A business subject situation prediction system and method based on digital twinning

Abstract

The application discloses a kind of business principal situation prediction system and method based on digital twinning, it is related to digital twinning technical field, including, the two-way causal entropy of adjacent business principal is calculated, and the dominant direction of causal entropy is determined, and the mutation causal entropy is screened out, and the mutation transmission suspension position of mutation causal entropy stop transmission is recorded as candidate propagation fracture;Find out the missing performance stage between the contract performance event on both sides of candidate propagation fracture from candidate propagation fracture, and mark as intermediate performance event gap, intercept the intermediate contract performance event fragment of same stage and same causal entropy dominant direction with intermediate performance event gap, and the intermediate contract performance event fragment is supplemented into intermediate performance event gap, the two-way causal entropy on both sides of candidate propagation fracture is recalculated, the candidate propagation fracture of mutation causal entropy recovery transmission is deleted, and the candidate propagation fracture of mutation causal entropy still stop transmission is solidified as propagation blocking position.The application improves the accuracy of propagation blocking position determination by recalculating two-way causal entropy.

Description

Technical Field

[0001] This invention relates to the field of digital twin technology, and in particular to a business entity situation prediction system and method based on digital twins. Background Technology

[0002] With the development of supply chain digitization, business process modeling, and complex relationship computation technologies, the research object of digital twins has gradually extended from physical entities to transaction subjects, performance relationships, and business states. Existing research typically uses event-driven modeling, temporal state representation, relationship graphs, and causal inference as its theoretical foundation. It abstracts business activities such as purchase orders, production fulfillment, logistics shipment, goods acceptance, and payment settlement into performance events with time identifiers, subject attributes, and stage semantics. It uses subject nodes, supply relationships, and state sequences to represent the structural associations and state evolution between business subjects. Around multi-subject business digital twins, related research has formed technical directions such as performance event modeling, temporal association analysis, causal strength calculation, and propagation path characterization.

[0003] However, existing digital twin business situation analysis typically identifies causal transmission relationships between business entities based on recorded sequence of fulfillment events. When intermediate fulfillment events such as production completion, goods shipment, and fulfillment acceptance are missing on both sides of a candidate propagation location, the temporal continuity and semantic connection of the fulfillment phase are disrupted, and bidirectional causal entropy may suddenly drop, reverse direction, or cease transmission at the missing location. Such abnormal responses caused by missing events have similar characteristics to propagation blockages caused by real business constraints, making it difficult to accurately distinguish between recorded gaps and actual propagation blockages, which can easily lead to misjudgments of propagation blockage locations. Summary of the Invention

[0004] In view of the aforementioned existing problems, the present invention is proposed.

[0005] Therefore, this invention provides a business entity situation prediction method based on digital twins to solve the problem of difficulty in accurately distinguishing between recording gaps and actual propagation blockages.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides a method for predicting the status quo of business entities based on digital twins, which includes, Obtain contract performance events, distinguish supply relationships according to the transaction direction from the supplier to the purchaser, arrange the contract performance events under each supply relationship into performance event segments, and establish a business digital twin based on the stage changes recorded in the performance event segments; In the business digital twin, calculate the bidirectional causal entropy of adjacent business entities and determine the dominant direction of causal entropy. Select the mutation causal entropy from the bidirectional causal entropy corresponding to the dominant direction of causal entropy, and record the mutation transmission termination position where the mutation causal entropy stops propagating as the candidate propagation break point. Identify the missing performance stage between the contract performance events on both sides of the candidate propagation break and mark it as an intermediate performance event gap. Extract the intermediate contract performance event fragment that is in the same stage as the intermediate performance event gap and in the same direction of causal entropy dominance from the performance event segment. Fill the intermediate contract performance event fragment into the intermediate performance event gap. Preserve the range of the order of the positions of the participants in the bidirectional causal entropy calculation on both sides of the candidate propagation break before filling it in as the causal entropy verification window. Update the performance status in the causal entropy verification window using the contract performance events after filling in the gap and recalculate the bidirectional causal entropy on both sides of the candidate propagation break. Delete the candidate propagation break where the mutation causal entropy resumes transmission and solidify the candidate propagation break where the mutation causal entropy still stops transmission as the propagation blocking position. Based on the location of the propagation blockade, the changes in the performance status in the business digital twin are deduced along the direction of causal entropy dominance, and the transmission of causal entropy of the mutation is cut off at the propagation blockade location to form the business entity situation prediction result.

[0007] As a preferred embodiment of the business entity situation prediction method based on digital twins described in this invention, the supplier is a business entity that provides goods to the purchasing entity in accordance with the contract in a supply relationship; The procuring entity is a business entity that purchases goods from a supplier in accordance with the contract in a supply relationship; The business entity mentioned above is an operating entity that participates in contract performance and is determined as either the supplier or the purchaser based on the direction of the transaction; The contract performance events include purchase orders, supply confirmation, production completion, goods shipment, goods receipt, performance acceptance, invoice issuance, and payment. The supply relationship is a transactional relationship formed when a supplier provides goods to a purchaser in accordance with a contract.

[0008] As a preferred embodiment of the business entity situation prediction method based on digital twins described in this invention, the step of arranging contract performance events under each supply relationship into performance event segments specifically involves: Read the contract number, order number, event type, and occurrence time recorded in the contract performance event record, and group contract performance events with the same contract number and order number under the same supply relationship into the same performance event set; The contract performance events in the performance event set are arranged according to the time of occurrence. Contract performance events with the same time of occurrence are then arranged in the order of performance: purchase order, supply confirmation, production completion, goods shipment, goods receipt, performance acceptance, invoice issuance and payment, to determine the chronological order of each contract performance event. By sequentially connecting the contract performance events according to their order of appearance, a performance event segment is formed.

