Scheduling decision-making method, device and equipment of distributed industry chain and storage medium

Abstract

The invention relates to a scheduling decision-making method, device and equipment of a distributed industry chain and a storage medium. The method comprises the following steps: receiving first decision-making required information sent by a first lower-level scheduling unit of a scheduling unit; receiving information required by a second decision sent by a second lower-level scheduling unit of the scheduling unit; when it is determined that at least one of the first lower-level scheduling unit and the second lower-level scheduling unit generates energy level change according to the first energy level information and the second energy level information, target sensing potential energy of the scheduling unit is determined according to the information needed by the first decision and the information needed by the second decision, and a first scheduling decision is generated according to the target sensing potential energy. By adopting the method for distributed state sensing and decision making, the transmission time of information of different regions of the industrial chain is shortened, the overall demand pressure of the whole lower-level industrial chain can be obtained through the back propagation mechanism and only by calculating the local basic potential energy, and the efficiency and accuracy of industrial chain scheduling decision making can be improved.

Description

Distributed supply chain scheduling decision-making methods, devices, equipment and storage media Technical Field

[0001] This application relates to the field of intelligent industrial chain technology, and in particular to a scheduling decision-making method, apparatus, equipment and storage medium for a distributed industrial chain. Background Technology

[0002] In the context of modern industrial production and globalized trade, the efficiency of supply chain coordination and scheduling directly affects a company's core competitiveness and the overall economic efficiency. To achieve precise matching and efficient flow of materials, information, and capital, information technology is widely used in supply chain scheduling and management, resulting in various advanced planning and scheduling systems.

[0003] Current supply chain scheduling solutions employ a centralized system architecture. This architecture relies on a powerful central control server or planning center. Its core operating mode involves periodically collecting massive amounts of status data from various links in the supply chain (such as suppliers, factories, warehouses, and logistics providers). Then, a complex global optimization algorithm is run on the central server to calculate a theoretically optimal production and transportation plan covering the entire supply chain. Finally, the generated scheduling instructions are distributed to each execution unit. This top-level design model leverages its advantage in overall coordination when dealing with relatively stable and predictable production scenarios.

[0004] However, in order to maintain a global perspective, traditional supply chain scheduling methods require decision-making systems to bear the enormous overhead of long-distance transmission of massive amounts of data and centralized computing, resulting in significant delays in decision-making. Summary of the Invention

[0005] Therefore, it is necessary to provide a distributed supply chain scheduling decision-making method, device, equipment, and storage medium that can improve the scheduling decision-making efficiency of the supply chain in response to the above-mentioned technical problems.

[0006] Firstly, this application provides a distributed supply chain scheduling decision-making method, applied to each scheduling unit in a supply chain decision-making system, including:

[0007] The system receives first decision-required information sent by the first subordinate scheduling unit of the scheduling unit; the first decision-required information includes the first energy potential information and the first basic potential energy of the first subordinate scheduling unit; the first subordinate scheduling unit is a scheduling unit without scheduling requirements.

[0008] The system receives second decision-required information sent by a second lower-level scheduling unit of the scheduling unit; the second decision-required information includes second energy bit information, second basic potential energy, and second requirement signature vector of the second lower-level scheduling unit; the second lower-level scheduling unit is a scheduling unit with scheduling requirements.

[0009] When it is determined, based on the first energy potential information and the second energy potential information, that at least one of the first lower-level scheduling unit and the second lower-level unit has generated an energy potential change, the target sensing potential energy of the scheduling unit is determined based on the first decision-required information and the second decision-required information, and a first scheduling decision is generated based on the target sensing potential energy.

[0010] In one embodiment, determining the target perception potential energy of the scheduling unit based on the first decision-required information and the second decision-required information includes:

[0011] The first sensing potential energy of the scheduling unit is determined based on the first energy potential information and the first basic potential energy.

[0012] The second sensing potential energy of the scheduling unit is determined based on the second energy level information, the second basic potential energy, and the second demand signature vector.

[0013] The target sensing potential energy is determined based on the first sensing potential energy and the second sensing potential energy.

[0014] In one embodiment, determining the first sensing potential energy of the scheduling unit based on the first energy potential information and the first basic potential energy includes:

[0015] For each of the first decision-required information, the first energy potential information and the first basic potential energy are used to determine the first target basic potential energy contributed by each first lower-level scheduling unit to the scheduling unit based on the first energy potential information and the first basic potential energy.

[0016] The first perceived potential energy is obtained by summing and calculating all the basic potential energies of the first target.

[0017] In one embodiment, determining the first target basic potential energy contributed by each first lower-level scheduling unit to the scheduling unit based on the first energy potential information and the first basic potential energy includes:

[0018] The importance of the first energy potential information is adjusted according to a preset first adjustment coefficient to obtain the adjusted energy potential;

[0019] The importance of the first basic potential energy is adjusted according to a preset second adjustment coefficient to obtain the adjusted potential energy;

[0020] The first target basic potential energy is obtained by performing a summation operation on the adjustable energy potential and the adjustable potential energy.

[0021] In one embodiment, determining the second sensing potential energy of the scheduling unit based on the second energy bit information, the second basic potential energy, and the second demand signature vector includes:

[0022] For each of the second decision-required information, the second potential information and the second basic potential energy are used to determine the second target basic potential energy contributed by each second lower-level scheduling unit to the scheduling unit based on the second potential information and the second basic potential energy.

[0023] The resonant gain of the scheduling unit is determined based on the cosine similarity between the second demand signature vector and the capability profile vector of the scheduling unit.

[0024] The second sensing potential energy is determined based on the second target fundamental potential energy and the resonant gain.

[0025] In one embodiment, generating a first scheduling decision based on the target perceived potential energy includes:

[0026] The equity pledge amount of the scheduling unit is determined based on the target perception potential energy of the scheduling unit and the preset virtual equity.

[0027] The first scheduling decision is generated based on the comparison between the pledged equity amount and the preset pledge threshold; the first scheduling decision includes the result of participating in the bidding and the result of not participating in the bidding.

[0028] In one embodiment, the method further includes:

[0029] The operating parameters of the scheduling unit are obtained and determined, and the third energy bit information of the scheduling unit is determined based on the operating parameters;

[0030] The third basic potential energy is determined based on the first potential energy information, the second potential energy information, the first basic potential energy, and the second basic potential energy.

[0031] When the scheduling unit's scheduling request indicates that the scheduling unit has no scheduling request, the third decision-making information is generated based on the third potential information and the third basic potential energy, and the third decision-making information is sent to the upper-level scheduling unit.

[0032] When the scheduling unit's scheduling request indicates that the scheduling unit has a scheduling need, a first request signature vector is generated, and a fourth decision-making information is generated based on the third energy bit information, the third basic potential energy, and the first request signature vector. The fourth decision-making information is then sent to the upper-level scheduling unit.

[0033] Secondly, this application also provides a scheduling and decision-making device for a distributed industrial chain, comprising:

[0034] The first receiving module is used to receive the first decision-required information sent by the first subordinate scheduling unit of the scheduling unit; the first decision-required information includes the first energy potential information and the first basic potential energy of the first subordinate scheduling unit; the first subordinate scheduling unit is a scheduling unit without scheduling requirements.

[0035] The second receiving module is used to receive the second decision-required information sent by the second lower-level scheduling unit of the scheduling unit; the second decision-required information includes the second energy bit information, the second basic potential energy, and the second requirement signature vector of the second lower-level scheduling unit; the second lower-level scheduling unit is a scheduling unit with scheduling requirements;

[0036] The determination module is configured to, when determining, based on the first energy potential information and the second energy potential information, that at least one of the first lower-level scheduling unit and the second lower-level unit has generated an energy potential change, determine the target sensing potential energy of the scheduling unit based on the first decision-required information and the second decision-required information, and generate a first scheduling decision based on the target sensing potential energy.

