vehicle dispatch system to guide vehicle scheduling and transportation. This effectively implements the optimization plan at the operational level, improving the collaborative efficiency of warehouse loading and vehicle dispatching.

[0080] The supply chain monitoring and optimization method of the embodiment of the present application obtains real-time Internet of Things data, constructs a dynamic bipartite graph and runs a dynamic matching algorithm to generate a package and vehicle allocation plan, thereby achieving dynamic matching of packages and vehicles. This can effectively improve the efficiency of package and vehicle allocation and enhance vehicle resource utilization, thereby achieving effective coordination between warehouse operations and transportation links and effectively ensuring on-time delivery of orders.

[0081] In some preferred embodiments, the real-time IoT data includes multiple information such as the destination, volume, weight, and timeliness requirements of the package, as well as the location, speed, remaining available space, and estimated time of arrival of the vehicle.

[0082] Specifically, the destination information of the package is used to determine the geographic location of the package delivery so that it can be spatially matched with the vehicle's scheduled route. The volume and weight information of the package is used to assess the impact of the package on the vehicle's loading space and load-bearing capacity, and is associated with the vehicle's remaining available space and load-bearing capacity. The package's timeliness requirement sets the latest delivery time for the package, which is used to compare with the vehicle's estimated arrival time to evaluate the degree of time matching. The vehicle's position and speed are used to track the vehicle's status in real time, calculate the distance between the vehicle and the warehouse or package location, and estimate the arrival time. The vehicle's remaining available space information indicates the vehicle's current space capacity available for loading packages. The vehicle's estimated arrival time information is the estimated time of arrival at a specific location based on the vehicle's current status and route.

[0083] More specifically, these specific data items provide the foundational information for constructing a dynamic bipartite graph that reflects the potential matching relationships between packages and vehicles. Based on this data, matching weights can be defined for the edges connecting package vertices and vehicle vertices. For example, matching weights can be calculated using metrics such as the spatial proximity between the package's destination and the vehicle's scheduled route, the degree of compatibility between the package's volumetric weight and the vehicle's remaining space, and the time margin between the package's timeliness requirement and the vehicle's estimated arrival time.

[0084] More specifically, in addition, when running the dynamic matching algorithm, these data items can be used as inputs and constraints, such as considering the remaining available space constraints of the vehicle and the timeliness requirements of the package, so as to generate a more reasonable and effective package and vehicle allocation plan, thereby improving warehouse loading efficiency and transportation delivery timeliness.

[0085] In some preferred embodiments, step S1 includes:

[0086] S11, scanning the label on the package to collect the unique identification code of each package in the set of packages to be shipped, and using the GPS positioning module to collect the location information of each vehicle in the set of transport vehicles;

[0087] S12. Upload the collected unique identification code and vehicle location information to the cloud data center for cleaning, conversion, and integration to form a set of packages to be shipped that includes the destination, volume, weight, and timeliness requirements of the packages, as well as a set of transport vehicles that includes the vehicle location, speed, remaining available space, and estimated arrival time.

[0088] Specifically, the label on the package is preferably an RFID tag.

[0089] More specifically, scanning the RFID tag on the package captures a unique identification code, providing the basis for uniquely identifying each package to be shipped. Using the GPS positioning module to collect vehicle location information provides basic data on the vehicle's real-time spatial location. This enables preliminary data collection on two key resources: packages and vehicles.

[0090] More specifically, the raw unique identification codes and vehicle location information are uploaded to a cloud data center, enabling centralized data processing. Cleaning in the cloud data center removes errors or redundant data from the collection process, improving data quality.

[0091] More specifically, the cloud data center converts the cleaned data, unifying raw data from various sources and formats. The cloud data center integrates the converted data by correlating the collected basic identification and location information with existing package attributes (such as destination, volume, weight, and timeliness requirements) and vehicle attributes (such as speed, remaining available space, and estimated time of arrival) in the system. This creates a structured and complete set of packages to be shipped and transport vehicles. This integration ensures the accuracy and completeness of the data required for subsequent construction of the dynamic bipartite graph and calculation of matching weights, while also reducing the storage and analysis pressures on local data.

[0092] More specifically, through cleaning, conversion, and integration in a cloud-based data center, the raw data is enriched and structured, forming a collection of packages to be shipped and a collection of transport vehicles that includes all necessary attributes. This provides an accurate, complete, and real-time data foundation for the subsequent construction of a dynamic bipartite graph and matching calculations, improving the effectiveness of the entire supply chain monitoring and optimization approach.