[0009] As a preferred embodiment of the business entity situation prediction method based on digital twins described in this invention, the establishment of the business digital twin specifically involves: The supplier and purchaser are mapped to digital twin business entities respectively, and adjacent digital twin business entities are connected by supply relationships to form a digital twin entity chain; At the stage change position of the performance event segment, the performance status of adjacent digital twin business entities before and after the stage change position is recorded respectively, forming a performance stage boundary state piece; Based on their sequential order, the boundary status slices of the performance phase are written into the main chain of the digital twin, thus completing the establishment of the business digital twin.

[0010] As a preferred embodiment of the business entity situation prediction method based on digital twins described in this invention, wherein: the term "candidate propagation breakpoint" is specifically defined as follows: Arrange the performance status of the digital twin business entities at both ends of the same supply relationship according to their sequential positions to form a dual-end performance status sequence; In the business digital twin, the preceding and subsequent performance states in the two-end performance state sequence are paired bidirectionally to measure the uncertainty reduction from the supplier to the purchaser and the uncertainty reduction from the purchaser to the supplier, thus forming a bidirectional causal entropy. By comparing the uncertainty reduction of bidirectional causal entropy in two transmission directions, the transmission direction in which the uncertainty reduction continuously exceeds the opposite direction is defined as the causal entropy dominant direction. The transmission direction is the direction of the change in performance status between adjacent digital twin business entities along the supply entity to the purchasing entity and along the purchasing entity to the supply entity. Select the bidirectional causal entropy that is higher than both the preceding and following positions from the bidirectional causal entropy corresponding to the direction of causal entropy dominance to form the mutation causal entropy; The transmission of mutation causal entropy is traced along the digital twin entity chain. The business entity connection position where the transmission of mutation causal entropy stops in the next supply relationship is determined as the mutation transmission termination position and output as a candidate propagation break point.

[0011] As a preferred embodiment of the business entity situation prediction method based on digital twins described in this invention, wherein: the marker is an intermediate fulfillment event gap, specifically, According to the order of stage changes recorded in the performance event segment, arrange the performance stages between the contract performance events before and after the candidate propagation break point to form the performance stages to be verified. The contract performance events recorded on both sides of the candidate propagation gap are checked one by one during the pending performance stage. The performance stages occupied by existing contract performance events are removed, and the performance stages not occupied by contract performance events are retained. The retained performance stages are marked as intermediate performance event gaps.

[0012] As a preferred embodiment of the business entity situation prediction method based on digital twins described in this invention, the step of filling the intermediate contract performance event fragment into the intermediate performance event gap specifically involves: The interception interval is defined by the performance stage corresponding to the gap in the intermediate performance event. The contract performance event at the beginning of the interception interval in the performance event segment is recorded as the interception start event, and the contract performance event at the end of the interception interval is recorded as the interception termination event. Between the start and end events of the interception, the business entity direction of each contract performance event is marked according to the transaction direction from the supplier to the purchaser. Contract performance events whose business entity direction is opposite to the direction of causal entropy dominance are eliminated, and the remaining contract performance events are arranged into candidate performance event segments. The contract performance events in the candidate performance event segment are projected to the corresponding positions of the intermediate performance event gaps, and the contract performance events that continuously cover the missing performance stages are extracted to form intermediate contract performance event segments; According to the order of the missing performance stages within the intermediate performance event gaps, the contract performance events in the intermediate contract performance event segments are filled into the intermediate performance event gaps one by one.

[0013] As a preferred embodiment of the business entity situation prediction method based on digital twins described in this invention, the recalculation of the bidirectional causal entropy on both sides of the candidate propagation breakpoint specifically involves: Before the intermediate contract performance event fragment is added, the first and last positions in the order of participating in the bidirectional causal entropy calculation on both sides of the candidate propagation breakpoint are selected respectively to determine the front and back boundaries of the causal entropy verification window. After filling the intermediate contract performance event fragment into the intermediate performance event gap, keep the front and back boundaries of the causal entropy verification window unchanged, and classify the filled contract performance events into the causal entropy verification window according to the newly determined order of positions. Update the performance stage boundary state pieces corresponding to the filled positions, and rearrange the supplier's performance status and the purchaser's performance status in the causal entropy verification window to form an updated double-end performance status sequence. From the updated two-way performance state sequence, select the valid performance state combination without missing or conflicting markers, redetermine the conditional probability, and measure the uncertainty reduction from the supplier to the purchaser and the uncertainty reduction from the purchaser to the supplier respectively to form the recalculated two-way causal entropy.

[0014] As a preferred embodiment of the business entity situation prediction method based on digital twins described in this invention, the step of forming the business entity situation prediction result specifically involves: Taking the performance status of the business entity at the starting position of the mutation causal entropy transmission as the starting point of transmission, the performance status change corresponding to the mutation causal entropy is transmitted level by level according to the direction of causal entropy dominance, forming a performance status transmission trajectory. The starting position of the mutation causal entropy transmission is the position of the business entity where the mutation causal entropy first appears along the direction of causal entropy dominance and begins to be transmitted to the next supply relationship. Write the performance status transmission trajectory into the business digital twin, and rewrite the performance stage of each business entity before the propagation blockage point in turn to form the performance status segment before the blockage. At the propagation blocking point, the change in the performance status corresponding to the mutation causal entropy is truncated, and the performance stage of the business entity that has not been rewritten after the propagation blocking point is arranged as the post-blocking performance status segment. The business entity status prediction result is formed by combining the pre-blocking and post-blocking performance status segments according to the sequential positions of the digital twin entity chain.

