Topology quantification method, device, equipment, storage medium and program product

Abstract

The application relates to a topology quantization method, device, equipment, storage medium and program product. The method comprises the following steps: firstly, acquiring multi-source heterogeneous data associated with a target industrial chain; then, generating at least one candidate topology corresponding to the target industrial chain based on the multi-source heterogeneous data; then, for each candidate topology, performing simulation operation on the candidate topology under a preset disturbance scene, determining an optimal scheduling strategy of each candidate topology; and finally, performing simulation operation on each candidate topology according to the optimal scheduling strategy of each candidate topology, and determining a quantization result of each candidate topology. The method can improve the accuracy of simulation evaluation of the target industrial chain.

Description

Technical Field

[0001] This application relates to the field of supply chain management technology, and in particular to a method, apparatus, equipment, storage medium, and program product for quantifying topology. Background Technology

[0002] Against the backdrop of increasing globalization, complexity, and uncertainty, industrial and supply chains, as the foundation of modern economic activity, face unprecedented challenges to their stability and resilience. Designing industrial chain topologies that combine high efficiency and resilience has become a core challenge for enterprises and industry managers. Furthermore, industrial chain simulation is necessary during the design and evaluation phases to verify its performance.

[0003] However, current supply chain simulation assessments suffer from low accuracy. Summary of the Invention

[0004] Therefore, it is necessary to provide a topology quantification method, device, equipment, storage medium, and program product that can improve the accuracy of supply chain simulation and evaluation, addressing the aforementioned technical problems.

[0005] In a first aspect, this application provides a topology quantization method, which includes:

[0006] Acquire multi-source heterogeneous data related to the target industry chain;

[0007] Based on multi-source heterogeneous data, at least one candidate topology corresponding to the target industry chain is generated;

[0008] For each candidate topology, simulations are performed on the candidate topology under a preset disturbance scenario to determine the optimal scheduling strategy for each candidate topology.

[0009] Simulations were performed on each candidate topology based on its optimal scheduling strategy to determine the quantification results of each candidate topology.

[0010] In one embodiment, based on multi-source heterogeneous data, at least one candidate topology corresponding to the target industry chain is generated, including:

[0011] Based on multi-source heterogeneous data, a baseline topology for the target industry chain is generated;

[0012] Edit and modify the baseline topology to generate at least one candidate topology.

[0013] In one embodiment, the candidate topologies are simulated under a preset disturbance scenario to determine the optimal scheduling strategy for each candidate topology, including:

[0014] The candidate topology is iteratively run under a preset perturbation scenario to obtain the policy function corresponding to the result of each run.

[0015] The policy function that reaches convergence is determined as the optimal dynamic scheduling policy for the candidate topology.

[0016] In one embodiment, each candidate topology is simulated and run according to its optimal scheduling strategy to determine the quantization result of each candidate topology, including:

[0017] For each candidate topology, the candidate topology is simulated under a preset disturbance scenario according to the optimal dynamic scheduling strategy corresponding to the candidate topology, and at least one key performance indicator of the candidate topology is obtained.

[0018] The comprehensive strategic fitness score of the candidate topology is determined based on each key performance indicator and its corresponding weight.

[0019] The quantitative results for each candidate topology are determined based on its comprehensive strategic fitness score.

[0020] In one embodiment, the method further includes:

[0021] Generate a topology resilience radar chart based on the key performance indicators corresponding to each candidate topology;

[0022] Based on the system state snapshot data of each candidate topology at different times during the simulation process, generate system state evolution animations corresponding to each candidate topology;

[0023] Displays a topology resilience radar chart and an animation of system state evolution.

[0024] In one embodiment, acquiring multi-source heterogeneous data related to the target industry chain includes:

[0025] Identify at least one target system that is associated with the target industry chain;

[0026] Obtain multi-source heterogeneous data from at least one target system.

[0027] Secondly, this application also provides a topology quantization device, which includes:

[0028] The acquisition module is used to acquire multi-source heterogeneous data related to the target industry chain;

[0029] The generation module is used to generate at least one candidate topology corresponding to the target industry chain based on multi-source heterogeneous data;

[0030] The first determining module is used to perform simulation operation on each candidate topology under a preset disturbance scenario to determine the optimal scheduling strategy for each candidate topology.

[0031] The second determination module is used to simulate and run each candidate topology according to the optimal scheduling strategy of each candidate topology, and determine the quantization result of each candidate topology.

[0032] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of the topology quantization method described in any of the first aspects above.

[0033] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the topology quantization method described in any of the first aspects above.

[0034] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of the topology quantization method described in any of the first aspects above.