[0037] 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 perform the following steps:

[0038] The system receives first decision-required information sent by the first subordinate scheduling unit of the scheduling unit; the first decision-required information includes the first energy potential information and the first basic potential energy of the first subordinate scheduling unit; the first subordinate scheduling unit is a scheduling unit without scheduling requirements.

[0039] The system receives second decision-required information sent by a second lower-level scheduling unit of the scheduling unit; the second decision-required information includes second energy bit information, second basic potential energy, and second requirement signature vector of the second lower-level scheduling unit; the second lower-level scheduling unit is a scheduling unit with scheduling requirements.

[0040] When it is determined, based on the first energy potential information and the second energy potential information, that at least one of the first lower-level scheduling unit and the second lower-level unit has generated an energy potential change, the target sensing potential energy of the scheduling unit is determined based on the first decision-required information and the second decision-required information, and a first scheduling decision is generated based on the target sensing potential energy.

[0041] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0042] The system receives first decision-required information sent by the first subordinate scheduling unit of the scheduling unit; the first decision-required information includes the first energy potential information and the first basic potential energy of the first subordinate scheduling unit; the first subordinate scheduling unit is a scheduling unit without scheduling requirements.

[0043] The system receives second decision-required information sent by a second lower-level scheduling unit of the scheduling unit; the second decision-required information includes second energy bit information, second basic potential energy, and second requirement signature vector of the second lower-level scheduling unit; the second lower-level scheduling unit is a scheduling unit with scheduling requirements.

[0044] When it is determined, based on the first energy potential information and the second energy potential information, that at least one of the first lower-level scheduling unit and the second lower-level unit has generated an energy potential change, the target sensing potential energy of the scheduling unit is determined based on the first decision-required information and the second decision-required information, and a first scheduling decision is generated based on the target sensing potential energy.

[0045] The aforementioned distributed supply chain scheduling decision-making method, apparatus, equipment, and storage medium are applied to each scheduling unit in the supply chain decision-making system. The system receives first decision-making information from the first subordinate scheduling unit. This first decision-making information includes the first energy level information and the first basic potential energy of the first subordinate scheduling unit. The first subordinate scheduling unit is a scheduling unit without scheduling needs. Simultaneously, the system receives second decision-making information from the second subordinate scheduling unit. This second decision-making information includes the second energy level information, the second basic potential energy, and the second demand signature vector of the second subordinate scheduling unit. The second subordinate scheduling unit is a scheduling unit with scheduling needs. Finally, when it is determined, based on the first and second energy level information, that at least one of the first and second subordinate scheduling units has experienced an energy level change, the system determines the target perception potential energy of the scheduling unit based on the first and second decision-making information, and generates a first scheduling decision based on the target perception potential energy. By treating each scheduling unit in a distributed supply chain as an independent execution entity, each unit independently perceives its own state information and the state information of other scheduling units, and makes independent decisions, and receives energy potential information and basic potential energy from lower-level scheduling units through backpropagation, when a lower-level scheduling unit experiences energy potential changes due to external disturbances, it determines its own perceived potential energy based on the information transmitted by the current lower-level scheduling unit. This allows the response to local disturbances in the supply chain to be executed locally, avoiding decision-making delays and improving the decision-making efficiency of supply chain scheduling. Compared to existing technologies where a central controller simultaneously receives information from all nodes and performs calculations, leading to decision-making delays, this distributed state perception and decision-making by scheduling units reduces the transmission time of information between different areas of the supply chain. Furthermore, through this backpropagation mechanism, the state change of any end scheduling unit in the supply chain is quantified and transmitted level by level to the next level. This allows the upper-level scheduling unit to obtain the overall demand pressure of the entire lower-level supply chain without relying on central instructions, simply by calculating its local basic potential energy, thus improving the efficiency and accuracy of supply chain scheduling decisions. Attached Figure Description

[0046] 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.

[0047] Figure 1 is a structural block diagram of the scheduling unit of a distributed supply chain in one embodiment;

[0048] Figure 2 is a flowchart illustrating the scheduling decision-making method for a distributed supply chain in one embodiment;

[0049] Figure 3 is a flowchart illustrating the process of determining the target sensing potential energy in one embodiment;

[0050] Figure 4 is a flowchart illustrating the process of determining the first sensing potential energy in one embodiment;

[0051] Figure 5 is a flowchart illustrating the process of determining the first target basic potential energy in one embodiment;

[0052] Figure 6 is a flowchart illustrating the process of determining the second sensing potential energy in one embodiment;

[0053] Figure 7 is a schematic diagram of the process of generating the first scheduling decision in one embodiment;

[0054] Figure 8 is a flowchart illustrating the process of determining the third decision-making basis potential energy in one embodiment;

[0055] Figure 9 is a structural block diagram of a distributed supply chain scheduling decision-making device in one embodiment;

[0056] Figure 10 is an internal structure diagram of a computer device in one embodiment. Detailed Implementation

[0057] 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.

[0058] 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.

[0059] In the context of modern industrial production and globalized trade, the efficiency of supply chain coordination and scheduling directly affects the core competitiveness of enterprises and the overall economic efficiency. To achieve precise matching and efficient flow of materials, information, and capital, information technology is widely applied to supply chain scheduling and management, resulting in various advanced planning and scheduling systems. Traditional supply chain scheduling decision-making solutions generally adopt a centralized system architecture. This architecture relies on a powerful central control server or planning center. Its core working mode is to periodically collect massive amounts of status data from various links in the supply chain (such as suppliers, factories, warehouses, and logistics providers), then run complex global optimization algorithms on the central server to calculate a theoretically optimal production and transportation plan covering the entire supply chain, and finally issue the generated scheduling instructions to each execution unit. This top-level design model can leverage its global coordination advantage when dealing with relatively stable and predictable production scenarios. However, the extreme dependence of traditional supply chain scheduling decision-making methods on a single central node makes the stability and reliability of the entire system very fragile. Once the central server fails or the network is interrupted, the entire scheduling capability faces the risk of paralysis. Meanwhile, to maintain a global perspective, the system must bear the enormous overhead of long-distance transmission of massive amounts of data and centralized computing. This not only consumes significant network and computing resources but also leads to substantial delays in decision-making. In the rapidly changing real world, this delay often means that centrally issued scheduling instructions arrive at the execution units lagging behind actual changes. Therefore, when the supply chain faces sudden local disturbances, such as temporary failures of critical equipment or unexpected disruptions to a transportation route, this architecture struggles to achieve rapid and flexible local adaptive adjustments. Its inherent rigidity makes it slow to respond to dynamic uncertainties, thus impacting the resilience and efficiency of the entire supply chain.

[0060] In view of the above-mentioned technical problems, this application provides a scheduling decision method for a distributed supply chain that can improve the transmission efficiency of scheduling decisions. The following embodiments will specifically illustrate the scheduling decision method for the distributed supply chain.

[0061] The distributed supply chain scheduling decision method provided in this application embodiment can be applied to the scheduling unit of the distributed supply chain as shown in Figure 1. The scheduling unit of the distributed supply chain includes a state awareness subunit 102, a potential energy awareness subunit 104, a collaborative decision-making subunit 106, and a data storage subunit 108. The data storage subunit 108 includes a capability profile storage subunit and a virtual equity account subunit. The state awareness subunit 102 is connected to the potential energy awareness subunit 104. The potential energy awareness subunit 104 is connected to the capability profile storage subunit, the collaborative decision-making subunit 106, and other scheduling units in the supply chain. The collaborative decision-making subunit 106 is connected to the virtual equity account subunit. The state perception subunit 102, potential energy perception subunit 104, and collaborative decision-making subunit 106 can each include some or all of sensors or processors such as data acquisition devices, microcontrollers, sensors, and transceivers, used for collecting, processing, and transmitting information from various scheduling units in the industry chain. The state perception subunit 102 collects the physical state or operating parameters of the scheduling unit and calculates the current energy level information of the scheduling unit based on the aforementioned physical state information or operating parameters, transmitting the current energy level information to its superior scheduling unit, and receiving energy level information transmitted from its subordinate scheduling units, transmitting this energy level information to the potential energy perception subunit 104. The potential energy perception subunit 104 receives the energy level information from the state perception subunit 102 and, in conjunction with its own capability profile information, calculates the pressure transmitted by the subordinate scheduling unit, i.e., senses the potential energy. The collaborative decision-making unit executes task bidding or scheduling contract transactions based on the sensed potential energy and virtual rights. The data storage subunit 108 can be any device used to store information about various scheduling units in the industry chain, such as a hard drive, solid-state drive, external hard drive, USB flash drive, memory card, optical disc, or cloud storage. The capability profile storage subunit stores a multi-dimensional vector representing the service characteristics of the scheduling unit; the virtual equity account unit stores, records, and updates the quantity of virtual equity held by the scheduling unit. It should be noted that the scheduling unit in the aforementioned distributed industry chain can be any level of scheduling unit in the distributed industry chain, enabling responses to local disturbances in the industry chain to be executed locally, reducing the complexity of decision-making.