[0015] Secondly, this invention provides a business entity situation prediction system based on digital twins, comprising: The acquisition module is used to acquire contract performance events, distinguish supply relationships according to the transaction direction from the supplier to the purchaser, arrange the contract performance events under each supply relationship into performance event segments, and establish a business digital twin based on the stage changes recorded in the performance event segments. The filtering module is used to calculate the bidirectional causal entropy of adjacent business entities in the business digital twin, determine the dominant direction of causal entropy, filter out the mutation causal entropy from the bidirectional causal entropy corresponding to the dominant direction of causal entropy, and record the mutation transmission termination position where the mutation causal entropy stops propagating as the candidate propagation breakpoint. The interception module is used to identify the missing performance stage between the contract performance events on both sides of the candidate propagation break and mark it as an intermediate performance event gap. It intercepts the intermediate contract performance event fragments that are in the same stage and in the same direction of causal entropy dominance as the intermediate performance event gap from the performance event segment. It fills the intermediate contract performance event fragments into the intermediate performance event gap, retains the range of the order of the positions of the participants in the bidirectional causal entropy calculation on both sides of the candidate propagation break before filling in the fragments as the causal entropy verification window, updates the performance status in the causal entropy verification window with the contract performance events after filling in the fragments, and recalculates the bidirectional causal entropy on both sides of the candidate propagation break. It deletes the candidate propagation break where the mutation causal entropy resumes transmission and fixes the candidate propagation break where the mutation causal entropy still stops transmission as the propagation blocking position. The prediction module is used to infer the changes in the performance status of the business digital twin based on the propagation blockage location and along the direction of causal entropy dominance, and to cut off the transmission of sudden causal entropy at the propagation blockage location to form the business entity situation prediction result.

[0016] The beneficial effects of this invention are as follows: By recalculating the bidirectional causal entropy, the accuracy of determining the propagation interruption location is improved. After the intermediate contract performance event fragment is filled into the gap in the intermediate performance event, the originally disconnected performance stages on both sides of the candidate propagation interruption are continuously connected, enabling the performance status changes between adjacent digital twin business entities to re-establish the corresponding relationship according to the complete stage sequence. The recalculated bidirectional causal entropy can more accurately reflect the transmission continuity in the direction of causal entropy dominance. When the abrupt causal entropy resumes transmission to the next supply relationship, the candidate propagation interruption is deleted; when the abrupt causal entropy still stops transmission, the candidate propagation interruption is solidified as the propagation interruption location. Thus, the absence of performance events in the digital twin no longer directly forms the erroneous propagation boundary, the identification range of the propagation interruption location is more convergent, and the consistency, stability, and reliability of the interruption determination and the prediction of the business entity's situation are improved. Attached Figure Description

[0017] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0018] Figure 1 This is a flowchart of a business entity status prediction method based on digital twins.

[0019] Figure 2 This is a schematic diagram of a business entity situation prediction system based on digital twins.

[0020] Figure 3 A flowchart for creating a business digital twin.

[0021] Figure 4 A flowchart for filling gaps in intermediate performance events and determining the locations for blocking their spread. Detailed Implementation

[0022] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0023] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0024] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places in this specification does not necessarily refer to the same embodiment, nor is it a single or selective embodiment that is mutually exclusive with other embodiments.

[0025] Reference Figures 1-4 As one embodiment of the present invention, this embodiment provides a business entity situation prediction method based on digital twins, including the following steps: S1. Obtain contract performance events, distinguish supply relationships according to the transaction direction from the supplier to the purchaser, arrange the contract performance events under each supply relationship into performance event segments, and establish a business digital twin based on the stage changes recorded in the performance event segments.

[0026] S1.1. Retrieve the contract number, order number, supplier, purchaser, event type, and occurrence time from the contract management database, order management records, supply confirmation records, production completion records, logistics shipment records, goods receipt records, performance acceptance records, invoice issuance records, and payment records. Then, categorize the purchase order records, supply confirmation records, production completion records, goods shipment records, goods receipt records, performance acceptance records, invoice issuance records, and payment records into their respective contract performance event types. Retain the contract number, order number, supplier, purchaser, and occurrence time recorded in each record to form pending contract performance events.

[0027] Using contract number, order number, supplier, and purchaser as corresponding conditions, each contract performance event to be verified is checked. When the contract number, order number, and supply relationship between the supplier and purchaser are consistent, the corresponding contract performance event to be verified is retained. When the event type, occurrence time, and recorded information of duplicate records are consistent, they are merged into one item. When there are differences in duplicate records, a conflict mark is retained. When the contract number, order number, supply relationship between the supplier and purchaser cannot be formed, or occurrence time is missing, the corresponding contract performance event to be verified is excluded and a missing mark is retained, thus forming a contract performance event.

[0028] Among them, the procuring entity is the business entity that purchases goods from the supplier in accordance with the contract in the supply relationship; the business entity is the operating entity that participates in the performance of the contract and is determined as the corresponding party between the supplier and the procuring entity according to the direction of the transaction; the contract performance events include purchase orders, supply confirmation, production completion, goods shipment, goods receipt, performance acceptance, invoice issuance and payment; the supply relationship is the transactional relationship formed by the supplier providing goods to the procuring entity in accordance with the contract.

[0029] S1.2 Read the contract number, order number, event type, and occurrence time recorded in the contract performance event record. Group contract performance events with the same contract number and order number under the same supply relationship into the same performance event set. Arrange the contract performance events in the performance event set according to the occurrence time, and arrange the contract performance events with the same occurrence time according to the performance order of purchase order, supply confirmation, production completion, goods shipment, goods receipt, performance acceptance, invoice issuance, and payment, and determine the sequential order of each contract performance event. Use the sequential order to connect each contract performance event in turn to form a performance event segment.

[0030] S1.3. Map the supplier and purchaser to digital twin business entities respectively, and connect adjacent digital twin business entities using supply relationships to form a digital twin entity chain. Specifically, The supplier and purchaser fields in the contract performance events are expanded separately. Contract performance events with the same supplier field share the same digital twin business entity, and contract performance events with the same purchaser field share the same digital twin business entity. When the same business entity is in the supplier position and purchaser position in different supply relationships, only one digital twin business entity is established, and different transaction positions are retained in the corresponding supply relationship. Suppliers and purchasers with the same contract number and order number are connected according to the transaction direction from supplier to purchaser. The same connection that appears repeatedly is retained once, and connections with any difference in contract number or order number are retained separately, so that each digital twin business entity is continuously connected along the supply relationship to form a digital twin entity chain.