[0035] The aforementioned topology quantification method, apparatus, equipment, storage medium, and program product first acquire multi-source heterogeneous data associated with the target industry chain. Then, based on the multi-source heterogeneous data, at least one candidate topology corresponding to the target industry chain is generated. Next, for each candidate topology, simulation operation is performed under a preset disturbance scenario to determine the optimal scheduling strategy for each candidate topology. Finally, simulation operation is performed on each candidate topology according to its optimal scheduling strategy to determine the quantification result of each candidate topology. In this way, the disturbance resistance capability of each candidate topology in the target industry chain is simulated, and the quantification results of each candidate topology operating under its respective optimal scheduling strategy are obtained, improving the accuracy of the simulation evaluation of the target industry chain. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 This is a flowchart illustrating a topology quantization method in one embodiment;

[0038] Figure 2 This is a flowchart illustrating the candidate topology generation steps in one embodiment;

[0039] Figure 3This is a flowchart illustrating the steps for generating the optimal dynamic scheduling strategy for a candidate topology in one embodiment.

[0040] Figure 4 This is a flowchart illustrating the steps for generating quantization results for each candidate topology in one embodiment.

[0041] Figure 5 This is a flowchart illustrating the topology quantization results in one embodiment;

[0042] Figure 6 This is a schematic diagram of the structure of a supply chain simulation optimization system in one embodiment;

[0043] Figure 7 This is a flowchart illustrating the topology optimization method in another embodiment;

[0044] Figure 8 This is a structural block diagram of a topology optimization device in one embodiment;

[0045] Figure 9 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0046] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0047] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0048] As the foundation of modern economic activity, the stability and operational efficiency of industrial chains directly impact regional and even global economies. Currently, globalized industrial chains face an increasingly complex external environment, with frequent unpredictable disturbances such as extreme weather events and public health crises, placing higher demands on their stable operation. Designing a strategic industrial chain structure that can ensure economic benefits while possessing sufficient resilience to external shocks has become a crucial technical issue. Furthermore, industrial chain simulation is necessary during the design and evaluation phases to verify its performance.

[0049] In related technologies, the design and evaluation of supply chain topologies typically employ techniques such as mathematical programming or discrete event simulation. Mathematical programming methods can optimize the structure for a single or a few objectives (such as minimizing cost) under specific constraints. Discrete event simulation is used to simulate the dynamic behavior of the supply chain under specific events (such as demand fluctuations or supply disruptions) given a topology and operating rules, in order to test its performance. However, existing technologies have significant limitations in application. For example, static optimization methods based on mathematical programming are usually based on deterministic assumptions about the future environment, making it difficult to effectively handle frequent and unpredictable disturbances in the real world. While discrete event simulation methods can simulate dynamic processes, the common practice when evaluating different candidate topologies is to pre-set the same set of fixed operating rules for all different topologies. Therefore, current supply chain simulation evaluations suffer from low accuracy.

[0050] In view of this, this application provides a topology quantification method. This method enables the simulation of the anti-disturbance capability of each candidate topology in the target industry chain, obtains the quantification results of each candidate topology operating under its own optimal scheduling strategy, and improves the accuracy of simulation evaluation of the target industry chain.

[0051] In one exemplary embodiment, such as Figure 1 As shown, a topology quantization method is provided. This embodiment illustrates the application of this method to a terminal. It is understood that this method can also be applied to a server, and further to a system including both a terminal and a server, and implemented through interaction between the terminal and the server. In this embodiment, the method includes the following steps:

[0052] Step 101: Obtain multi-source heterogeneous data related to the target industry chain.

[0053] Multi-source heterogeneous data refers to data from multiple sources with different structures. This includes data acquired from various upstream and downstream links of the target industry chain, different industry sectors, and various data platforms. The goal is to collect and integrate diverse and complex data information to provide a unified data foundation for simulating the target industry chain. Optionally, multi-source heterogeneous data may include the status data of various physical entities in the target industry chain, business process data within the target industry chain, and external environmental data affecting the operation of the target industry chain.

[0054] Step 102: Based on multi-source heterogeneous data, generate at least one candidate topology corresponding to the target industry chain.

[0055] For the multi-source heterogeneous data obtained above, it is necessary to process and integrate the multi-source heterogeneous data. First, data cleaning is performed on each source data to remove outliers and fill in missing values. Second, format normalization is performed to convert data in different formats (such as JSON, XML, CSV) into a unified structured data table. Finally, timestamp alignment is performed to synchronize all data records in the time dimension with a unified time base, forming a complete snapshot dataset that is time-series and includes the physical entities of the target industry chain, business processes, and external environment status.

[0056] Optionally, for the processed dataset, a baseline topology is defined to characterize the target industry chain. Based on the baseline topology, at least one candidate topology with a different structure from the baseline topology can be generated.

[0057] Step 103: For each candidate topology, perform simulation operation on the candidate topology under a preset disturbance scenario to determine the optimal scheduling strategy for each candidate topology.

[0058] The preset disturbance scenario can include at least one disturbance event, used to simulate external shocks or internal changes during simulation to test the dynamic response capability of the candidate topology. Optionally, the preset disturbance scenario can be defined as an event vector, which is a set with the structure S={e1,e2,…en}, where n is the total number of events, e represents a specific disturbance event, and each disturbance event e is defined as a triple with the structure e=(content, time, parameter). The content defines the nature or type of the disturbance event. For example, the content of a disturbance event could be a supplier's raw material supply disruption, a transportation route interruption, or a sudden increase in market demand for a product. The time defines the precise occurrence time of the disturbance event on the simulation timeline. For example, day 30 indicates that the disturbance event is triggered on day 30 of the simulation. The parameter is a set containing one or more key-value pairs used to quantify the disturbance event. For example, for an event where a supplier cuts off the supply of raw materials, the parameters can be set as {interruption duration: 15 days, impact percentage: 100%}; for an event where market demand for a product suddenly increases, the parameters can be set as {growth multiple: 2.5, duration: 30 days}.