[0062] The following embodiments will specifically illustrate the scheduling decision method for the distributed supply chain based on the scheduling unit of the above-mentioned distributed supply chain.

[0063] In an exemplary embodiment, a scheduling decision method for a distributed supply chain is provided. Taking the application of this method to the scheduling unit of the distributed supply chain in Figure 1 as an example, as shown in Figure 2, the method includes:

[0064] Step S201: Receive the first decision-making information sent by the first subordinate scheduling unit of the scheduling unit.

[0065] The distributed industrial chain can be a coal production and transportation industrial chain, a power dispatch industrial chain, or any other industrial chain; there are no restrictions here. A distributed industrial chain includes multiple dispatch units, each of which is a computing entity with independent computing and communication capabilities, deployed on a specific functional or resource unit within the industrial chain. For example, in the coal production and transportation industrial chain, this chain includes coal mine production, railway transportation, port transshipment, maritime shipping, and downstream power plant consumption. A coal mine loading line, an independent railway line, a port berth, a cargo ship en route, and a power plant's coal bunker can all be configured as a dispatch unit. Multiple dispatch units are interconnected based on the actual paths of material and information flows within the industrial chain, forming a distributed network. The connection relationships between dispatch units define their hierarchical relationships. For example, if dispatch unit A provides materials or services to dispatch unit B, then dispatch unit A is defined as the superior dispatch unit of dispatch unit B, and dispatch unit B is defined as the subordinate dispatch unit of dispatch unit A. Each dispatch unit can correspond to zero, one, or multiple superior and subordinate dispatch units. The first lower-level scheduling unit is the direct lower-level scheduling unit of the scheduling unit that has no scheduling needs, meaning there are no other levels of scheduling units between the scheduling unit and the first lower-level scheduling unit. The information required for the first decision includes the first energy potential information and the first basic potential energy of the first lower-level scheduling unit; the first energy potential information can be the resource saturation or buffer capacity of the scheduling unit at the corresponding time; the first basic potential energy can be the scalar value of the original demand pressure borne by the scheduling unit, originating from all its direct lower-level scheduling units that have no demand, without matching degree correction.

[0066] In the embodiments of this application, the state perception module of each scheduling unit in the distributed industrial chain periodically acquires the real-time operating parameters of the physical facilities bound to the scheduling unit through one or more sensor interfaces, and converts these real-time operating parameters into a unified, normalized scalar value, i.e., energy level information, according to the preset calculation logic. The scheduling unit can be any type of storage physical unit, processing physical unit, and transportation physical unit. The real-time operating parameters and corresponding preset calculation logic processed by different types of scheduling units are different. Optionally, the first method is that the scheduling unit is a storage physical unit, such as a coal yard in a port or a fuel warehouse in a power plant. The real-time operating parameters acquired by the state perception subunit are mainly the current inventory. The calculation relationship of its energy level is expressed by the following formula (1):

[0067] (1);

[0068] in, The real-time inventory level at time t can be obtained directly by sensors such as level gauges. The maximum design capacity of this storage unit is a preset static parameter. At this point, the energy level... The value range is [0,1]. The closer the value is to 1, the closer the inventory of the storage unit is to saturation.

[0069] Alternatively, in a second approach, the scheduling unit is a processing physical unit, such as a loading system in a coal mine or a ship loader in a port. The real-time operating parameters acquired by the state-aware module 10 primarily reflect its current workload. The calculation relationship is expressed by the following formula (2):

[0070] (2);

[0071] in, The instantaneous operating rate or the amount of work in progress at time t can be obtained from the equipment controller or the job management system. The peak processing capacity of this processing unit is a preset static parameter. At this point, the energy level... The value range is also [0,1]. The closer the value is to 1, the closer the operating load of the processing unit is to its capacity limit.

[0072] Alternatively, a third approach is to use a transportation-type physical unit as the scheduling unit, such as a railway section or a ship en route. The energy potential calculation can comprehensively consider both its spatial and temporal occupancy rates. For example, for a railway section, its energy potential can be determined based on the ratio of the number of trains currently operating within that section to the maximum train capacity of that section.

[0073] Optionally, the first basic potential energy can be obtained by progressively transferring and superimposing the potential energy of all subordinate scheduling units of the scheduling unit. For example, a distributed industrial chain includes a coal mine production scheduling unit, a railway transportation scheduling unit, a port transshipment scheduling unit, a maritime shipping scheduling unit, and a subordinate power plant consumption scheduling unit. The scheduling units decrease in level from front to back. The first basic potential energy of the port transshipment scheduling unit is the result of progressively transferring and superimposing the potential energy of the subordinate power plant consumption scheduling unit and the maritime shipping scheduling unit.

[0074] In the embodiments of this application, after the first lower-level scheduling unit calculates the information required for the first decision, it transmits the information required for the first decision to its direct superior scheduling unit. After receiving the information required for the first decision transmitted by the first lower-level scheduling unit, the direct superior scheduling unit calculates the corresponding target sensing potential energy based on the first energy potential information and the first basic potential energy.

[0075] Step S202: Receive the second decision-making information sent by the second lower-level scheduling unit of the scheduling unit.

[0076] The second lower-level scheduling unit is the direct lower-level scheduling unit of the scheduling unit that has scheduling needs; that is, there are no other levels of scheduling units between the scheduling unit and the second lower-level scheduling unit. The information required for the second decision includes the second energy level information, the second basic potential energy, and the second demand signature vector of the second lower-level scheduling unit. The second energy level information can be the resource saturation or buffer capacity of the scheduling unit at the corresponding time. The second basic potential energy can be the scalar value of the original demand pressure borne by the scheduling unit, originating from all its direct lower-level scheduling units with needs, without matching degree correction. The second demand signature vector is used to describe the service nature of the scheduling unit required by the second lower-level scheduling unit.

[0077] In the embodiments of this application, when a scheduling unit needs the service of a higher-level scheduling unit, its collaborative decision-making subunit broadcasts a task message to its higher-level scheduling unit. This task message contains a structured dataset, including: a unique task identifier, a task description, a final deadline for task completion, and a requirement signature vector. Optionally, the state-aware module in the scheduling unit integrates the second energy level information and the second basic potential energy with the task message to generate the information required for the second decision, and then transmits this information to the higher-level scheduling unit. The calculation process for the second energy level information and the second basic potential energy is the same as that for the first energy level information and the second basic potential energy, and will not be elaborated here. After calculating the information required for the second decision, the second lower-level scheduling unit transmits this information to its direct higher-level scheduling unit. Upon receiving the information required for the second decision from the second lower-level scheduling unit, the direct higher-level scheduling unit calculates the corresponding target-aware potential energy based on the second energy level information and the second basic potential energy.

[0078] It should be noted that steps S201 and S202 can be executed sequentially, that is, step S201 can be executed first and then step S202; or step S202 can be executed first and then step S201; or they can be executed simultaneously, without any restriction.