[0031] S1.4. At the stage change position of the performance event segment, record the performance status of adjacent digital twin business entities before and after the stage change position to form a performance stage boundary state slice, specifically, The sequence of event types for adjacent contract performance events in the performance event segment is determined by the order in which the event types transition from the previous performance stage to the next performance stage. Purchase orders, goods receipt, performance acceptance, and payment are associated with the purchasing entity, while supply confirmation, production completion, goods shipment, and invoice issuance are associated with the supply entity. Using the occurrence time of the previous contract performance event as the pre-stage change state point, find the latest-occurring contract performance events for the supplier and the purchaser from the preceding contract performance events. Record the event type, occurrence time, and chronological order as the pre-stage change performance state for the supplier and the purchaser, respectively. Using the occurrence time of the next contract performance event as the post-stage change state point, find the latest-occurring contract performance events for the supplier and the purchaser, respectively. Record the event type, occurrence time, and chronological order as the post-stage change performance state for the supplier and the purchaser, respectively.

[0032] If no corresponding contract performance event exists for any digital twin business entity before the corresponding state time point, the corresponding performance state is recorded as not having reached the corresponding performance stage; if the corresponding contract performance event has a missing or conflicting marker, the missing or conflicting marker is retained in the corresponding performance state; the contract number, order number, supplier, purchaser, and supply relationship are combined with the performance state before and after the stage change of each supplier and purchaser to form the same state record, and arranged in chronological order to form a performance stage boundary state piece.

[0033] Among them, the performance status is the most recent performance stage that the digital twin business entity has reached up to the corresponding status point in time, as well as the occurrence time, sequential position and record mark corresponding to the most recent performance stage, which is used to characterize the respective stage progress of the supplier and the purchaser in the contract performance process.

[0034] S1.5. Based on the sequential arrangement, write the boundary state slices of the performance phase into the digital twin main chain to complete the establishment of the business digital twin. Specifically, The corresponding connections in the digital twin entity chain are located based on the supplier, purchaser, and supply relationship recorded in the boundary state slice of the performance stage. The boundary state slices of the performance stage with the same contract number and order number are placed in the same connection and arranged from front to back according to their order of position. When the overlapping performance states of the same digital twin business entity in adjacent boundary state slices of the performance stage are consistent, the overlapping performance states are merged and retained once. When the overlapping performance states are inconsistent, they are retained separately according to their order of position and the corresponding conflict markers are continued. The unilateral performance states of the first and last positions of the performance event segment maintain their original order of position.

[0035] The performance status formed by actual contract performance events is marked as the actual performance status, and the occurrence time of the corresponding contract performance event is recorded as the status update time. When a new contract performance event occurs, the corresponding performance event segment and the supply relationship in the digital twin entity chain are located by contract number, order number, supplier and purchaser. The order of the segments is re-determined according to the occurrence time of the new contract performance event, and a new performance stage boundary state piece affected by the new contract performance event is formed. The newly formed performance stage boundary state piece is written into the corresponding digital twin entity chain, while the actual performance status before the update and the corresponding status update time are retained, so that the digital twin entity chain continuously records the performance status changes caused by actual contract performance events.

[0036] For subsequent performance status projections, projection status markers are added to the corresponding digital twin business entity positions and recorded separately from the actual performance status, ensuring that the projection performance status does not cover the actual performance status formed by actual contract performance events. The business digital twin is composed of the digital twin business entity, the continuously updated digital twin entity chain, and the performance stage boundary status pieces that record the actual performance status and projection performance status respectively. This is used to record the supply relationship between the supplier and the purchaser, the sequence of contract performance events, changes in actual performance status, and changes in projection performance status.

[0037] S2. Calculate the bidirectional causal entropy of adjacent business entities in the business digital twin, determine the dominant direction of causal entropy, select the mutation causal entropy from the bidirectional causal entropy corresponding to the dominant direction of causal entropy, and record the mutation transmission termination position where the mutation causal entropy stops propagating as the candidate propagation break point.

[0038] S2.1 Arrange the performance statuses of the digital twin business entities at both ends of the same supply relationship according to their sequential positions to form a two-end performance status sequence; in the business digital twin, pair the preceding performance status with the subsequent performance status in the two-end performance status sequence bidirectionally, measure the uncertainty reduction from the supplier to the purchaser, and the uncertainty reduction from the purchaser to the supplier, forming a bidirectional causal entropy, specifically... The boundary state pieces of the performance stage at both ends of the same supply relationship in the business digital twin are unfolded according to their sequential positions. The performance status of the supplier and the performance status of the purchaser recorded in each performance stage boundary state piece are placed in the same sequential position, so that the performance status of the supplier and the performance status of the purchaser under the same sequential position are kept in correspondence. When a performance status on one side is missing, the missing mark is continued. Performance status with conflict mark retains the corresponding sequential position but does not participate in the measurement of uncertainty reduction, forming a two-end performance status sequence.

[0039] The current sequential position is defined as the position in the sequence of consecutive performance states that can be obtained from the previous and next performance states. The causal entropy calculation window is then defined with the current sequential position as the center. From the causal entropy calculation window, consecutive performance states that are not marked with missing or conflicting tags are selected. The performance states of the supplier and the purchaser in the previous sequential position are respectively defined as the previous performance states of the supplier and the purchaser. The performance states of the supplier and the purchaser in the next sequential position are respectively defined as the subsequent performance states of the supplier and the purchaser, thus forming the effective performance state combination corresponding to the current sequential position.

[0040] The proportion of the occurrence of the same valid performance state combination within the statistical causal entropy calculation window to the total number of valid performance state combinations is used to determine the conditional probability corresponding to the current sequential position. The uncertainty reduction from the supplier to the purchaser and the uncertainty reduction from the purchaser to the supplier are measured respectively. The uncertainty reductions in the two transmission directions are combined to form the bidirectional causal entropy corresponding to the current sequential position. The current sequential position is moved one by one along the two-end performance state sequence, and the causal entropy calculation window is moved synchronously to form a bidirectional causal entropy corresponding to each valid sequential position.