[0059] For each candidate topology, simulations are performed under a preset perturbation scenario to determine the optimal scheduling strategy that optimizes the performance of the candidate topology. Optionally, a simulation environment can be constructed by combining the candidate topology with the preset perturbation scenario, and a deep reinforcement learning model can be run to determine the optimal scheduling strategy corresponding to the candidate topology.

[0060] Step 104: Simulate and run each candidate topology according to the optimal scheduling strategy of each candidate topology to determine the quantization results of each candidate topology.

[0061] Based on the optimal scheduling strategy of each candidate topology, simulations are performed on each candidate topology under a preset disturbance scenario. The operational performance parameters of each candidate topology during the simulation are obtained. Based on the operational performance parameters, the quantitative results of each candidate topology under a specific evaluation system can be determined, including the values ​​of various performance indicators, the comprehensive score, or comparative analysis with other candidate topologies, thereby providing a basis for the topology selection and optimization of the target industry chain.

[0062] In the above embodiments, firstly, multi-source heterogeneous data associated with the target industry chain is acquired. Then, based on the multi-source heterogeneous data, at least one candidate topology corresponding to the target industry chain is generated. Next, for each candidate topology, simulation operation is performed under a preset disturbance scenario to determine the optimal scheduling strategy for each candidate topology. Finally, simulation operation is performed on each candidate topology according to its optimal scheduling strategy to determine the quantitative result of each candidate topology. In this way, the disturbance resistance capability of each candidate topology in the target industry chain is simulated, and the quantitative result of each candidate topology operating under its respective optimal scheduling strategy is obtained, improving the accuracy of the simulation evaluation of the target industry chain.

[0063] In the embodiments of this application, the process of generating at least one candidate topology corresponding to the target industry chain based on multi-source heterogeneous data is as follows: Figure 2 As shown, it may include:

[0064] Step 201: Generate a baseline topology for the target industry chain based on multi-source heterogeneous data.

[0065] The acquisition of multi-source heterogeneous data associated with the target industry chain may include: identifying at least one target system associated with the target industry chain; and acquiring multi-source heterogeneous data from at least one target system.

[0066] The target system can be any system in the upstream or downstream of the target industry chain. Since multi-source heterogeneous data can include status data of each physical entity in the target industry chain, business process data within the target industry chain, and external environmental data affecting the operation of the target industry chain, this data can be used. Status data of each physical entity can include status data of facilities such as factories, warehouses, and equipment. Business process data can include order, logistics, and production plan execution data. External environmental data affecting the operation of the industry chain includes market demand and policy / regulatory data.

[0067] The status data of each physical entity in the supply chain can be collected from the production site control system via interfaces, acquiring equipment operating parameters such as temperature, pressure, and speed. The production site control system may include SCADA (Supervisory Control And Data Acquisition) and PLC (Programmable Logic Controller), as well as real-time inventory level data of materials obtained from WMS (Warehouse Management System). Business process data can be extracted from ERP (Enterprise Resource Planning) systems via interfaces, including order information such as order number, product type, required quantity, and delivery date; production order execution status and material consumption records from MES (Manufacturing Execution System); and logistics order location and estimated arrival time from TMS (Transportation Management System). External environmental data affecting the supply chain operation can be obtained from publicly available data service platforms via web crawlers or application programming interfaces, such as price indices of key raw materials, macroeconomic indicators of target markets, or weather forecast data for specific regions.

[0068] After processing and fusing multi-source heterogeneous data, a complete snapshot dataset with consistent structure and temporal stacking is obtained. Based on this dataset, a baseline topology for the target industry chain is generated. The baseline topology can be defined as an initial attribute graph G=(V,E,P). In the initial attribute graph, V represents the set of nodes, E represents the set of edges, and P represents the set of global attributes and rules.

[0069] Specifically, a node This can represent an entity within a supply chain. For example, v1 could be a factory, whose attributes could include: type: factory, ID: F01, capacity: 1000 units / day, location: city A. Another node... This can represent a warehouse, whose attributes may include {Type: Warehouse, ID: WO1, Capacity: 5000 items, Operating Cost: 10 yuan / item}. An edge It can represent the relationship between entities, such as the transportation path connecting nodes F01 and W01. Its attributes can include {Type: Logistics, Transportation Time: 2 days, Transportation Cost: 5 yuan / piece}. Global attributes and rules P define the macro-constraints of the entire target industrial chain, such as {Total carbon emission cap: 1000 tons / year}.

[0070] Step 202: Edit and modify the baseline topology to generate at least one candidate topology.