[0079] Step S203: When it is determined, based on the first energy potential information and the second energy potential information, that at least one of the first lower-level scheduling unit and the second lower-level unit has generated an energy potential change, the target sensing potential energy of the scheduling unit is determined based on the first decision-making information and the second decision-making information, and a first scheduling decision is generated based on the target sensing potential energy.

[0080] Among them, the target perception potential energy can be the total scheduling pressure borne by the scheduling unit at the corresponding moment; the first scheduling decision includes the result of participating in the bidding and the result of not participating in the bidding.

[0081] In the embodiments of this application, the real-time operating parameters of the scheduling unit may change abruptly due to actual external disturbances (e.g., a port is closed due to weather). Since the potential energy information is the result of calculation of the real-time operating parameters, the potential energy information of the scheduling unit will also change. When it is determined that at least one of the first lower-level scheduling unit and the second lower-level unit has a change in potential energy based on the first potential energy information and the second potential energy information, that is, it is determined whether the first potential energy information of the first lower-level scheduling unit at the current moment is consistent with the potential energy information at the previous moment, and whether the second potential energy information of the second lower-level scheduling unit at the current moment is consistent with the potential energy information at the previous moment. When the first potential energy information of the first lower-level scheduling unit at the current moment is inconsistent with the potential energy information at the previous moment, or the second potential energy information of the second lower-level scheduling unit at the current moment is inconsistent with the potential energy information at the previous moment, the potential energy of the scheduling unit will be affected in this case. The target perceived potential energy of the scheduling unit is determined based on the information required for the first decision and the information required for the second decision, and the first scheduling decision is generated based on the target perceived potential energy. When the first energy level information of the first lower-level scheduling unit at the current moment is consistent with the energy level information of the previous moment, and the second energy level information of the second lower-level scheduling unit at the current moment is consistent with the energy level information of the previous moment, the potential energy of the scheduling unit will not be affected in this case. The scheduling unit does not need to calculate the perceived potential energy at the current moment, but continues to use the perceived potential energy obtained at the previous calculation moment as the target perceived potential energy, and generates the first scheduling decision based on the target perceived potential energy.

[0082] Optionally, when generating the first scheduling decision based on the target perception potential energy, the collaborative decision-making subunit calculates the equity pledge amount of the scheduling unit based on the target perception potential energy output by the potential energy perception subunit and the virtual equity stored by the data storage subunit, and compares the equity pledge amount with the preset pledge threshold to generate the first scheduling decision.

[0083] The aforementioned distributed supply chain scheduling decision-making method is applied to each scheduling unit in the supply chain decision-making system. It receives first decision-making information from the first subordinate scheduling unit. This first decision-making information includes the first energy level information and the first basic potential energy of the first subordinate scheduling unit, which is a scheduling unit without scheduling needs. Simultaneously, it receives second decision-making information from the second subordinate scheduling unit. This second decision-making information includes the second energy level information, the second basic potential energy, and the second demand signature vector of the second subordinate scheduling unit, which is a scheduling unit with scheduling needs. Finally, when it is determined, based on the first and second energy level information, that at least one of the first and second subordinate scheduling units has experienced an energy level change, the target perception potential energy of the scheduling unit is determined based on the first and second decision-making information, and a first scheduling decision is generated based on the target perception potential energy. By treating each scheduling unit in a distributed supply chain as an independent execution entity, each unit independently perceives its own state information and the state information of other scheduling units, and makes independent decisions, and receives energy potential information and basic potential energy from lower-level scheduling units through backpropagation, when a lower-level scheduling unit experiences energy potential changes due to external disturbances, it determines its own perceived potential energy based on the information transmitted by the current lower-level scheduling unit. This allows the response to local disturbances in the supply chain to be executed locally, avoiding decision-making delays and improving the decision-making efficiency of supply chain scheduling. Compared to existing technologies where a central controller simultaneously receives information from all nodes and performs calculations, leading to decision-making delays, this distributed state perception and decision-making by scheduling units reduces the transmission time of information between different areas of the supply chain. Furthermore, through this backpropagation mechanism, the state change of any end scheduling unit in the supply chain is quantified and transmitted level by level to the next level. This allows the upper-level scheduling unit to obtain the overall demand pressure of the entire lower-level supply chain without relying on central instructions, simply by calculating its local basic potential energy, thus improving the efficiency and accuracy of supply chain scheduling decisions.

[0084] In an exemplary embodiment, as shown in FIG3, the above step S201 "determines the target perception potential energy of the scheduling unit based on the information required for the first decision and the information required for the second decision" includes:

[0085] S301, determine the first sensing potential energy of the scheduling unit based on the first potential information and the first basic potential energy.

[0086] Among them, the first sensing potential energy can be the total scheduling pressure borne by the scheduling unit at the corresponding moment when the lower-level scheduling unit has no demand.

[0087] In the embodiments of this application, after receiving the first decision information transmitted by the first lower-level scheduling unit, the potential energy sensing subunit determines the first target basic potential energy contributed by each first lower-level scheduling unit to the scheduling unit based on the first energy potential information and the first basic potential energy in the information required for each first decision; then, it accumulates and calculates all the first target basic potential energies to obtain the first sensing potential energy.

[0088] S302, determine the second sensing potential energy of the scheduling unit based on the second energy potential information, the second basic potential energy, and the second demand signature vector.

[0089] Among them, the second sensing potential energy can be the total scheduling pressure borne by the scheduling unit at the corresponding moment when the lower-level scheduling unit has a demand.

[0090] In the embodiments of this application, after receiving the second decision information transmitted by the second lower-level scheduling unit, the potential energy sensing subunit determines the second target basic potential energy contributed by each second lower-level scheduling unit to the scheduling unit based on the second energy potential information and the second basic potential energy in the information required for each second decision; determines the resonant gain of the scheduling unit based on the cosine similarity between the second demand signature vector and the capability profile vector of the scheduling unit; and finally determines the second sensing potential energy based on the second target basic potential energy and the resonant gain.

[0091] S303, determine the target sensing potential energy based on the first sensing potential energy and the second sensing potential energy.

[0092] In the embodiments of this application, the potential energy sensing subunit performs a summation operation on the first sensing potential energy and the second sensing potential energy to obtain the target sensing potential energy.

[0093] By calculating the sensing potential energy of scheduling units in different application scenarios, the accuracy of the analysis of scheduling unit capabilities is improved, which facilitates the determination of subsequent scheduling decision results.

[0094] In an exemplary embodiment, as shown in FIG4, the above step S301 "determining the first sensing potential energy of the scheduling unit based on the first potential information and the first basic potential energy" includes:

[0095] S401, for the first potential information and the first basic potential energy in the information required for each first decision, determine the first target basic potential energy contributed by each first lower-level scheduling unit to the scheduling unit based on the first potential information and the first basic potential energy.

[0096] Among them, the first target basic potential energy can be the original demand pressure borne by the scheduling unit, which originates from all its direct subordinate scheduling units that have no demand and has not been matched.

[0097] In the embodiments of this application, the potential energy sensing subunit receives first energy level information and first basic potential energy broadcast by each of the no-demand scheduling units in its set of all direct subordinate scheduling units through a network interface. Based on the received information, the importance of the first energy level information is adjusted according to a preset first adjustment coefficient to obtain the adjusted energy level; the importance of the first basic potential energy is adjusted according to a preset second adjustment coefficient to obtain the adjusted potential energy; and finally, the adjusted energy level and the adjusted potential energy are summed to obtain the first target basic potential energy.

[0098] S402, sum and calculate all the first target basic potential energies to obtain the first perceived potential energy.