[0041] The uncertainty reduction from supplier to purchaser is expressed as: ; The uncertainty reduction from the procuring entity to the supplying entity is expressed as: ; The expression for conditional information entropy is: ; ; In the formula, This represents the reduction in uncertainty regarding the supplier's role in the purchaser's role. Indicates the supplier. Indicates the procuring entity. Indicates the order of precedence. Indicates the subsequent sequential order. This indicates the previous performance status of the procuring entity. Indicates the subsequent performance status of the procuring entity. Represents conditional information entropy. This indicates the preceding performance status of the business entity whose subsequent performance status is being measured. This indicates the subsequent performance status of the business entity whose uncertainty is being measured. Represents the performance status variable The specific prior performance status obtained. Represents the performance status variable The specific subsequent performance status obtained, Indicates the performance status Performance status The probability of them appearing together Indicates the performance status Performance status when it occurs The conditional probability of occurrence This indicates that all valid performance status combinations are summed. Represents the logarithm to the base 2. This represents the reduction in uncertainty regarding the procuring entity's connection to the supply entity. This indicates the supplier's previous performance status. This indicates the supplier's subsequent performance status. This indicates the prior performance status of adjacent digital twin business entities. Represents the performance status variable The obtained prior performance status of adjacent digital twin business entities, Indicates the performance status Performance status Performance status The probability of them appearing together Indicates the performance status Performance status Performance status when both appear together The conditional probability of occurrence.

[0042] In this invention, the causal entropy calculation window consists of the current sequential position and three consecutive sequential positions before and after the current sequential position, for a total of seven consecutive sequential positions. When the current position is at the beginning or end of a two-way performance state sequence and it is impossible to obtain all seven consecutive sequential positions, the window extends from the current sequential position toward the side where consecutive performance states can be obtained, maintaining the causal entropy calculation window containing seven consecutive sequential positions. When there are at least five valid performance state combinations without missing or conflicting markers within the causal entropy calculation window, the bidirectional causal entropy is measured for the current sequential position. When there are fewer than five valid performance state combinations, the current sequential position does not participate in the bidirectional causal entropy measurement. When calculating conditional information entropy, the corresponding items of performance state combinations with a co-occurrence probability of zero are treated as zero and are not included in the logarithmic operation.

[0043] S2.2. Compare the uncertainty reduction of the bidirectional causal entropy in both transmission directions. The direction in which the uncertainty reduction continuously exceeds that in the opposite direction is defined as the causal entropy-dominant direction. Specifically, The uncertainty reduction from the supplier to the purchaser corresponding to the same sequential position is compared with the uncertainty reduction from the purchaser to the supplier. If the uncertainty reduction from the supplier to the purchaser exceeds the uncertainty reduction from the purchaser to the supplier, the transmission direction from the supplier to the purchaser is indicated. If the uncertainty reduction from the purchaser to the supplier exceeds the uncertainty reduction from the purchaser to the purchaser, the transmission direction from the purchaser to the supplier is indicated. If the uncertainty reductions in the two transmission directions are the same, the equivalent position is retained, and no transmission direction is assigned.

[0044] Connecting bidirectional causal entropies that maintain the same transmission direction among adjacent sequential positions forms a continuous transmission segment. The continuous transmission segment is interrupted at the position where the equivalent position and the transmission direction are reversed. A continuous transmission segment is formed when there are three consecutive sequential positions that maintain the same transmission direction; if there are fewer than three consecutive sequential positions, a continuous transmission segment is not formed. When only one transmission direction forms a continuous transmission segment within the same performance event segment, the corresponding transmission direction is recorded as the causal entropy dominant direction. When both transmission directions form continuous transmission segments, the transmission direction that covers more sequential positions is recorded as the causal entropy dominant direction. If the number of sequential positions covered is the same, a causal entropy dominant direction is not formed.

[0045] Among them, the transmission direction is the direction of the change in the performance status between adjacent digital twin business entities, from the supplier to the purchaser and from the purchaser to the supplier.

[0046] S2.3. From the bidirectional causal entropy corresponding to the dominant causal entropy direction, select the bidirectional causal entropy that is simultaneously higher than the preceding and following positions to form the mutation causal entropy. Specifically, From the continuous transmission segment corresponding to the dominant causal entropy direction, bidirectional causal entropies that simultaneously possess both the preceding and following sequential positions are selected sequentially. The uncertainty reduction of the current sequential position is compared with the uncertainty reduction of the preceding and following sequential positions, respectively. If the uncertainty reduction of the current sequential position exceeds the uncertainty reduction of both preceding and following sequential positions, the current sequential position and its corresponding transmission direction are retained, forming abrupt causal entropy. If the uncertainty reduction of the current sequential position does not simultaneously exceed the uncertainty reduction of both preceding and following sequential positions, it is not considered abrupt causal entropy. Bidirectional causal entropies with missing or conflict markers at the first and last positions of the continuous transmission segment, as well as adjacent positions, are not included in the abrupt causal entropy selection.

[0047] S2.4. Tracing the transmission of mutation causal entropy along the digital twin main chain, the business main connection position where the transmission of mutation causal entropy stops in the next supply relationship is determined as the mutation transmission termination position, and is output as a candidate propagation breakpoint. Specifically, Starting from the connection position of the digital twin business entity corresponding to the mutation causal entropy, the system moves along the direction of causal entropy dominance to the adjacent digital twin business entity. It then enters the next supply relationship through the shared digital twin business entity in the adjacent supply relationship. The system then compares and connects the performance status of the shared digital twin business entity in the previous supply relationship after the stage change with the performance status before the stage change in the next supply relationship, which occurred later than the occurrence time corresponding to the mutation causal entropy and is the first in the sequence.