[0093] In some preferred embodiments, the spatial indicators include one or more of the degree of match between the package's destination and the vehicle's scheduled route, the degree of match between the package's volume and the remaining available space on the vehicle, and the distance between the vehicle's current location and the warehouse where the package is located. The temporal indicators include the degree of match between the package order's timeliness requirements and the vehicle's estimated delivery time.

[0094] Specifically, the definition of these indicators enables the calculation of matching weights to reflect the actual spatial and temporal constraints between packages and vehicles. Specifically, for the problem of allocating packages and transport vehicles in warehouse operation management systems, traditional methods are vague in the content of spatial and temporal indicators when defining the matching weights between packages and vehicles, resulting in the allocation plan failing to fully consider constraints such as actual loading space, transportation routes and delivery timeliness, affecting operational efficiency and delivery punctuality. The supply chain monitoring and optimization method of the embodiment of the present application solves this problem by clearly defining spatial and temporal indicators. Among them, the degree of matching between the package destination and the vehicle's scheduled route is used to evaluate the degree of convenience of the vehicle's route for package delivery, the degree of matching between the package volume and the remaining available space on the vehicle is used to evaluate whether the vehicle has the ability to load the package, the distance between the vehicle's current location and the warehouse where the package is located is used to evaluate the convenience of the vehicle arriving at the warehouse, and the degree of matching between the package order timeliness requirements and the vehicle's estimated delivery time is used to evaluate whether the vehicle can complete the delivery task within the specified time.

[0095] More specifically, by calculating the degree of match between these specific indicators, we can generate matching weights that more accurately reflect the actual compatibility between packages and vehicles. Subsequent dynamic matching algorithms can then generate allocation plans based on these more precise weights. These plans can better account for loading space constraints, optimize transportation routing, and ensure on-time package delivery, thereby improving the efficiency and effectiveness of warehouse loading operations and vehicle scheduling.

[0096] In some preferred embodiments, step S3 comprises:

[0097] S31. Obtain the destination coordinates of each package in the set of packages to be shipped, obtain multiple route coordinate points on the scheduled route of each vehicle in the set of transport vehicles, calculate the minimum geographical distance between each destination coordinate and the multiple route coordinate points of each vehicle, and obtain a distance set;

[0098] S32. Based on the distance set, use a Gaussian function to calculate the spatial matching degree between the destination of each package and the vehicle's scheduled route;

[0099] S33. Obtain the latest delivery time for each package in the set of packages to be shipped, calculate the estimated delivery time for each package by each vehicle in the set of transport vehicles, calculate the time margin between the latest delivery time and the estimated delivery time, and obtain a time margin set;

[0100] S34, using a sigmoid function to calculate the time matching degree between each package and each vehicle according to the time margin set;

[0101] S35. Based on the spatial matching degree and the temporal matching degree, weighted fusion is performed to calculate the matching weight between each package and each vehicle.

[0102] Specifically, the matching weights defined in the above steps are the basis for the subsequent dynamic matching algorithm to generate an allocation plan.

[0103] More specifically, step S31 obtains the coordinates of the package destination and the coordinates of points on the vehicle's planned route to prepare data for spatial matching calculations. This process primarily calculates the minimum geographic distance between the package destination and the vehicle's route points, thereby quantifying the degree of spatial proximity.

[0104] More specifically, step S32 uses a Gaussian function to calculate the spatial matching degree based on the minimum geographical distance, and quantifies the proximity of the geographical locations into a numerical value. The smaller the distance, the higher the spatial matching degree.

[0105] More specifically, step S33 obtains the latest delivery time of the package and the estimated delivery time of the vehicle to the package, calculates the time margin, prepares data for time matching calculation, and thus quantifies time feasibility.

[0106] More specifically, step S34 uses a sigmoid function to calculate the time matching degree based on the time margin, and quantifies the satisfaction degree of the time window into a numerical value. The larger the time margin, the higher the time matching degree.

[0107] More specifically, step S35 performs weighted fusion on the calculated spatial matching degree and temporal matching degree to obtain a final matching weight.

[0108] More specifically, this matching weight comprehensively reflects the spatial proximity between the package's destination and the vehicle's route, as well as the temporal alignment between the package's timeliness requirements and the vehicle's delivery capabilities. This provides a quantitative basis for the subsequent dynamic matching algorithm, solving the problem of how to specifically quantify and integrate spatial and temporal metrics. This improves the accuracy of the matching weight and, in turn, optimizes the parcel-to-vehicle allocation scheme. Through these specific steps, this method provides an operational and quantifiable method for calculating matching weights, providing reliable input for the subsequent dynamic matching algorithm, thereby enabling the generation of more reasonable parcel-to-vehicle allocation schemes.