[0099] In the embodiments of this application, after obtaining the first target basic potential energy of each first lower-level scheduling unit to the current scheduling unit, the potential energy sensing subunit obtains the connection weight of the first lower-level scheduling unit to the current scheduling unit, and obtains the first sensing potential energy based on the weighted summation of the first target basic potential energy and the connection weight. Optionally, the first sensing potential energy is represented by the following equation (3):

[0100] (3);

[0101] In the formula, The connection weight is a preset normalized coefficient between 0 and 1, representing the weight from the scheduling unit. to its lower-level scheduling unit Material or service flow in the scheduling unit The historical proportion in the total output. This weight determines the contribution of different lower-level scheduling units to the potential energy of the higher-level scheduling unit; It is a monotonic potential energy transfer function, whose function is to convert the state (energy potential) of the lower-level scheduling unit and the pressure it bears (basic potential energy) into a pressure signal on the upper-level scheduling unit. This serves as the primary fundamental potential energy for the first lower-level scheduling unit. This is the primary target potential energy of the first lower-level scheduling unit.

[0102] Through this backpropagation mechanism, the state change of any end scheduling unit will be quantified and transmitted to the next level, enabling the upper-level scheduling unit to know the overall demand pressure of the entire lower-level industrial chain without relying on central instructions, simply by calculating the local basic potential energy, thus improving the decision-making efficiency of the industrial chain scheduling unit.

[0103] In an exemplary embodiment, as shown in FIG5, the above step S401 "determining the first target basic potential energy contributed by each first lower-level scheduling unit to the scheduling unit based on the first potential information and the first basic potential energy" includes:

[0104] S501, adjust the importance of the first potential information according to the preset first adjustment coefficient to obtain the adjusted potential.

[0105] The preset first adjustment coefficient can be a non-negative adjustment coefficient, used to characterize the relative importance of the energy potential gap.

[0106] In the embodiments of this application, the importance of the first energy potential information is adjusted according to a preset first adjustment coefficient, that is, the preset first adjustment coefficient and the relevant parameters of the first energy potential information are multiplied to obtain the adjusted energy potential. The relevant parameters of the first energy potential information can be... , indicating the lower-level scheduling unit The "shortage" of resource buffers or inventory. When its capacity is... The lower (e.g., the power plant's coal bunker is empty), (Approaching 0), the larger the value of this item, the greater the basic pulling potential energy it generates for the upper-level scheduling unit.

[0107] S502, adjust the importance of the first basic potential energy according to the preset second adjustment coefficient to obtain the adjusted potential energy.

[0108] The preset second adjustment coefficient can be a non-negative adjustment coefficient, used to characterize the relative importance of the transferred potential energy.

[0109] In the embodiments of this application, the importance of the first basic potential energy is adjusted according to a preset second adjustment coefficient, that is, the preset second adjustment coefficient and the first basic potential energy are multiplied to obtain the adjustment potential. The adjustment potential energy can be... , indicating the lower-level scheduling unit The potential energy it already carries, derived from its subordinate level, will be proportionally... This ensures that the most urgent demand pressures at the end of the industrial chain can be transmitted to the source of the industrial chain without attenuation or according to a pre-set strategy.

[0110] S503 performs a summation operation on the adjustable potential and the adjustable energy to obtain the first target fundamental potential energy.

[0111] In the embodiments of this application, after obtaining the adjustment potential and adjustment energy, the potential energy sensing subunit performs a summation operation on the adjustment potential and adjustment energy to obtain the first target basic potential energy of the first lower-level scheduling unit for the scheduling unit. Optionally, the first target basic potential energy of the current scheduling unit is calculated using the following formula (4):

[0112] (4);

[0113] in, The preset first adjustment parameter, This is a preset second adjustment parameter used to adjust the relative importance of the energy potential gap and the transferred potential energy. This is the first energy bit information of the first lower-level scheduling unit. This serves as the primary fundamental potential energy for the first lower-level scheduling unit. This represents the first target fundamental potential energy of the first lower-level scheduling unit relative to the current scheduling unit.

[0114] Through this backpropagation mechanism, the state change of any end scheduling unit will be quantified and transmitted to the next level, enabling the upper-level scheduling unit to know the overall demand pressure of the entire lower-level industrial chain without relying on central instructions, simply by calculating the local basic potential energy, thus improving the decision-making efficiency of the industrial chain scheduling unit.

[0115] In an exemplary embodiment, as shown in FIG6, the above step S302 "determining the second sensing potential energy of the scheduling unit based on the second energy potential information, the second basic potential energy, and the second demand signature vector" includes:

[0116] S601, for the second potential information and the second basic potential energy in the information required for each second decision, determine the second target basic potential energy contributed by each second lower-level scheduling unit to the scheduling unit based on the second potential information and the second basic potential energy.

[0117] Among them, the second target basic potential energy can be the original demand pressure borne by the scheduling unit, which originates from all its directly subordinate scheduling units with demand and is not subject to matching degree correction.

[0118] In the embodiments of this application, the potential energy sensing subunit receives the second energy level information and the second basic potential energy broadcast by each second lower-level scheduling unit with demand from all its direct lower-level scheduling units through a network interface. For the second energy level information and the second basic potential energy in the information required for each second decision, the importance of the second energy level information is adjusted according to a preset first adjustment coefficient to obtain the adjusted energy level; the importance of the second basic potential energy is adjusted according to a preset second adjustment coefficient to obtain the adjusted potential energy. Finally, the adjusted energy level and the adjusted potential energy are summed to obtain the second target basic potential energy. The calculation process of the second target basic potential energy is the same as that of the first target basic potential energy, that is, it is represented by equation (4). The calculation only requires replacing the energy level information and basic potential energy of the first lower-level scheduling unit with the energy level information and basic potential energy of the second lower-level scheduling unit, which will not be elaborated here.

[0119] S602, determine the resonant gain of the scheduling unit based on the cosine similarity between the second demand signature vector and the capability profile vector of the scheduling unit.

[0120] The capability profile vector can be a multi-dimensional capability profile vector used to describe the inherent service characteristics of the scheduling unit, and can be represented by the following equation (5):

[0121] (5);

[0122] In the formula, For the index of the scheduling unit, This represents the total number of capability dimensions. Each element in the vector... It is a quantified value, corresponding to a specific technical or economic indicator. For example, It can be expressed as the maximum physical throughput (unit: tons / hour). It can be expressed as unit operating cost (unit: yuan / ton). This can be represented as the historical on-time rate of jobs (a value between 0 and 1). This capability profile vector is relatively static throughout the scheduling unit's lifecycle and is used for subsequent matching degree calculations.

[0123] The second demand signature vector can be the intrinsic attribute information of the demand initiated by the second lower-level scheduling unit, and the second demand signature vector corresponds one-to-one with the dimension of the capability profile vector. The resonant gain can be the cosine similarity between the second demand signature vector and the capability profile vector of the scheduling unit, and is used to characterize the degree of fit between the capability characteristics of the current scheduling unit and the specific demand attributes transmitted by the lower-level scheduling unit.

[0124] In the embodiments of this application, when the potential energy sensing subunit of the current scheduling unit obtains the second demand front vector of the second lower-level scheduling unit and the capability profile vector of the current scheduling unit, it uses a cosine similarity algorithm to obtain the resonance gain. Optionally, the resonance gain can be expressed by the following equation (6):

[0125] (6);

[0126] in: This represents the total number of dimensions of the vector. and These are the capability profile vectors. and requirement signature vector In the Component values ​​in each dimension. This represents the dot product of two vectors. and Let represent the Euclidean norms of the two vectors, respectively. The resonant gain is a scalar value between -1 and 1 (with a range of [0,1] when all components are non-negative). The closer this value is to 1, the stronger the current scheduling unit. Capability characteristics and requirements of the second-level scheduling unit The more compatible the attributes are, the stronger the "resonance" between them. Conversely, if the value is close to 0 or negative, it indicates that the two are mismatched or have conflicting capabilities.

[0127] S603, determine the second sensing potential energy based on the second target fundamental potential energy and the resonant gain.

[0128] Among them, the second sensing potential energy can be the total scheduling pressure borne by the scheduling unit at the corresponding moment when the lower-level scheduling unit has a demand.