[0048] The business entity connection position after the stage change is consistent with the performance status before the stage change, and the uncertainty reduction in the same transmission direction in the next supply relationship exceeds the uncertainty reduction in the opposite transmission direction, is taken as the continued transmission position and tracked along the digital twin entity chain. When the performance status after the stage change is inconsistent with the performance status before the stage change, the next supply relationship lacks a connectable performance status, and the uncertainty reduction in the same transmission direction in the next supply relationship does not exceed the uncertainty reduction in the opposite transmission direction, the business entity connection position shared by the previous supply relationship and the next supply relationship is determined as the mutation transmission termination position and output as a candidate propagation break point. When there is no next supply relationship at the end of the digital twin entity chain, no candidate propagation break point is formed.

[0049] S3. Identify the missing performance stage between the contract performance events on both sides of the candidate propagation break and mark it as an intermediate performance event gap. Extract the intermediate contract performance event fragment that is in the same stage as the intermediate performance event gap and in the same direction of causal entropy dominance from the performance event segment. Fill the intermediate contract performance event fragment into the intermediate performance event gap. Preserve the range of the order of the positions of the participants in the bidirectional causal entropy calculation on both sides of the candidate propagation break before filling it in as the causal entropy verification window. Update the performance status in the causal entropy verification window using the contract performance events after filling in the gap and recalculate the bidirectional causal entropy on both sides of the candidate propagation break. Delete the candidate propagation break where the mutation causal entropy resumes transmission. Solidify the candidate propagation break where the mutation causal entropy still stops transmission as the propagation blocking position.

[0050] S3.1. According to the order of stage changes recorded in the performance event segment, arrange the performance stages between the contract performance events before and after the candidate propagation break point to form the performance stages to be verified, specifically as follows: From the performance event segments corresponding to the supply relationship on the front side of the candidate propagation break, select the contract performance event that occurs latest and is ranked last in sequence, and record it as the front boundary contract performance event; from the performance event segments corresponding to the supply relationship on the back side of the candidate propagation break, select the contract performance event that occurs earliest and is ranked first in sequence, and record it as the back boundary contract performance event; arrange the performance stages from the performance stage to the performance stage of the back boundary contract performance event in the order of purchase order, supply confirmation, production completion, goods shipment, goods receipt, performance acceptance, invoice issuance and payment, to form the performance stages to be verified, which are used to limit the performance stages that need to be verified on both sides of the candidate propagation break.

[0051] S3.2. Verify the contract performance events actually recorded on both sides of the candidate propagation gap against the pending performance stage, remove performance stages occupied by existing contract performance events, retain performance stages not occupied by contract performance events, and mark the retained performance stages as intermediate performance event gaps. Specifically, The performance stages in the pending performance phase are matched one by one with the event types recorded in the performance event segments before and after the candidate propagation break. When the event type of a contract performance event is consistent with the performance stage in the pending performance phase and does not have a conflict mark, the performance stage already occupied by the contract performance event is removed. When the event type does not correspond to the contract performance event, the corresponding contract performance event has a missing mark, or the corresponding contract performance event has a conflict mark, the corresponding performance stage is retained. The consecutively retained performance stages are connected in their original order and marked as intermediate performance event gaps.

[0052] S3.3. Define the cutoff interval based on the performance stage corresponding to the gap in the intermediate performance event. Record the contract performance event at the beginning of the cutoff interval as the cutoff start event, and the contract performance event at the end of the cutoff interval as the cutoff end event. Specifically, Using the first missing performance stage in the intermediate performance event gap as the starting point of the interception interval and the last missing performance stage in the intermediate performance event gap as the ending point of the interception interval, the contract performance events with the same event type as the starting point of the interception interval and the contract performance events with the same event type as the ending point of the interception interval are searched from the performance event segments in the business digital twin. When the event types are consistent and the contract performance event corresponding to the starting point is located before the contract performance event corresponding to the ending point, the contract performance event corresponding to the starting point is recorded as the interception start event and the contract performance event corresponding to the ending point is recorded as the interception termination event. When the intermediate performance event gap contains only one missing performance stage, the same contract performance event is recorded as both the interception start event and the interception termination event.

[0053] S3.4. Between the start and end events of the interception, mark the business entity direction of each contract performance event according to the transaction direction from the supplier to the purchaser. Eliminate contract performance events whose business entity direction is opposite to the dominant causal entropy direction. Arrange the remaining contract performance events into a candidate performance event segment, specifically... The purchase order, goods receipt, performance acceptance, and payment between the start and end events of the interception are marked as the business entity pointing from the purchaser to the supplier. The supply confirmation, production completion, goods shipment, and invoice issuance are marked as the business entity pointing from the supplier to the purchaser. When the business entity pointing of a contract performance event is consistent with the direction of causal entropy dominance, the corresponding contract performance event is retained. When the business entity pointing of a contract performance event is opposite to the direction of causal entropy dominance, the corresponding contract performance event is removed. When there are no contract performance events in the same direction between the start and end events of the interception, the corresponding interception interval is not retained. The remaining contract performance events are arranged into candidate performance event segments according to their occurrence time and chronological order.

[0054] S3.5. Project the contract performance events in the candidate performance event segment to the corresponding positions of the intermediate performance event gaps, extract the contract performance events that continuously cover the missing performance stages, and form intermediate contract performance event segments, specifically as follows: Contract performance events in candidate performance event segments are assigned to the missing performance stages with the same name within the intermediate performance event gaps according to event type. When multiple contract performance events correspond to the same missing performance stage, the contract performance event with the earliest chronological order and without conflict markers is retained. When each missing performance stage within the intermediate performance event gap has a corresponding contract performance event, and the chronological order of the corresponding contract performance events is consistent with the chronological order of the missing performance stages, continuous contract performance events are extracted from the contract performance event corresponding to the first missing performance stage to the contract performance event corresponding to the last missing performance stage. When there is a missing performance stage that does not correspond to a contract performance event or when the chronological order of the contract performance events is inconsistent with the chronological order of the missing performance stages, the corresponding candidate performance event segment is excluded, forming an intermediate contract performance event segment that continuously covers the missing performance stages.