[0109] More specifically, in step S35, the matching weight can be simply calculated based on the following formula: matching weight = α * spatial matching degree + β * temporal matching degree; wherein α and β are the weight coefficients of spatial matching degree and temporal matching degree respectively, and the sum of the two is preferably 1. α and β can be set according to actual usage requirements, such as 0.8 and 0.2.

[0110] In some preferred embodiments, step S35 includes:

[0111] S351: Based on a preset weight coefficient of the spatial matching degree and a preset weight coefficient of the temporal matching degree, a weighted average method is used to fuse the spatial matching degree and the temporal matching degree to obtain an initial matching weight;

[0112] S352: Acquire a vehicle type preference parameter, where the vehicle type preference parameter indicates the priority of different vehicle types;

[0113] S353. Adjust the initial matching weight according to the vehicle type preference parameter to obtain the final matching weight.

[0114] Specifically, the above process first uses a weighted average method to calculate the spatial match between the package's destination and the vehicle's scheduled route, as well as the temporal match between the package's order timeliness requirement and the vehicle's estimated delivery time, based on preset spatial and temporal match weights. This method then calculates the initial matching weight between the package and the vehicle. Subsequently, these initial matching weights are adjusted based on the acquired vehicle type preference parameters. This final matching weight not only takes into account the geographic and temporal compatibility between the package and the vehicle, but also the suitability or importance of the vehicle type for the transport task.

[0115] More specifically, the vehicle type preference parameter sets the priority for different vehicle types (e.g., cold chain trucks, regular trucks, small vehicles, large vehicles) during the matching process. The vehicle type preference parameter can be a numeric value, with a higher value indicating a higher priority for that vehicle type when matching a specific package.

[0116] More specifically, the process of adjusting the initial matching weight may be to multiply the initial matching weight by the corresponding vehicle type preference parameter, or to adjust the weight using other functional relationships.

[0117] More specifically, the final matching weight can comprehensively reflect the spatial proximity, temporal matching, and vehicle type preference, and is used in the subsequent dynamic matching algorithm.

[0118] More specifically, technical contribution: By introducing vehicle type preference parameters, the advantages of different types of vehicles can be better utilized. For example, cold chain vehicles can be used to transport fresh products, or small vehicles can be used for terminal distribution, thereby improving transportation efficiency and reducing transportation costs. In this embodiment, the process of obtaining vehicle type preference parameters in step S352 can be to obtain the vehicle type preference parameters of each vehicle based on the package type. This adjustment makes the final matching weight more comprehensive. It reflects the matching degree between the package and the vehicle, comprehensively considers the space, time and vehicle type factors, provides more accurate input for the subsequent dynamic matching algorithm, and helps to generate a more reasonable and efficient package and vehicle allocation plan, thereby improving the utilization efficiency of different types of transportation resources.

[0119] In some preferred embodiments, step S4 includes:

[0120] S41. Obtain warehouse loading resource restriction information from the warehouse operation management system;

[0121] S42. Obtain vehicle route information from the vehicle dispatching system, where the vehicle route information includes multiple unloading points of the vehicle and a time window requirement for each unloading point;

[0122] S43. Using the warehouse loading resource restriction information and vehicle driving route information as constraints and combining them with matching weights, a dynamic matching algorithm is used to generate an initial allocation plan for each package.

[0123] S44. Based on the initial allocation plan, evaluate the degree of matching between the arrival time of each vehicle at each unloading point and the time window requirement. If any unloading point does not meet the time window requirement, adjust the matching relationship between packages and vehicles in the initial allocation plan. Reassign the packages that do not meet the time window requirement to other vehicles that meet the time window requirement, and generate an adjusted allocation plan.

[0124] S45. Evaluate the warehouse loading resource utilization rate based on the adjusted allocation plan. If the warehouse loading resource utilization rate is lower than a preset threshold, adjust the matching relationship between packages and vehicles in the adjusted allocation plan. Reallocate packages allocated to vehicles with low resource utilization rates to other vehicles with high resource utilization rates to improve the warehouse loading resource utilization rate and generate a final allocation plan.

[0125] Specifically, step S41 obtains the physical resource constraints of the warehouse for loading operations, such as the number of loading platforms and the number of loading personnel.

[0126] More specifically, step S42 obtains the time constraints that must be met during the vehicle transportation process, including the unloading point and the corresponding time window requirements.