[0129] In the embodiments of this application, after the potential energy sensing unit completes the calculation of the second basic potential energy and all corresponding resonant gains, it performs calculations on each second lower-level scheduling unit. The potential energy contribution is independently weighted and corrected, specifically for the energy contribution from the second lower-level scheduling unit. The contribution value of the second fundamental potential energy Through resonant gain The contribution value of the second fundamental potential energy is corrected by multiplication to obtain the second perceived potential energy. Optionally, the second perceived potential energy is expressed by the following relationship (7):

[0130] (7);

[0131] in, This represents the second fundamental potential energy of the second lower-level scheduling unit relative to the current scheduling unit. The formula represents the resonant gain of the second lower-level scheduling unit relative to the current scheduling unit. The calculation process of this formula shows that the second sensing potential energy... Instead of making a general correction to a summed basic potential energy, it makes a refined and differentiated correction to each second basic potential energy before summing them.

[0132] This synthesis mechanism achieves the following technical effect: a second-level lower-level scheduling unit. Even if extremely high fundamental potential energy is generated (i.e. (The value is large), but if its demand signature matches the current scheduling unit Capability profile mismatch (i.e., resonant gain) If the potential energy of a lower-level branch approaches zero, then the contribution of that lower-level branch to the final perceived potential energy will also approach zero. Conversely, a lower-level demand with moderate basic potential energy, if highly matched with the capability of the current scheduling unit (i.e., resonant gain), will also approach zero. If the value approaches 1, its contribution to the final perceived potential energy will be significantly amplified.

[0133] By introducing a resonant gain mechanism, the scheduling unit's decision-making not only considers the magnitude of demand pressure but also the degree of matching between its own capabilities and the nature of the demand, thus achieving precise allocation of resources within the industrial chain.

[0134] In an exemplary embodiment, as shown in FIG7, the above step S203 "generating a first scheduling decision based on the target perception potential energy" includes:

[0135] S701 determines the equity pledge amount of the scheduling unit based on the target perception potential energy of the scheduling unit and the preset virtual equity.

[0136] The preset virtual equity can be a numerical value used to quantify the historical contributions and commitment capabilities of the scheduling unit. The preset virtual equity is a non-negative scalar value, assigned an initial value during scheduling unit initialization, and dynamically increased or decreased during subsequent collaborative decision-making processes based on the scheduling unit's behavioral outcomes (e.g., successful task completion or contract signing). The equity pledge amount is a value used to assess the scheduling unit's level of enthusiasm for participating in task bidding.

[0137] In the embodiments of this application, after receiving the task message transmitted by the lower-level scheduling unit, the collaborative decision-making subunit performs a multiplication operation on the target perception potential energy calculated by the demand signature vector in the task message and the preset virtual rights to obtain the rights pledge amount of the scheduling unit in this bidding. Optionally, the rights pledge amount is represented by the following formula (8):

[0138] (8);

[0139] in, These are system parameters used to regulate bidding enthusiasm. For pre-set virtual rights, This is the target perceived potential energy of the scheduling unit. This formula shows that the higher a bidder's perceived potential energy for a task, and the more virtual equity it has accumulated historically, the greater the amount of equity it is willing to pledge for that task.

[0140] S702, based on the comparison results between the pledged equity amount and the preset pledge threshold, generate the first scheduling decision.

[0141] The first scheduling decision includes the outcome of participating in the bidding and the outcome of not participating in the bidding. The preset staking threshold can be limited according to actual needs; it is not limited here.

[0142] In the embodiments of this application, when the collaborative decision-making subunit calculates the equity pledge amount of the scheduling unit, it compares the equity pledge amount with a preset pledge threshold. When the equity pledge amount is greater than the preset pledge threshold, the first scheduling decision is for the scheduling unit to participate in the bidding; when the equity pledge amount is not greater than the preset pledge threshold, the first scheduling decision is for the scheduling unit not to participate in the bidding.

[0143] By introducing a task bidding and scheduling contract mechanism based on virtual rights, a quantitative collaborative means is provided for resource competition and risk management among distributed scheduling units, thereby improving the system's operational stability in dynamic environments.

[0144] Optionally, the collaborative decision-making subunit of the lower-level scheduling unit that issued the task message selects the scheduling unit with the highest pledged equity as the winning scheduling unit. Once the winning scheduling unit successfully completes its task as agreed, the task issuer (lower-level scheduling unit) will send a task completion confirmation signal to the winning scheduling unit. Upon receiving this signal, the collaborative decision-making subunit of the winning scheduling unit instructs the virtual equity account subunit to unlock the previously locked pledged equity. Then, according to the task agreement, their virtual rights are increased, that is... ,in As a reward for completing the task.

[0145] If the task execution of the winning scheduling unit fails (e.g., due to timeout or failure to meet quality requirements), the task issuer will send a task failure signal. The collaborative decision-making subunit of the winning scheduling unit will instruct the virtual equity account subunit to deduct the previously locked pledged equity, i.e. ,in, This refers to the amount of equity pledged.

[0146] For scheduling units that do not win the bid, their collaborative decision-making subunits can directly instruct the virtual equity account unit to unlock its pledged equity after receiving the bidding results.

[0147] In an exemplary embodiment, as shown in FIG8, the method further includes:

[0148] S801, obtain the operating parameters of the scheduling unit, and determine the third energy bit information of the scheduling unit based on the operating parameters.

[0149] The operating parameters can be inventory levels, workload, space utilization, or time utilization. The third information can be the resource saturation or buffer capacity of the scheduling unit at the corresponding moment.

[0150] In the embodiments of this application, the state-aware subunit of the scheduling unit determines the type of the scheduling unit, obtains the operating parameters and energy level calculation method corresponding to that type of scheduling unit, and calculates the operating parameters of the scheduling unit according to the energy level calculation method to obtain the current third energy level information of the scheduling unit. The determination of the operating parameters and energy level calculation method is consistent with the calculation principle of the first energy level information in step S201, and will not be described in detail here.

[0151] S802, based on the first potential information, the second potential information, the first fundamental potential energy, and the second fundamental potential energy, determine the third fundamental potential energy.

[0152] Among them, the third basic potential energy can be the scalar value of the original demand pressure borne by the scheduling unit, which originates from all its direct subordinate scheduling units and is not subject to matching degree correction.

[0153] In the embodiments of this application, the potential energy sensing subunit of the scheduling unit adjusts the importance of the first energy potential information and the second energy potential information according to a preset first adjustment coefficient to obtain the first adjustable energy potential and the second adjustable energy potential; it adjusts the importance of the first basic potential energy and the second basic potential energy according to a preset second adjustment coefficient to obtain the first adjustable potential energy and the second adjustable potential energy; it performs a weighted summation operation on all the first adjustable energy potentials and the corresponding first adjustable potential energy to obtain the first potential energy; it performs a weighted summation operation on all the second adjustable energy potentials and the corresponding second adjustable potential energy to obtain the second potential energy; and it performs a summation operation on the first potential energy and the second potential energy to obtain the third basic potential energy.

[0154] S803, when the scheduling unit's scheduling request indicates that the scheduling unit has no scheduling request, the information required for the third decision is generated based on the third potential information and the third basic potential energy, and the information required for the third decision is sent to the superior scheduling unit.

[0155] In the embodiments of this application, when the scheduling unit's scheduling request indicates that the scheduling unit has no scheduling request, there is no scheduling unit with task requirements in the distributed industry chain. The scheduling unit will not generate the corresponding task message, that is, it will not generate the first requirement signature vector corresponding to the task. In this case, it is only necessary to generate the third decision-making information based on the third energy bit information and the third basic potential energy, and send the third decision-making information to the upper-level scheduling unit. The upper-level scheduling unit can directly calculate the current sensing potential energy and the scheduling decision result based on the third decision-making information.

[0156] S804, when the scheduling unit's scheduling request indicates that the scheduling unit has a scheduling requirement, a first requirement signature vector is generated, and the fourth decision-making information is generated based on the third energy information, the third basic potential energy and the first requirement signature vector, and the fourth decision-making information is sent to the superior scheduling unit.