[0071] After generating the initial attribute graph, at least one new attribute graph G is generated based on the initial attribute graph by performing preset modification operations. ’ Each new attribute graph G ’ This constitutes a candidate topology. Optionally, modification operations specifically include modifying the node set V, edge set E, or global attribute and rule set P of the initial attribute graph G. For example, node modification operations may include: adding a new node (e.g., creating a new factory F02), deleting an existing node (e.g., closing warehouse W01), or modifying node attributes (e.g., increasing the capacity of factory F01 to 1500 units / day). Edge modification operations may include: adding a new edge (e.g., establishing a new transportation path from factory F02 to warehouse W01), or modifying edge attributes (e.g., adjusting the cost of a transportation path).

[0072] In one embodiment, the steps for simulating candidate topologies under a preset disturbance scenario and determining the optimal scheduling strategy for each candidate topology are as follows: Figure 3 As shown, it may include:

[0073] Step 301: Iterate through the candidate topology under a preset perturbation scenario to obtain the policy function corresponding to each run result.

[0074] Step 302: The policy function that has reached the convergence state is determined as the optimal dynamic scheduling policy for the candidate topology.

[0075] In this process, a simulation environment is constructed by combining candidate topologies with disturbance scenarios. A deep reinforcement learning model is then used for iterative operation to determine the optimal dynamic scheduling strategy corresponding to the candidate topologies.

[0076] During the iterative process, the relevant elements of reinforcement learning include the system state S. t and scheduling action a t Among them, the system state S t This refers to a vector that provides a complete description of the operational status of the target supply chain at simulation time t. Specifically, this vector may include: the current capacity utilization rate and work-in-process inventory level of each factory node; the material inventory level of each warehouse node; and the information on in-transit batches of materials and their estimated arrival times along each transportation path.

[0077] Scheduling action a t This refers to the observation of the system state S. tSubsequently, the deep reinforcement learning model can execute a set of scheduling decisions. This set can specifically include: adjusting the production priority or production rate of a factory; initiating an emergency purchase order for an alternative supplier; or changing the transportation mode or route of a batch of goods in transit.

[0078] The optimal dynamic scheduling strategy is ultimately expressed as a policy function π, which is used to express the system state S at any given time. t Mapped to a scheduling action a t That is, a t =π(S t The objective of this strategy function is to maximize the preset long-term cumulative return R. t The calculation method is shown in the following formula.

[0079]

[0080] in, It is the immediate reward obtained from the simulation environment after the action is executed at the (k+1)th time step in the future. For example, a positive reward is obtained for successfully delivering an order, while a negative reward is obtained for inventory backlog or order delay. The discount factor is a constant between 0 and 1, used to balance the importance of immediate returns versus future returns. k is a time step counter variable used to iterate and sum over all immediate returns from the current moment into the future.

[0081] In determining the policy function, the deep reinforcement learning model repeatedly cycles through observing states, executing actions, and receiving rewards in the simulation environment. Utilizing the reward signals, it continuously updates its internal network parameters using algorithms such as policy gradient or Q-learning until the policy function π converges. The converged policy function π represents the optimal dynamic scheduling policy for the current candidate topology and the preset perturbation scenario.

[0082] In one embodiment, after determining the optimal dynamic scheduling strategy for each candidate topology, the steps for simulating and running each candidate topology according to its optimal scheduling strategy, and determining the quantization results for each candidate topology are as follows: Figure 4 As shown, it includes:

[0083] Step 401: For each candidate topology, simulate the candidate topology under a preset disturbance scenario according to the optimal dynamic scheduling strategy corresponding to the candidate topology, and obtain at least one key performance indicator of the candidate topology.

[0084] After determining each candidate topology and its corresponding optimal dynamic scheduling strategy, a simulation of each candidate topology under a preset disturbance scenario is run. After the simulation is completed, at least one key performance indicator (KPI) is extracted from the recorded simulation data through quantitative calculation. The KPI may include one or more of the following: economic indicators, resilience indicators, or efficiency indicators.

[0085] The economic indicators include calculating the total profit over the entire simulation period, which is calculated by subtracting total costs (including production costs, inventory holding costs, transportation costs, and order delay penalties) from total sales revenue. The resilience indicators include calculating the system recovery time, defined as the time elapsed from the occurrence of the disturbance event until key output indicators in the supply chain (such as order fulfillment rate) recover to 95% of their pre-disruption levels. The efficiency indicators include calculating the on-time delivery rate, which is calculated by dividing the number of orders delivered before the specified delivery date by the total number of orders.

[0086] Step 402: Determine the comprehensive strategic fitness score of the candidate topology based on each key performance indicator and its corresponding weight.

[0087] After determining each key performance indicator (KPI), the KPIs of different dimensions are normalized. Then, based on the pre-set weights of the multiple KPIs, the normalized KPIs are weighted and summed as shown below.

[0088]

[0089] Where F is the comprehensive strategic fitness score of the candidate topology; m is the number of key performance indicators; k i The value of the key performance indicator after normalization; w i These are the preset weight values ​​corresponding to the key performance indicators, and the sum of all weights is 1; the weight set {w1, w2, ..., w...} i This reflects specific strategic preferences, namely, preferences for each key performance indicator.