[0127] More specifically, step S43 combines the actual operational constraints with the matching weights, runs a dynamic matching algorithm, and generates a preliminary package-to-vehicle allocation plan, incorporating the actual operational constraints into the matching process. The initial allocation plan includes which vehicle should transport each package, as well as the order and quantity of vehicle loading.

[0128] More specifically, step S44 verifies and adjusts the time window of the initial allocation plan, evaluating whether the vehicles can meet the preset time window requirements at each unloading point when transporting according to the initial plan. If there is any non-compliance, the matching relationship between the package and the vehicle is adjusted, and the packages that cause time window conflicts are reallocated to other vehicles that can meet the time window to ensure transportation timeliness.

[0129] More specifically, step S45 evaluates and optimizes the warehouse resource utilization of the adjusted allocation plan, and evaluates the utilization efficiency of the warehouse loading resources when executing the current allocation plan. If the resource utilization is lower than the preset level, the matching relationship between the package and the vehicle is further adjusted, and the packages allocated to the vehicles with lower resource utilization are transferred to the vehicles with higher resource utilization, thereby improving the overall utilization efficiency of the warehouse loading resources.

[0130] More specifically, the above approach optimizes the allocation of packages and vehicles by incorporating practical operational constraints. First, resource constraints such as the number of loading bays and personnel are obtained from the warehouse operation management system. Simultaneously, vehicle route information, including each unloading point and its corresponding time window requirements, is obtained from the vehicle scheduling system. These resource constraints and time window requirements are used as constraints and, along with package-vehicle matching weights calculated based on spatial and temporal metrics, are input into a dynamic matching algorithm to generate a preliminary package allocation plan. This initial plan specifies which vehicle will transport each package, as well as the order and number of vehicles to be loaded. Next, this initial plan is checked for time window compliance. For each vehicle involved in the plan, its estimated arrival time at each unloading point is calculated and compared to the preset time window requirements. If a vehicle is found to be unable to meet the time window requirement at a particular unloading point, the packages causing the time window violation are identified and removed from the current vehicle. These packages are then reassigned to other vehicles that meet the time window constraints, generating a time-window-adjusted allocation plan. Finally, the adjusted allocation plan is evaluated for warehouse loading resource utilization. The solution calculates the utilization of resources such as warehouse loading platforms and personnel during the execution of the plan. If overall resource utilization falls below a preset efficiency threshold, indicating room for optimization in resource allocation, the matching of packages and vehicles is further adjusted. For example, packages assigned to vehicles with lower resource utilization rates may be transferred to vehicles with higher resource utilization rates to better utilize warehouse resources. This ultimately generates an allocation plan to guide actual loading and scheduling. By introducing, verifying, and optimizing layer-by-layer constraints, the solution generates an allocation plan that comprehensively considers matching, warehouse operating capacity, and transportation time constraints, addressing resource constraints and timeliness in actual operations.

[0131] In some preferred embodiments, the warehouse loading resource restriction information includes the number of loading platforms and the number of loading personnel, and step S43 includes:

[0132] S431. Allocate multiple alternative vehicles for each package based on the matching weights, sort the multiple alternative vehicles based on the matching weights, and generate a set of alternative transportation solutions based on the sorted alternative vehicles.

[0133] S432. For each alternative transportation plan in the set of alternative transportation plans, based on the number of loading platforms and the number of loading personnel, the loading efficiency of the alternative transportation plan in the warehouse is evaluated using a pre-built loading efficiency evaluation model to obtain a loading efficiency evaluation result.

[0134] S433. For each alternative transportation plan, based on the multiple unloading points of the vehicle and the time window requirements of each unloading point, the time window satisfaction degree of the alternative transportation plan during the transportation process is evaluated based on a pre-established time window satisfaction degree evaluation model to obtain a time window satisfaction degree evaluation result;

[0135] S435. Based on the loading efficiency evaluation results and the time window satisfaction evaluation results, a weighted average method is used to calculate the comprehensive score of each alternative transportation plan, and the alternative transportation plan with the highest comprehensive score is used as the initial allocation plan for the corresponding package.

[0136] Specifically, this approach addresses the problem of failing to effectively balance warehouse loading efficiency and vehicle transportation time window requirements when generating initial allocation plans. First, multiple alternative vehicles are generated for each package based on their matching weights in reverse order, and a set of alternative transportation plans is formed based on these alternative vehicles. This provides a wide range of possible package and vehicle allocation combinations, laying the foundation for subsequent evaluation and selection. Next, for each alternative transportation plan in the set, a pre-built loading efficiency evaluation model is used to evaluate the warehouse's loading efficiency, taking into account the number of loading platforms and the number of loading personnel. This quantifies the plan's utilization of warehouse resources.