[0157] Among them, the first demand signature vector can be the intrinsic attribute information of the demand initiated by the lower-level scheduling unit with the demand.

[0158] In the embodiments of this application, when the scheduling unit's scheduling request indicates that the scheduling unit has a scheduling need, at least one scheduling unit with a task requirement exists in the distributed industry chain. The scheduling unit will not generate a corresponding task message; that is, it will generate a first requirement signature vector corresponding to the task. Since the first requirement signature vector is related to the capability characteristics of the upper-level scheduling unit, in this case, it is necessary to generate the fourth decision-making information based on the third energy bit information, the third basic potential energy, and the first requirement signature vector, and send the fourth decision-making information to the upper-level scheduling unit. The upper-level scheduling unit can directly calculate the current perceived potential energy and the scheduling decision result based on the fourth decision-making information.

[0159] The following section will use a specific scheduling scenario to illustrate the execution details of the system and method provided by this invention.

[0160] The supply chain in this embodiment includes the following scheduling unit 100: a lower-level power plant scheduling unit, labeled as... A port dispatch unit, marked as , it is The direct superior; two railway line dispatching units, respectively marked as and Both are The superior.

[0161] During system initialization, the parameters of each scheduling unit are configured as follows:

[0162] Capability profile vector Defined as [punctuality rate, cost coefficient, capacity], its quantified value is [0.98, 300, 50000], indicating that it has the characteristics of high punctuality rate and high cost. Its initial virtual rights... It is 1000 units. Capability profile vector Defined as [punctuality rate, cost coefficient, capacity], its quantified value is [0.85, 220, 50000], indicating that it has the characteristics of average punctuality rate and low cost. Its initial virtual rights... It is 1000 units.

[0163] The implementation steps are as follows:

[0164] Step 1: Changes in the status of lower-level units and the generation of demands.

[0165] At any moment Power plant dispatching unit Due to continuous power generation, its internal state sensing module 10 detected that the coal inventory had dropped to 10% of its maximum design capacity. The module then calculated and updated its energy potential. .

[0166] Because the energy level is lower than the preset replenishment threshold, The collaborative decision-making module 30 automatically generates an emergency coal supply transportation task. The task's requirement signature vector... The vector is set as [timeliness requirement, cost sensitivity, capacity demand], with quantified values ​​of [0.95, 0.2, 0.9]. The high timeliness requirement component (0.95) and low cost sensitivity component (0.2) objectively describe the urgency of this demand. The task is then assigned to its superior port scheduling unit. .

[0167] Step 2: Uploading potential energy and transferring tasks.

[0168] Port Dispatch Unit The potential energy sensing module 20 receives from The low-energy bit information. This module is based on... Calculate The underlying potential energy it carries, derived from the lower level.

[0169] To meet the requirements corresponding to this high potential energy, The collaborative decision-making module 30 assigns the coal transportation task to its two superior railway line dispatching units. and The broadcast included information from [source name]. Requirement signature vector .

[0170] Step 3: Differentiated potential energy perception of the upper-level scheduling unit.

[0171] Railway line dispatching unit and All potential energy sensing modules 20 receive this task. Each of them performs the calculation of the resonant gain:

[0172] The potential energy sensing module 20 uses its own capability profile =[0.98,300,50000] and the received demand signature =[0.95,0.2,0.9] for vector similarity calculation. Due to its high on-time performance, which perfectly matches the high timeliness requirements, the calculated resonant gain... This is a high value; here it is 0.92.

[0173] The potential energy sensing module 20 uses its own capability profile =[0.85,220,50000] has the same requirement signature. Calculations are performed. Due to the low match between its on-time performance characteristics and the required timeliness, the calculated resonant gain... This is a low value, here it is 0.65.

[0174] Subsequently, both components multiply their respective base potential energy contribution values ​​from the port by their resonant gains to obtain the final sensed potential energy. The calculation result is as follows: Perceived potential energy Significantly higher than Perceived potential energy .

[0175] Step 4: Bidding based on perceived potential and virtual rights.

[0176] The collaborative decision-making modules 30 of the two railway line dispatching units calculate the amount of equity required to bid for the task based on their respective perceived potential energy and virtual equity stock.

[0177] Calculate its equity pledge amount .because A higher value results in a higher calculated pledge amount.

[0178] Calculate its equity pledge amount .because The lower the value, the lower the calculated amount of collateral.

[0179] Both parties pledged their respective equity amounts and Submitted as bidding information . After comparing 30 collaborative decision-making modules, the module with the highest pledge amount is selected. The successful bidder.

[0180] Step 5: Task execution and equity account update.

[0181] After winning the bid, transportation was immediately organized, and the coal was successfully delivered to the port within the stipulated time. Upon completion of the task, Towards Send a task completion confirmation signal. The collaborative decision-making module 30 instructs its virtual equity account unit 42 to perform an equity increase operation, updating its virtual equity to... ,in It is the preset task reward value.

[0182] In an alternative outcome, if If the transportation cannot be completed on time due to unforeseen circumstances, then Send a task failure signal. Virtual equity account unit 42 will be instructed to perform an equity deduction operation, reducing the amount of virtual equity staked during the bidding process and updating it to [amount missing]. .

[0183] It should be understood that although the steps in the flowcharts of the above embodiments 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 above embodiments 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.

[0184] Based on the same inventive concept, this application also provides a distributed supply chain scheduling decision-making device for implementing the above-mentioned distributed supply chain scheduling decision-making method. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more embodiments of the distributed supply chain scheduling decision-making device provided below can be found in the limitations of the distributed supply chain scheduling decision-making method described above, and will not be repeated here.

[0185] In an exemplary embodiment, as shown in FIG9, a distributed supply chain scheduling decision-making device is provided, comprising: a first receiving module 91, a second receiving module 92, and a determining module 93, wherein:

[0186] The first receiving module 91 is used to receive the first decision-making information sent by the first subordinate scheduling unit of the scheduling unit; the first decision-making information includes the first energy potential information and the first basic potential energy of the first subordinate scheduling unit; the first subordinate scheduling unit is a scheduling unit without scheduling requirements.

[0187] The second receiving module 92 is used to receive the second decision-required information sent by the second lower-level scheduling unit of the scheduling unit; the second decision-required information includes the second energy bit information, the second basic potential energy, and the second requirement signature vector of the second lower-level scheduling unit; the second lower-level scheduling unit is a scheduling unit with scheduling requirements;

[0188] The determination module 93 is used to determine the target sensing potential energy of the scheduling unit based on the first decision-making information and the second decision-making information when it is determined that at least one of the first lower-level scheduling unit and the second lower-level unit has generated a change in energy potential according to the first energy potential information and the second energy potential information, and to generate a first scheduling decision based on the target sensing potential energy.

[0189] In one exemplary embodiment, the determining module 93 includes:

[0190] The first determining unit is used to determine the first sensing potential energy of the scheduling unit based on the first energy potential information and the first basic potential energy.

[0191] The second determining unit determines the second sensing potential energy of the scheduling unit based on the second energy potential information, the second basic potential energy, and the second demand signature vector.

[0192] The third determining unit is used to determine the target sensing potential energy based on the first sensing potential energy and the second sensing potential energy.

[0193] In an exemplary embodiment, the first determining unit includes:

[0194] The sub-unit is determined for the purpose of determining the first target basic potential energy contributed by each first lower-level scheduling unit to the scheduling unit based on the first energy potential information and the first basic potential energy in the information required for each first decision.

[0195] The calculation subunit is used to accumulate and calculate all the first target basic potential energies to obtain the first perceived potential energy.

[0196] In an exemplary embodiment, the aforementioned determining subunit is specifically used to: adjust the importance of the first energy potential information according to a preset first adjustment coefficient to obtain an adjusted energy potential; adjust the importance of the first basic potential energy according to a preset second adjustment coefficient to obtain an adjusted potential energy; and perform a summation operation on the adjusted energy potential and the adjusted potential energy to obtain the first target basic potential energy.