[0055] S3.6. According to the order of the missing performance stages within the intermediate performance event gaps, fill in the missing contract performance events in the intermediate contract performance event segments one by one into the intermediate performance event gaps. Specifically, Using the direction of causal entropy dominance as the direction of contract performance events, the position for each missing performance stage within the gap of intermediate performance events is determined. From the original contract performance events on both sides of the candidate propagation break, the contract performance event located before and closest to the position to be filled is found. The found contract performance event is used as the attribution reference event, and the contract number, order number, supplier, and purchaser recorded in the attribution reference event are determined as the contract number, order number, supplier, and purchaser of the corresponding filled contract performance event. Thus, the filled contract performance event belongs to the performance event segment directly connected to the position to be filled in the direction of causal entropy dominance.

[0056] The number of contract performance events that need to be added is counted between each group of adjacent original contract performance events. The occurrence time of the previous original contract performance event is used as the starting point of the time interval, and the occurrence time of the next original contract performance event is used as the ending point of the time interval. The time interval is divided into equal intervals according to the number of contract performance events that need to be added plus an interval. According to the order of the missing performance stages in the gap of the intermediate performance events, the intermediate time corresponding to each division position is recorded as the occurrence time of the corresponding contract performance event to be added, so that the occurrence time of the added contract performance event is located between the occurrence times of the previous and next original contract performance events.

[0057] Contract performance events of the same type in the intermediate contract performance event segment are sequentially placed into the corresponding missing performance stages. The event types and sequential order of each contract performance event in the intermediate contract performance event segment are preserved. The sequential order of each supplementary contract performance event is re-determined according to the determined occurrence time. The supplementary contract performance events are marked as supplementary. The original contract performance events on both sides of the intermediate performance event gap remain unchanged. This completes the filling of the intermediate performance event gap in the intermediate contract performance event segment.

[0058] S3.7 Recalculate the bidirectional causal entropy on both sides of the candidate propagation breakpoint, delete the candidate propagation breakpoints where the mutation causal entropy resumes propagation, and solidify the candidate propagation breakpoints where the mutation causal entropy still stops propagating as propagation blocking locations. Specifically, Before the intermediate contract performance event fragment is added, the first and last positions in the order of participating in the bidirectional causal entropy calculation on both sides of the candidate propagation breakpoint are selected respectively to determine the front and back boundaries of the causal entropy verification window. After filling the intermediate contract performance event fragment into the intermediate performance event gap, keep the front and back boundaries of the causal entropy verification window unchanged, and classify the filled contract performance events into the causal entropy verification window according to the newly determined order of positions. Update the performance stage boundary state pieces corresponding to the filled positions, and rearrange the supplier's performance status and the purchaser's performance status in the causal entropy verification window to form an updated double-end performance status sequence. From the updated two-way performance status sequence, select the valid performance status combination without missing or conflicting markers. Re-determine the conditional probability according to the minimum number of valid samples and the zero-probability processing method used in the causal entropy calculation window. Measure the uncertainty reduction from the supplier to the purchaser and the uncertainty reduction from the purchaser to the supplier respectively to form the recalculated two-way causal entropy.

[0059] From the recalculated bidirectional causal entropy, check whether the dominant direction of causal entropy remains continuous before and after the candidate propagation break. If the uncertainty reduction in the same transmission direction in the next supply relationship after the candidate propagation break exceeds the uncertainty reduction in the opposite transmission direction, it is determined that the mutation causal entropy has resumed transmission and the candidate propagation break is deleted. If the uncertainty reduction in the same transmission direction in the next supply relationship after the candidate propagation break does not exceed the uncertainty reduction in the opposite transmission direction, or if the dominant direction of causal entropy before and after the candidate propagation break is reversed, it is determined that the mutation causal entropy has stopped transmission and the candidate propagation break is fixed as the propagation blocking position.

[0060] S4. Based on the propagation blockage location, extrapolate the changes in the performance status in the business digital twin along the direction of causal entropy dominance, and cut off the transmission of sudden causal entropy at the propagation blockage location to form the business entity situation prediction result.

[0061] S4.1. Taking the performance status of the business entity at the starting point of the mutation causal entropy transmission as the starting point, the performance status change corresponding to the mutation causal entropy is transmitted level by level according to the direction of causal entropy dominance, forming a performance status transmission trajectory. The starting point of the mutation causal entropy transmission is the business entity position where the mutation causal entropy first appears along the direction of causal entropy dominance and begins to be transmitted to the next supply relationship. Specifically... The latest actual contract performance event in the business digital twin is determined as the prediction time point. The position of the digital twin business entity where the mutation causal entropy first appears along the direction of causal entropy dominance is taken as the starting position of mutation causal entropy transmission. The performance state before the stage change and the performance state after the stage change are read from the performance stage boundary state slice corresponding to the starting position of mutation causal entropy transmission. The performance state after the stage change is determined as the predicted performance state of the corresponding digital twin business entity after the prediction time point.

[0062] From the historical performance event segments stored in the business digital twin, the number of times the same performance status before the stage change transitions to the performance status after each stage change is counted. The proportion of the number of times the performance status transitions to the current stage after the stage change to the total number of subsequent performance status changes is determined as the state transition probability. The average time interval between the contract performance events corresponding to the transition from the same performance status before the stage change to the current stage after the stage change is also counted to determine the expected transition duration.