[0137] At the same time, for each alternative transportation option, a pre-built time window satisfaction evaluation model is used, combined with the vehicle's unloading point and time window requirements, to evaluate the extent to which the option meets the time window requirements during transportation. This quantifies the feasibility and timeliness of the option in the transportation process. Finally, a weighted average method is used to calculate the overall score of each alternative transportation option using the obtained loading efficiency evaluation results and time window satisfaction evaluation results. This provides a mechanism to balance the two potentially conflicting objectives of warehouse loading efficiency and vehicle time window satisfaction. Ultimately, the alternative transportation option with the highest overall score is selected as the initial allocation plan for the corresponding package. Through the generation, evaluation, and selection process, this method is able to generate an initial allocation plan that strikes a balance between warehouse loading efficiency and vehicle time window satisfaction, thus addressing the lack of balance problem in existing methods.

[0138] In some preferred embodiments, step S5 includes:

[0139] S51. Analyze the allocation plan and extract the unique identification code of the package to be loaded and the corresponding vehicle identification information;

[0140] S52. Obtaining physical attribute information of the package from the warehouse operation management system based on the unique identification code of the package;

[0141] S53. Obtaining vehicle loading layout information from the warehouse operation management system based on the vehicle identification information;

[0142] S54: Generate a loading sequence instruction based on the physical attribute information of the package and the loading layout information of the vehicle;

[0143] S55. Convert the loading sequence instruction into a scheduling instruction executable by the warehouse operation management system.

[0144] Specifically, the above processing method refines step S5 and solves the problem of failing to consider loading constraints and operation details when generating scheduling instructions based only on the allocation plan.

[0145] More specifically, step S51 receives an allocation plan and extracts the unique identification code of the package to be loaded and the corresponding vehicle identification information. Step S52 obtains the physical attribute information of the package from the warehouse operation management system based on the extracted unique identification code of the package. The physical attribute information may include the size, weight, and stacking requirements of the package. Step S53 obtains the loading layout information of each vehicle from the warehouse operation management system based on the extracted vehicle identification information. The loading layout information may include the available loading space and weight limit of the vehicle. Then, step S54 generates a loading sequence instruction based on the physical attribute information of the package and the vehicle loading layout information. The loading sequence instruction indicates the loading position of each package in the vehicle in combination with the vehicle's weight limit and available loading space. Finally, step S55 converts the loading sequence instruction into a data structure in a format that can be recognized and executed by the warehouse operation management system.

[0146] More specifically, in step S54, the loading sequence instruction is used to guide warehouse personnel or automated equipment to load packages into the vehicle in a determined order, and determine the loading position of each package in the vehicle taking into account the size, weight, stacking requirements of the package, as well as the load-bearing limit and available space of the vehicle, such as placing heavier packages or packages with stacking requirements at the bottom.

[0147] More specifically, the allocation plan is refined into loading instructions that take into account physical constraints and operation sequences, ensuring the execution of the allocation plan at the warehouse end and improving loading efficiency and accuracy.

[0148] In some preferred embodiments, the physical attribute information includes the size, weight and stacking requirements of the package; the loading layout information includes the available loading space and weight limit of the vehicle; the loading sequence instruction indicates the specific loading position of each package in the vehicle in combination with the vehicle's weight limit and available loading space. If there is a stacking requirement for the package, the stackable package will be placed at the bottom first, and the heavier package will be placed at the bottom.

[0149] Specifically, physical attribute information includes the package's dimensions, weight, and stacking requirements. This information is captured through scanning or input and describes the package's physical characteristics. Loading layout information includes the vehicle's available loading space and weight restrictions. This information is captured through the warehouse operations management system or preset parameters and describes the vehicle's physical constraints. Generating loading sequence instructions involves a decision-making process that determines where each package should be placed inside the vehicle and in what order, based on the package's physical attributes and the vehicle's loading layout.

[0150] More specifically, the decision logic considers the overall weight limit of the vehicle and the load-bearing capacity of different areas to ensure that the total weight of the load does not exceed the limit. At the same time, the decision logic considers the three-dimensional available space inside the vehicle and plans the placement of packages to maximize space utilization. In addition, the decision logic follows specific loading rules. For example, for packages marked as stackable, the system will prioritize them in a lower position so that other packages can be stacked on top of them. For heavier packages, the system will also prioritize them at the bottom of the vehicle or close to the center of gravity of the vehicle to improve transportation stability. The combination of these rules and information enables the generated loading sequence instructions to guide the actual loading operation and ensure the safety and efficiency of loading.