[0197] In an exemplary embodiment, the second determining unit described above includes:

[0198] For each second decision-making requirement, the second potential information and the second basic potential energy are used to determine the second target basic potential energy contributed by each second lower-level scheduling unit to the scheduling unit based on the second potential information and the second basic potential energy.

[0199] The resonant gain of the scheduling unit is determined based on the cosine similarity between the second demand signature vector and the capability profile vector of the scheduling unit.

[0200] The second sensing potential energy is determined based on the second target fundamental potential energy and the resonant gain.

[0201] In an exemplary embodiment, the determining module 93 is specifically used to: determine the equity pledge amount of the scheduling unit based on the target perception potential energy of the scheduling unit and the preset virtual equity; generate a first scheduling decision based on the comparison result of the equity pledge amount and the preset pledge threshold; the first scheduling decision includes the result of participating in the bidding and the result of not participating in the bidding.

[0202] In an exemplary embodiment, the above-described apparatus further includes an acquisition module, specifically configured to: acquire and determine the operating parameters of the scheduling unit, and determine the third energy level information of the scheduling unit based on the operating parameters; determine the third basic potential energy based on the first energy level information, the second energy level information, the first basic potential energy, and the second basic potential energy; when the scheduling unit's scheduling request indicates that the scheduling unit has no scheduling request, generate the third decision-making information based on the third energy level information and the third basic potential energy, and send the third decision-making information to the superior scheduling unit; when the scheduling unit's scheduling request indicates that the scheduling unit has a scheduling request, generate the first request signature vector, and generate the fourth decision-making information based on the third energy level information, the third basic potential energy, and the first request signature vector, and send the fourth decision-making information to the superior scheduling unit.

[0203] Each module in the aforementioned distributed supply chain scheduling and decision-making 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 corresponding operations of each module.

[0204] In an exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram is shown in Figure 10. The computer device includes a processor, memory, input / output interfaces (I / O), and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is connected to the system bus via the I / O interfaces. The processor of the computer device provides computing and control capabilities. The memory of the computer device includes non-volatile storage media and internal memory. The non-volatile storage media stores an operating system, computer programs, and a database. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage media. The database of the computer device stores data such as energy level information, sensed potential energy, and virtual rights of the scheduling unit. The I / O interfaces of the computer device are used for information exchange between the processor and external devices. The communication interface of the computer device is used for communication with external terminals via a network connection. When the computer program is executed by the processor, it implements a distributed supply chain scheduling decision-making method.

[0205] Those skilled in the art will understand that the structure shown in Figure 10 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 may combine certain components, or may have different component arrangements.

[0206] In one exemplary embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above-described method embodiments.

[0207] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0208] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0209] 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. When executed, the computer program 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.

[0210] 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.

[0211] 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 scheduling decision-making method for a distributed supply chain, characterized in that, The method, applied to each scheduling unit in a supply chain decision-making system, includes: receiving first decision-required information sent by a first subordinate scheduling unit of the scheduling unit; the first decision-required information includes first energy level information and first basic potential energy of the first subordinate scheduling unit; the first subordinate scheduling unit is a scheduling unit without scheduling needs; receiving second decision-required information sent by a second subordinate scheduling unit of the scheduling unit; the second decision-required information includes second energy level information, second basic potential energy, and second requirement signature vector of the second subordinate scheduling unit; the second subordinate scheduling unit is a scheduling unit with scheduling needs; when it is determined, based on the first energy level information and the second energy level information, that at least one of the first subordinate scheduling unit and the second subordinate unit has generated an energy level change, determining the target perception potential energy of the scheduling unit based on the first decision-required information and the second decision-required information, and generating a first scheduling decision based on the target perception potential energy.

2. The method according to claim 1, characterized in that, Determining the target perception potential energy of the scheduling unit based on the first decision-required information and the second decision-required information includes: determining the first perception potential energy of the scheduling unit based on the first energy level information and the first basic potential energy; determining the second perception potential energy of the scheduling unit based on the second energy level information, the second basic potential energy, and the second requirement signature vector; and determining the target perception potential energy based on the first perception potential energy and the second perception potential energy.

3. The method according to claim 2, characterized in that, The step of determining the first perceived potential energy of the scheduling unit based on the first energy potential information and the first basic potential energy includes: for each of the first decision-required information and the first basic potential energy, determining the first target basic potential energy contributed by each first lower-level scheduling unit to the scheduling unit based on the first energy potential information and the first basic potential energy; and summing all the first target basic potential energies to obtain the first perceived potential energy.

4. The method according to claim 3, characterized in that, The step of determining the first target basic potential energy contributed by each first lower-level scheduling unit to the scheduling unit based on the first energy potential information and the first basic potential energy includes: adjusting the importance of the first energy potential information according to a preset first adjustment coefficient to obtain an adjusted energy potential; adjusting the importance of the first basic potential energy according to a preset second adjustment coefficient to obtain an adjusted potential energy; and performing a summation operation on the adjusted energy potential and the adjusted potential energy to obtain the first target basic potential energy.

5. The method according to claim 2, characterized in that, The step of determining the second sensing potential energy of the scheduling unit based on the second energy level information, the second basic potential energy, and the second demand signature vector includes: for each second decision-required information, determining the second target basic potential energy contributed by each second lower-level scheduling unit to the scheduling unit based on the second energy level information and the second basic potential energy; determining the resonant gain of the scheduling unit based on the cosine similarity between the second demand signature vector and the capability profile vector of the scheduling unit; and determining the second sensing potential energy based on the second target basic potential energy and the resonant gain.

6. The method according to claim 1, characterized in that, The step of generating a first scheduling decision based on the target perceived potential energy includes: determining the equity pledge amount of the scheduling unit based on the target perceived potential energy of the scheduling unit and the preset virtual equity; generating the first scheduling decision based on the comparison result of the equity pledge amount and the preset pledge threshold; the first scheduling decision includes the result of participating in the bidding and the result of not participating in the bidding.

7. The method according to claim 1, characterized in that, The method further includes: acquiring and determining the operating parameters of the scheduling unit, and determining the third energy level information of the scheduling unit based on the operating parameters; determining the third basic potential energy based on the first energy level information, the second energy level information, the first basic potential energy, and the second basic potential energy; when the scheduling demand of the scheduling unit indicates that the scheduling unit has no scheduling demand, generating the third decision-making information based on the third energy level information and the third basic potential energy, and sending the third decision-making information to the upper-level scheduling unit; when the scheduling demand of the scheduling unit indicates that the scheduling unit has a scheduling demand, generating the first demand signature vector, and generating the fourth decision-making information based on the third energy level information, the third basic potential energy, and the first demand signature vector, and sending the fourth decision-making information to the upper-level scheduling unit.

8. A scheduling and decision-making device for a distributed industrial chain, characterized in that, The device, applied to each scheduling unit in a supply chain decision-making system, comprises: a first receiving module for receiving first decision-required information sent by a first subordinate scheduling unit of the scheduling unit; the first decision-required information includes first energy level information and first basic potential energy of the first subordinate scheduling unit; the first subordinate scheduling unit is a scheduling unit without scheduling needs; a second receiving module for receiving second decision-required information sent by a second subordinate scheduling unit of the scheduling unit; the second decision-required information includes second energy level information, second basic potential energy, and second requirement signature vector of the second subordinate scheduling unit; the second subordinate scheduling unit is a scheduling unit with scheduling needs; and a determining module for determining the target sensing potential energy of the scheduling unit based on the first decision-required information and the second decision-required information when it is determined, based on the first energy level information and the second energy level information, that at least one of the first subordinate scheduling unit and the second subordinate unit has generated an energy level change, and generating a first scheduling decision based on the target sensing potential energy.

9. 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 7.

10. 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 7.

Images