[0090] Step 403: Determine the quantitative results of each candidate topology based on the comprehensive strategic fitness score of each candidate topology.

[0091] Based on the above simulation and calculation, all candidate topologies and their corresponding comprehensive strategic fitness scores are obtained. All candidate topologies are then sorted in descending or ascending order according to their scores to generate the final optimal ranking result, which is also the quantitative result of each candidate topology.

[0092] In an exemplary embodiment, after obtaining the quantization results of each candidate topology of the target industry chain, the quantization results of each candidate topology can be visualized, such as... Figure 5 As shown, the method also includes:

[0093] Step 501: Generate a topology resilience radar chart based on the key performance indicators corresponding to each candidate topology.

[0094] By extracting multiple key performance indicators corresponding to each candidate topology from the simulation process of each candidate topology, a comparative radar chart is generated for multiple candidate topologies.

[0095] Specifically, each key performance indicator (KPI), such as economic indicators, resilience indicators, and efficiency indicators, is set as an independent coordinate axis on the radar chart. For each candidate topology, its quantified value on that KPI determines its data point position on that coordinate axis. By connecting the data points of a single candidate topology on all coordinate axes, a closed polygon is formed. Overlaying different candidate topologies onto the same radar chart allows for differentiation of different candidate topologies using different colors or line styles, thus enabling a direct comparison of the performance of each candidate topology across multiple performance dimensions.

[0096] Step 502: Based on the system state snapshot data of each candidate topology at different times during the simulation process, generate system state evolution animations corresponding to each candidate topology;

[0097] The system acquires system snapshot data at different moments during the simulation process for each candidate topology, dynamically demonstrating the process of different candidate topologies responding to the same disturbance scenario. Specifically, it acquires timestamped system state snapshot data recorded by each candidate topology during the simulation run. Then, in chronological order, it renders the system state of each frame—such as the inventory level, production status, and logistics disruption status of each node—as a graphical interface for a candidate topology. By continuously playing these rendered graphical interfaces, a dynamic animation is generated. Through the system state demonstration animations corresponding to each candidate topology, it is possible to review and compare the system state evolution paths and response differences of different candidate topologies under disturbance scenarios, such as after a supplier supply disruption.

[0098] Step 503 displays the topology resilience radar plot and system state evolution animation.

[0099] The aforementioned topology resilience radar chart and system status demonstration animation can be displayed through the front-end display interface.

[0100] In embodiments of this application, the topology quantization method can be applied to, for example... Figure 6The supply chain simulation and optimization system shown is deployed in the terminal. The supply chain simulation and optimization system may include a digital twin data module, a strategic topology sandbox module, a scenario integration engine, a dynamic strategy engine, a strategic adaptability assessment module, and a decision support presentation module.

[0101] The digital twin data module integrates heterogeneous data from multiple sources, including physical entity data, business process data, and external environment data, originating from the target industry chain, providing a unified data foundation for the industry chain simulation and optimization system. The digital twin data module may further include a physical entity data acquisition unit for acquiring the status data of each physical entity; a business process data acquisition unit for acquiring business process data; an external environment data acquisition unit for acquiring external environment data affecting the operation of the industry chain; and a data integration unit, connected to the above three acquisition units, for performing structured processing and timestamp alignment and fusion of the acquired physical entity data, business process data, and external environment data to form the unified data foundation.

[0102] The strategic topology sandbox module, based on the aforementioned digital twin data module, establishes a unified data foundation and generates at least one candidate topology with a structure different from the target industry chain benchmark topology. The strategic topology sandbox module also includes a topology definition unit for generating an initial attribute graph, and a topology generation unit for generating at least one candidate topology from the initial attribute graph.

[0103] Scene synthesis engine, used to generate preset perturbation scenes.

[0104] The dynamic policy engine configures a deep reinforcement learning model and interacts with the candidate topology in a simulation environment under perturbation scenarios by running the deep reinforcement learning model to determine the optimal dynamic scheduling strategy for each candidate topology.

[0105] The strategic fitness assessment module is used to simulate and run each candidate topology according to the optimal scheduling strategy of each candidate topology, and determine the quantitative results of each candidate topology. The strategic fitness assessment module includes a key performance indicator extraction unit, which is used to obtain the key performance indicators of each candidate topology, and a score calculation unit, which is used to calculate the comprehensive strategic fitness score of each candidate topology based on the corresponding key performance indicators, and generate the quantitative results.

[0106] The decision support presentation module is used to acquire the quantitative results of each candidate topology and generate multi-dimensional visualized decision information. This includes a topology resilience radar chart generation unit, which generates a topology resilience radar chart based on multiple key performance indicators to compare the performance of each candidate topology; and a key event simulation and debriefing animation generation unit, which dynamically displays key event simulation and debriefing animations of the response processes of different candidate topologies under the same disturbance scenario, thus dynamically demonstrating the response processes of different candidate topologies.