[0151] More specifically, after receiving the package-to-vehicle allocation plan, the system parses it to extract the package identifiers and corresponding vehicle identifiers. Using the package identifiers, the system queries the warehouse database to retrieve physical attribute data for each package to be loaded, including dimensions, weight, and stacking requirements. Using the vehicle identifiers, the system retrieves the corresponding vehicle's three-dimensional model of available loading space and loading layout data, including weight limits for each area, from the warehouse operations management system. The system then runs a loading planning algorithm based on the obtained package physical attribute information and vehicle loading layout information. This algorithm aims to maximize space utilization and adhere to weight limits, while also considering package stacking requirements and weight distribution rules. The algorithm iteratively finds the appropriate loading location within the vehicle for each package and determines the optimal loading sequence. For example, the algorithm assesses whether placing a package in a certain location will result in partial or overall overweight, whether it will exceed the available space, and whether it complies with stacking rules (e.g., stackable packages should be placed at the bottom, heavy packages at the bottom). By evaluating the feasibility and advantages of placing different packages in different locations, the algorithm ultimately generates detailed loading sequence instructions. The instructions clearly indicate where each package should be loaded onto the vehicle, as well as the order in which packages should be loaded. This technical solution solves the problem of translating abstract allocation plans into specific loading instructions that warehouse workers can understand and execute, improving the accuracy and efficiency of loading operations.

[0152] Second, please refer to Figure 3 Some embodiments of the present application further provide a supply chain monitoring and optimization system, which is applied in a warehouse operation management system. The system includes:

[0153] Data acquisition module 201, used to obtain real-time IoT data of the set of packages to be shipped and the set of transport vehicles in the warehouse;

[0154] A bipartite graph construction module 202 is used to construct a dynamic bipartite graph based on real-time IoT data, wherein a vertex set on one side of the dynamic bipartite graph represents a set of packages to be shipped, and a vertex set on the other side represents a set of transport vehicles;

[0155] A weight definition module 203 is configured to define a matching weight of an edge connecting a vertex corresponding to the set of packages to be shipped and a vertex corresponding to the set of transport vehicles based on the matching degree of the spatial index and / or the temporal index;

[0156] Matching algorithm module 204, for running a dynamic matching algorithm to generate a parcel-to-vehicle allocation plan based on matching weights. The allocation plan is used to guide warehouse loading operations and vehicle scheduling;

[0157] The instruction sending module 205 is used to convert the allocation plan into a scheduling instruction of the warehouse operation management system.

[0158] The supply chain monitoring and optimization system of the embodiment of the present application obtains real-time Internet of Things data, constructs a dynamic bipartite graph and runs a dynamic matching algorithm to generate a package and vehicle allocation plan, thereby achieving dynamic matching of packages and vehicles. This can effectively improve the efficiency of package and vehicle allocation and improve vehicle resource utilization, thereby achieving effective coordination between warehouse operations and transportation links and effectively ensuring that orders are delivered on time.

[0159] In addition, the units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed across multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0160] Furthermore, the functional modules in each embodiment of the present application can be integrated together to form an independent part, or each module can exist independently, or two or more modules can be integrated to form an independent part.

[0161] In this document, relational terms such as first and second, etc. are used merely to distinguish one entity or operation from another entity or operation, but do not necessarily require or imply any actual relationship or order between these entities or operations.

[0162] The above description is merely an embodiment of the present application and is not intended to limit the scope of protection of the present application. For those skilled in the art, various modifications and variations of the present application are possible. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application shall be included in the scope of protection of the present application.