[0107] like Figure 7 The diagram shows a flowchart of the topology quantification method provided in this application embodiment. First, the digital twin data module integrates and models the target industry chain data. The digital twin data module obtains physical entity data, business process data, and external environment data from various information systems in the target industry chain, and processes and merges these multi-source heterogeneous data to form a unified data foundation, providing data support for subsequent steps.

[0108] Next, based on the unified data foundation established in the above steps, the strategic topology sandbox module defines and generates the industry chain topology. The module first formally defines the current target industry chain structure as a baseline topology. Then, based on a series of preset strategic assumptions, such as adding new suppliers, establishing new production bases, or adjusting the logistics network, it modifies the nodes or edges of the baseline topology, thereby generating one or more candidate topologies to be evaluated.

[0109] Meanwhile, the scenario integration engine generates one or more disturbance scenarios. Each disturbance scenario is an event vector containing specific content, occurrence time, and relevant parameters, used to simulate external shocks to the industrial chain in the simulation, such as sharp fluctuations in raw material prices and the closure of major transportation ports.

[0110] Subsequently, the supply chain simulation optimization system enters a cyclical evaluation process for all candidate topologies. For each candidate topology to be evaluated, the dynamic policy engine receives the candidate topology and the generated disturbance scenario, and determines an exclusive optimal dynamic scheduling policy for the specific candidate topology and disturbance scenario combination through its internal deep reinforcement learning model.

[0111] After obtaining the optimal dynamic scheduling strategy, the strategic fitness assessment module, driven by this strategy, runs a simulation of the candidate topology under a perturbation scenario. After the simulation, the key performance indicator extraction unit within the strategic fitness assessment module extracts the quantified values ​​of several preset key performance indicators from the simulation results. The scoring calculation unit then uses the extracted key performance indicators to calculate the comprehensive strategic fitness score of the current candidate topology through a weighted summation.

[0112] The supply chain simulation optimization system determines whether all candidate topologies have been evaluated. If not, it returns to the next candidate topology and repeats the process of strategy determination, simulation evaluation, and score calculation. If all candidate topologies have been evaluated, they are ranked according to their comprehensive strategic fitness scores to generate the final optimal ranking result.

[0113] Finally, the decision support presentation module receives the optimal ranking results and the key performance indicator data recorded during the simulation. Based on these inputs, the decision support presentation module generates multi-dimensional visualized decision information such as topology resilience radar charts and key event deduction and recap animations, and outputs them. At this point, the entire collaborative simulation optimization process is complete.

[0114] In the embodiments of this application, by setting up a digital twin data module, a strategic topology sandbox module, and a scenario integration engine, the real-world industrial chain can be abstracted into a computable topological structure, and its dynamic response to various unexpected disturbance events can be simulated. This enables the system to comprehensively simulate the anti-disturbance capabilities of different candidate topologies as a whole, solving the problem that existing technologies struggle to systematically and quantitatively assess the resilience of the industrial chain. Simultaneously, the dynamic strategy engine employs a deep reinforcement learning model, enabling each candidate topology to autonomously learn and determine a unique optimal dynamic scheduling strategy under specific disturbance scenarios. This ensures that the final evaluation of each candidate topology is based on its optimal operational performance, rather than on a fixed, suboptimal operating rule, thus making the comparison between candidate topologies fairer and the evaluation results more accurately reflect the true potential of each candidate topology. Furthermore, through the strategic fitness evaluation module, the complex simulation process is ultimately converged into a quantified comprehensive strategic fitness score and a clear optimal ranking result. The decision support presentation module generates visualizations such as topology resilience radar charts and key event deduction and recap animations, presenting data in an intuitive way. This provides managers with objective data basis and clear logical support for selecting the optimal industrial chain structure, reducing the subjectivity of decision-making.

[0115] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0116] Based on the same inventive concept, this application also provides a topology quantization apparatus for implementing the topology quantization method described above. The solution provided by this apparatus is similar to the implementation scheme described in the above method; therefore, the specific limitations in one or more topology quantization apparatus embodiments provided below can be found in the limitations of the topology quantization method described above, and will not be repeated here.

[0117] In one exemplary embodiment, such as Figure 8 As shown, a topology quantization device is provided, comprising: an acquisition module, a generation module, a first determination module, and a second determination module, wherein:

[0118] The acquisition module is used to acquire multi-source heterogeneous data related to the target industry chain;

[0119] The generation module is used to generate at least one candidate topology corresponding to the target industry chain based on multi-source heterogeneous data;

[0120] The first determining module is used to perform simulation operation on each candidate topology under a preset disturbance scenario to determine the optimal scheduling strategy for each candidate topology.

[0121] The second determination module is used to simulate and run each candidate topology according to the optimal scheduling strategy of each candidate topology, and determine the quantization result of each candidate topology.

[0122] In one embodiment, the generation module is specifically used to generate a baseline topology of the target industry chain based on multi-source heterogeneous data; and to edit and modify the baseline topology to generate at least one candidate topology.

[0123] In one embodiment, the first determining module is specifically used to iteratively run the candidate topology under a preset disturbance scenario to obtain the strategy function corresponding to each running result; and to determine the strategy function that has reached the convergence state as the optimal dynamic scheduling strategy of the candidate topology.

[0124] In one embodiment, the second determining module is specifically used to, for each candidate topology, simulate and run the candidate topology under a preset disturbance scenario according to the optimal dynamic scheduling strategy corresponding to the candidate topology, and obtain at least one key performance indicator of the candidate topology; determine the comprehensive strategic fitness score of the candidate topology according to each key performance indicator and the weight value corresponding to each key performance indicator; and determine the quantitative result of each candidate topology according to the comprehensive strategic fitness score of each candidate topology.

[0125] In one embodiment, the device further includes a display module for generating a topology resilience radar chart based on the key performance indicators corresponding to each candidate topology; generating a system state evolution animation corresponding to each candidate topology based on system state snapshot data at different times during the simulation process; and displaying the topology resilience radar chart and the system state evolution animation.

[0126] In one embodiment, the acquisition module is specifically used to identify at least one target system associated with the target industry chain; and to acquire multi-source heterogeneous data from the at least one target system.

[0127] Each module in the aforementioned topology quantization device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the operations corresponding to each module.

[0128] In one exemplary embodiment, a computer device is provided, which may be a terminal, and its internal structure diagram may be as follows: Figure 9 As shown, the computer device includes a processor, memory, input / output interfaces, a communication interface, a display unit, and an input device. The processor, memory, and input / output interfaces are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interfaces are used for exchanging information between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When executed by the processor, the computer program implements a topology quantization method. The display unit is used to form a visually visible image and can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of the computer device can be a touch layer covering the display screen, or buttons, trackballs, or touchpads set on the casing of the computer device, or external keyboards, touchpads, or mice, etc.

[0129] Those skilled in the art will understand that Figure 9The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0130] In an exemplary embodiment, a computer device is provided, including a memory and a processor. The memory stores a computer program, and the processor executes the computer program to perform the following steps: acquiring multi-source heterogeneous data associated with a target industry chain; generating at least one candidate topology corresponding to the target industry chain based on the multi-source heterogeneous data; for each candidate topology, simulating the candidate topology under a preset disturbance scenario to determine the optimal scheduling strategy for each candidate topology; and simulating the candidate topology according to the optimal scheduling strategy for each candidate topology to determine the quantization result of each candidate topology.

[0131] In one embodiment, when the processor executes the computer program, it also performs the following steps: generating a baseline topology for the target industry chain based on multi-source heterogeneous data; editing and modifying the baseline topology to generate at least one candidate topology.

[0132] In one embodiment, when the processor executes the computer program, it further performs the following steps: iteratively running the candidate topology under a preset disturbance scenario to obtain the policy function corresponding to each running result; and determining the policy function that reaches the convergence state as the optimal dynamic scheduling policy for the candidate topology.

[0133] In one embodiment, when the processor executes the computer program, it further performs the following steps: for each candidate topology, it simulates the candidate topology under a preset disturbance scenario according to the optimal dynamic scheduling strategy corresponding to the candidate topology, and obtains at least one key performance indicator of the candidate topology; it determines the comprehensive strategic fitness score of the candidate topology according to each key performance indicator and the weight value corresponding to each key performance indicator; and it determines the quantitative result of each candidate topology according to the comprehensive strategic fitness score of each candidate topology.

[0134] In one embodiment, when the processor executes the computer program, it further performs the following steps: generating a topology resilience radar chart based on the key performance indicators corresponding to each candidate topology; generating a system state evolution animation corresponding to each candidate topology based on system state snapshot data at different times during the simulation process; and displaying the topology resilience radar chart and the system state evolution animation.

[0135] In one embodiment, when the processor executes the computer program, it further performs the following steps: identifying at least one target system associated with the target industry chain; and acquiring multi-source heterogeneous data from the at least one target system.

[0136] In one embodiment, a computer-readable storage medium is provided, on which a computer program is stored. When the computer program is executed by a processor, it performs the following steps: acquiring multi-source heterogeneous data associated with a target industry chain; generating at least one candidate topology corresponding to the target industry chain based on the multi-source heterogeneous data; for each candidate topology, simulating the candidate topology under a preset disturbance scenario to determine the optimal scheduling strategy for each candidate topology; and simulating the candidate topology according to the optimal scheduling strategy for each candidate topology to determine the quantization result of each candidate topology.

[0137] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: generating a baseline topology for the target industry chain based on multi-source heterogeneous data; editing and modifying the baseline topology to generate at least one candidate topology.

[0138] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: iteratively running the candidate topology under a preset disturbance scenario to obtain the policy function corresponding to each running result; and determining the policy function that reaches the convergence state as the optimal dynamic scheduling policy for the candidate topology.

[0139] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: for each candidate topology, simulates the candidate topology under a preset disturbance scenario according to the optimal dynamic scheduling strategy corresponding to the candidate topology, and obtains at least one key performance indicator of the candidate topology; determines the comprehensive strategic fitness score of the candidate topology according to each key performance indicator and the weight value corresponding to each key performance indicator; and determines the quantitative result of each candidate topology according to the comprehensive strategic fitness score of each candidate topology.