Claims

1. A supply chain monitoring and optimization method, applied in a warehouse operation management system, characterized in that: The method comprises the following steps: S1. Obtain real-time IoT data on the collection of packages waiting to be shipped and the collection of transport vehicles within the warehouse; S2. Construct a dynamic bipartite graph based on the real-time IoT data, where a vertex set on one side of the dynamic bipartite graph represents the set of packages to be shipped, and a vertex set on the other side of the dynamic bipartite graph represents the set of transport vehicles; S3. Defining matching weights of edges connecting vertices corresponding to the set of packages to be shipped and vertices corresponding to the set of transport vehicles based on the matching degree of spatial indicators and / or temporal indicators; S4. Running a dynamic matching algorithm to generate a package-to-vehicle allocation plan based on the matching weights, wherein the allocation plan is used to guide warehouse loading operations and vehicle scheduling; S5. Converting the allocation plan into a dispatch instruction for a warehouse operation management system; Step S3 includes: S31. Obtain the destination coordinates of each package in the set of packages to be shipped, obtain multiple route coordinate points on a predetermined route of each vehicle in the set of transport vehicles, and calculate the minimum geographical distance between each of the destination coordinates and the multiple route coordinate points of each vehicle to obtain a distance set. S32. Calculate the degree of spatial matching between the destination of each package and the vehicle's scheduled route using a Gaussian function based on the distance set; S33. Obtain the latest delivery time for each package in the set of packages to be shipped, calculate the estimated delivery time for each package by each vehicle in the set of transport vehicles, and calculate the time margin between the latest delivery time and the estimated delivery time to obtain a time margin set. S34. Calculate the time matching degree between each package and each vehicle using a sigmoid function according to the time margin set; S35. Calculate the matching weight between each package and each vehicle by weighted fusion based on the spatial matching degree and the temporal matching degree; Step S4 includes: S41. Obtain warehouse loading resource restriction information from the warehouse operation management system; S42. Acquire vehicle route information from the vehicle dispatching system, where the vehicle route information includes multiple unloading points of the vehicle and a time window requirement for each unloading point; S43. Using the warehouse loading resource restriction information and the vehicle driving route information as constraints and combining them with the matching weights, a dynamic matching algorithm is used to generate an initial allocation plan for each package. S44. Based on the initial allocation plan, evaluate the degree of matching between the arrival time of each vehicle at each unloading point and the time window requirement. If any unloading point does not meet the time window requirement, adjust the matching relationship between packages and vehicles in the initial allocation plan. Reallocate the packages that do not meet the time window requirement to other vehicles that meet the time window requirement, and generate an adjusted allocation plan. S45. Evaluate the warehouse loading resource utilization rate based on the adjusted allocation plan. If the warehouse loading resource utilization rate is lower than a preset threshold, adjust the matching relationship between packages and vehicles in the adjusted allocation plan, and reallocate packages allocated to vehicles with low resource utilization rates to vehicles with high resource utilization rates, thereby generating a final allocation plan.

2. A supply chain monitoring and optimization method according to claim 1, characterized in that: The real-time IoT data includes the destination, volume, weight, timeliness requirements of the package, as well as the location, speed, remaining available space, and estimated arrival time of the vehicle.

3. A supply chain monitoring and optimization method according to claim 2, characterized in that: Step S1 includes: S11, scanning the label on the package to collect the unique identification code of each package in the set of packages to be shipped, and using the GPS positioning module to collect the location information of each vehicle in the set of transport vehicles; S12. Upload the collected unique identification codes and vehicle location information to a cloud data center for cleaning, conversion, and integration to form a set of packages to be shipped that includes the destination, volume, weight, and timeliness requirements of the packages, and a set of transport vehicles that includes the vehicle location, speed, remaining available space, and estimated arrival time.

4. A supply chain monitoring and optimization method according to claim 1, characterized in that: The spatial indicators include one or more of the following: the degree of match between the package's destination and the vehicle's scheduled route, the degree of match between the package's volume and the remaining available space on the vehicle, and the distance between the vehicle's current location and the warehouse where the package is located. The temporal indicators include the degree of match between the package order's timeliness requirements and the vehicle's estimated delivery time.

5. A supply chain monitoring and optimization method according to claim 1, characterized in that: Step S35 includes: S351: Based on a preset weight coefficient of the spatial matching degree and a preset weight coefficient of the temporal matching degree, a weighted average method is used to fuse the spatial matching degree and the temporal matching degree to obtain an initial matching weight; S352: Acquire a vehicle type preference parameter, where the vehicle type preference parameter indicates the priority of different vehicle types; S353. Adjust the initial matching weight according to the vehicle type preference parameter to obtain the final matching weight.

6. A supply chain monitoring and optimization method according to claim 1, characterized in that: The warehouse loading resource restriction information includes the number of loading platforms and the number of loading personnel. Step S43 includes: S431. Allocate multiple alternative vehicles for each package based on the matching weights, sort the multiple alternative vehicles based on the matching weights, and generate a set of alternative transportation solutions based on the sorted alternative vehicles. S432. For each alternative transportation plan in the set of alternative transportation plans, based on the number of loading platforms and the number of loading personnel, the loading efficiency of the alternative transportation plan in the warehouse is evaluated using a pre-built loading efficiency evaluation model to obtain a loading efficiency evaluation result. S433. For each alternative transportation plan, based on the multiple unloading points of the vehicle and the time window requirements of each unloading point, the time window satisfaction degree of the alternative transportation plan during the transportation process is evaluated based on a pre-established time window satisfaction degree evaluation model to obtain a time window satisfaction degree evaluation result; S435. Based on the loading efficiency evaluation results and the time window satisfaction evaluation results, a weighted average method is used to calculate the comprehensive score of each alternative transportation plan, and the alternative transportation plan with the highest comprehensive score is used as the initial allocation plan for the corresponding package.