[0140] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: generating a topology resilience radar chart based on the key performance indicators corresponding to each candidate topology; generating a system state evolution animation corresponding to each candidate topology based on system state snapshot data at different times during the simulation process; and displaying the topology resilience radar chart and the system state evolution animation.

[0141] In one embodiment, when the computer program is executed by a processor, it further performs the following steps: identifying at least one target system associated with a target industry chain; and acquiring multi-source heterogeneous data from at least one target system.

[0142] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, performs the following steps: acquiring multi-source heterogeneous data associated with a target industry chain; generating at least one candidate topology corresponding to the target industry chain based on the multi-source heterogeneous data; for each candidate topology, performing simulation operation on the candidate topology under a preset disturbance scenario to determine the optimal scheduling strategy for each candidate topology; and performing simulation operation on each candidate topology according to the optimal scheduling strategy for each candidate topology to determine the quantification result of each candidate topology.

[0143] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: generating a baseline topology for the target industry chain based on multi-source heterogeneous data; editing and modifying the baseline topology to generate at least one candidate topology.

[0144] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: iteratively running the candidate topology under a preset disturbance scenario to obtain the policy function corresponding to each running result; and determining the policy function that reaches the convergence state as the optimal dynamic scheduling policy for the candidate topology.

[0145] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: for each candidate topology, simulates the candidate topology under a preset disturbance scenario according to the optimal dynamic scheduling strategy corresponding to the candidate topology, and obtains at least one key performance indicator of the candidate topology; determines the comprehensive strategic fitness score of the candidate topology according to each key performance indicator and the weight value corresponding to each key performance indicator; and determines the quantitative result of each candidate topology according to the comprehensive strategic fitness score of each candidate topology.

[0146] In one embodiment, when the computer program is executed by the processor, it further performs the following steps: generating a topology resilience radar chart based on the key performance indicators corresponding to each candidate topology; generating a system state evolution animation corresponding to each candidate topology based on system state snapshot data at different times during the simulation process; and displaying the topology resilience radar chart and the system state evolution animation.

[0147] In one embodiment, when the computer program is executed by a processor, it further performs the following steps: identifying at least one target system associated with a target industry chain; and acquiring multi-source heterogeneous data from at least one target system.

[0148] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0149] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0150] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0151] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for quantizing topology, characterized in that, The method includes: Acquire multi-source heterogeneous data related to the target industry chain; Based on the multi-source heterogeneous data, at least one candidate topology corresponding to the target industry chain is generated; For each candidate topology, the candidate topology is simulated under a preset disturbance scenario to determine the optimal scheduling strategy for each candidate topology; The candidate topologies are simulated and run according to their optimal scheduling strategies to determine the quantization results of each candidate topology.

2. The method according to claim 1, characterized in that, The step of generating at least one candidate topology corresponding to the target industry chain based on the multi-source heterogeneous data includes: Based on the multi-source heterogeneous data, a baseline topology for the target industry chain is generated; The baseline topology is edited and modified to generate at least one candidate topology.

3. The method according to claim 1, characterized in that, The step of simulating the candidate topologies under a preset disturbance scenario to determine the optimal scheduling strategy for each candidate topology includes: The candidate topology is iteratively run under a preset perturbation scenario to obtain the strategy function corresponding to each run result; The policy function that reaches convergence is determined as the optimal dynamic scheduling policy for the candidate topology.

4. The method according to claim 1, characterized in that, The step of simulating and running each candidate topology according to the optimal scheduling strategy of each candidate topology to determine the quantization result of each candidate topology includes: For each candidate topology, the candidate topology is simulated under a preset disturbance scenario according to the optimal dynamic scheduling strategy corresponding to the candidate topology, and at least one key performance indicator of the candidate topology is obtained. The comprehensive strategic fitness score of the candidate topology is determined based on each of the key performance indicators and the corresponding weight values ​​of each key performance indicator. The quantitative results of each candidate topology are determined based on the comprehensive strategic fitness score of each candidate topology.

5. The method according to claim 4, characterized in that, The method further includes: A topology resilience radar chart is generated based on the key performance indicators corresponding to each candidate topology. Based on the system state snapshot data of each candidate topology at different times during the simulation process, generate system state evolution animations corresponding to each candidate topology; The topology resilience radar chart and the system state evolution animation are displayed.

6. The method according to any one of claims 1 to 5, characterized in that, The acquisition of multi-source heterogeneous data related to the target industry chain includes: Identify at least one target system associated with the target industry chain; The multi-source heterogeneous data is obtained from the at least one target system.

7. A topology quantization device, characterized in that, The device includes: The acquisition module is used to acquire multi-source heterogeneous data related to the target industry chain; The generation module is used to generate at least one candidate topology corresponding to the target industry chain based on the multi-source heterogeneous data; The first determining module is used to perform simulation operation on each candidate topology under a preset disturbance scenario to determine the optimal scheduling strategy for each candidate topology. The second determining module is used to perform simulation operation on each of the candidate topologies according to the optimal scheduling strategy of each candidate topology, and determine the quantization result of each candidate topology.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images