7. A supply chain monitoring and optimization method according to claim 1, characterized in that: Step S5 includes: S51, parsing the allocation plan, extracting the unique identification code of the package to be loaded and the corresponding vehicle identification information; S52. Obtaining physical attribute information of the package from a warehouse operation management system based on the unique identification code of the package; S53. Acquire vehicle loading layout information from a warehouse operation management system based on the vehicle identification information; S54: Generate a loading sequence instruction based on the physical attribute information of the package and the loading layout information of the vehicle; S55: Convert the loading sequence instruction into a scheduling instruction executable by the warehouse operation management system.

8. A supply chain monitoring and optimization system, applied in a warehouse operation management system, characterized in that: The system comprises: The data acquisition module is used to obtain real-time IoT data on the collection of packages to be shipped and the collection of transport vehicles in the warehouse; A bipartite graph construction module, configured to construct a dynamic bipartite graph based on the real-time IoT data, wherein a vertex set on one side of the dynamic bipartite graph represents the set of packages to be shipped, and a vertex set on the other side of the dynamic bipartite graph represents the set of transport vehicles; a weight definition module, configured to define a matching weight of an edge connecting a vertex corresponding to the set of packages to be shipped and a vertex corresponding to the set of transport vehicles based on a matching degree of a spatial index and / or a temporal index; A matching algorithm module, configured to run a dynamic matching algorithm and generate a distribution plan for packages and vehicles based on the matching weights, wherein the distribution plan is used to guide warehouse loading operations and vehicle scheduling; An instruction sending module, used for converting the allocation plan into a scheduling instruction for the warehouse operation management system; The step of defining the matching weight of the edge connecting the vertices corresponding to the set of packages to be shipped and the vertices corresponding to the set of transport vehicles based on the matching degree of the spatial index and / or the temporal index includes: S31. Obtain the destination coordinates of each package in the set of packages to be shipped, obtain multiple route coordinate points on a predetermined route of each vehicle in the set of transport vehicles, and calculate the minimum geographical distance between each of the destination coordinates and the multiple route coordinate points of each vehicle to obtain a distance set. S32. Calculate the degree of spatial matching between the destination of each package and the vehicle's scheduled route using a Gaussian function based on the distance set; S33. Obtain the latest delivery time for each package in the set of packages to be shipped, calculate the estimated delivery time for each package by each vehicle in the set of transport vehicles, and calculate the time margin between the latest delivery time and the estimated delivery time to obtain a time margin set. S34. Calculate the time matching degree between each package and each vehicle using a sigmoid function according to the time margin set; S35. Calculate the matching weight between each package and each vehicle by weighted fusion based on the spatial matching degree and the temporal matching degree; The steps of running the dynamic matching algorithm and generating a distribution plan for packages and vehicles based on the matching weights, wherein the distribution plan is used to guide warehouse loading operations and vehicle scheduling, include: S41. Obtain warehouse loading resource restriction information from the warehouse operation management system; S42. Acquire vehicle route information from the vehicle dispatching system, where the vehicle route information includes multiple unloading points of the vehicle and a time window requirement for each unloading point; S43. Using the warehouse loading resource restriction information and the vehicle driving route information as constraints and combining them with the matching weights, a dynamic matching algorithm is used to generate an initial allocation plan for each package. S44. Based on the initial allocation plan, evaluate the degree of matching between the arrival time of each vehicle at each unloading point and the time window requirement. If any unloading point does not meet the time window requirement, adjust the matching relationship between packages and vehicles in the initial allocation plan. Reallocate the packages that do not meet the time window requirement to other vehicles that meet the time window requirement, and generate an adjusted allocation plan. S45. Evaluate the warehouse loading resource utilization rate based on the adjusted allocation plan. If the warehouse loading resource utilization rate is lower than a preset threshold, adjust the matching relationship between packages and vehicles in the adjusted allocation plan, and reallocate packages allocated to vehicles with low resource utilization rates to vehicles with high resource utilization rates, thereby generating a final allocation plan.

Images