[0063] The predicted performance status is connected to the next supply relationship along the direction of causal entropy dominance. At the digital twin business entity position shared by adjacent supply relationships, the performance status after the stage change of the previous supply relationship is connected to the performance status before the stage change of the next supply relationship. When the connection is consistent, the corresponding predicted performance status, state transition probability and expected transition duration in the next supply relationship are determined. When the connection is inconsistent, the corresponding performance status change is terminated. The digital twin business entity position is continuously transmitted to the propagation blocking position according to the order of the digital twin main chain. The digital twin business entity position, predicted performance status, state transition probability and expected transition duration along the way are arranged in sequence to form the performance status transmission trajectory. The time from the prediction time point to the expected arrival of the performance status transmission trajectory at the propagation blocking position is determined as the prediction period.

[0064] S4.2. Write the performance status transmission trajectory into the business digital twin, and rewrite the performance stages of each business entity before the propagation blocking point in sequence to form the performance status segment before the blocking point, specifically, The positions of the digital twin business entities in the performance status transmission trajectory are matched one by one with the digital twin business entities in the business digital twin. The performance stage recorded in the performance status transmission trajectory is located to the corresponding performance stage boundary state piece. The actual performance status formed by the actual contract performance event in the performance stage boundary state piece is retained, and a performance status record with a deduced status mark is established at the corresponding digital twin business entity position.

[0065] In the performance status record with the deduced status marker, the predicted performance status recorded by the performance status transmission trajectory is used to rewrite the performance stage of the corresponding digital twin business entity. At the same time, the corresponding state transition probability and the expected transition duration are recorded, so that the performance status formed by the deduction and the actual performance status are retained respectively. The performance stages of the business entity that have been rewritten between the start position of the mutation causal entropy transmission position and the propagation blocking position are arranged continuously according to the sequential position of the digital twin entity chain to form the performance status segment before the blocking.

[0066] S4.3. At the propagation blocking point, the change in performance status corresponding to the mutation causal entropy is truncated. The performance stages of the business entities that have not been rewritten after the propagation blocking point are arranged into post-blocking performance status segments. The pre-blocking performance status segments and post-blocking performance status segments are combined according to the order of the digital twin entity chain to form the business entity situation prediction result, specifically, Using the propagation blocking position as the termination boundary of the performance status change, the performance stage of the business entity corresponding to the propagation blocking position is retained as the last position of the performance status segment before the blocking, and the performance status transmission trajectory is stopped from extending to the digital twin business entity after the propagation blocking position. Starting from the first digital twin business entity position after the propagation blocking position, the performance stages of the business entities that have not formed a deduced performance status are read in sequence. The performance stages of the business entities that maintain the actual performance status unchanged within the prediction period are arranged according to the order of the digital twin entity chain to form the performance status segment after the blocking.

[0067] The pre-blocking performance status segment is placed before the post-blocking performance status segment to maintain the continuity of the digital twin business entity positions and performance stage order before and after the propagation blocking position. The prediction time point, prediction period, predicted performance status, state transition probability, and expected transition duration corresponding to each digital twin business entity are recorded. The lowest state transition probability is selected from the state transition probabilities contained in the performance status transmission trajectory as the prediction credibility of the corresponding performance status transmission trajectory. The pre-blocking performance status segment, post-blocking performance status segment, prediction time point, prediction period, and prediction credibility are combined to form the business entity situation prediction result.

[0068] This embodiment also provides a business entity situation prediction system based on digital twins, including: The acquisition module is used to acquire contract performance events, distinguish supply relationships according to the transaction direction from the supplier to the purchaser, arrange the contract performance events under each supply relationship into performance event segments, and establish a business digital twin based on the stage changes recorded in the performance event segments. The filtering module is used to calculate the bidirectional causal entropy of adjacent business entities in the business digital twin, determine the dominant direction of causal entropy, filter out the mutation causal entropy from the bidirectional causal entropy corresponding to the dominant direction of causal entropy, and record the mutation transmission termination position where the mutation causal entropy stops propagating as the candidate propagation breakpoint. The interception module is used to identify the missing performance stage between the contract performance events on both sides of the candidate propagation break and mark it as an intermediate performance event gap. It intercepts the intermediate contract performance event fragments that are in the same stage and in the same direction of causal entropy dominance as the intermediate performance event gap from the performance event segment. It fills the intermediate contract performance event fragments into the intermediate performance event gap, retains the range of the order of the positions of the participants in the bidirectional causal entropy calculation on both sides of the candidate propagation break before filling in the fragments as the causal entropy verification window, updates the performance status in the causal entropy verification window with the contract performance events after filling in the fragments, and recalculates the bidirectional causal entropy on both sides of the candidate propagation break. It deletes the candidate propagation break where the mutation causal entropy resumes transmission and fixes the candidate propagation break where the mutation causal entropy still stops transmission as the propagation blocking position. The prediction module is used to infer the changes in the performance status of the business digital twin based on the propagation blockage location and along the direction of causal entropy dominance, and to cut off the transmission of sudden causal entropy at the propagation blockage location to form the business entity situation prediction result.

[0069] This embodiment also provides a computer device applicable to the business entity situation prediction method based on digital twins, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the business entity situation prediction method based on digital twins as proposed in the above embodiment.

[0070] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0071] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements the business entity situation prediction method based on digital twins as proposed in the above embodiments. The storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as Static Random Access Memory (SRAM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Red-Only Memory (PROM), Read-Only Memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0072] In summary, this invention improves the accuracy of propagation interruption location determination by recalculating bidirectional causal entropy. After the intermediate contract performance event fragment is filled into the gap in the intermediate performance event, the previously disconnected performance stages on both sides of the candidate propagation breakpoint are continuously connected, enabling the performance status changes between adjacent digital twin business entities to re-establish the corresponding relationship according to the complete stage sequence. The recalculated bidirectional causal entropy can more accurately reflect the transmission continuity in the direction of causal entropy dominance. When the abrupt causal entropy resumes transmission to the next supply relationship, the candidate propagation breakpoint is deleted; when the abrupt causal entropy still stops transmission, the candidate propagation breakpoint is fixed as the propagation interruption location. Thus, the absence of performance events in the digital twin no longer directly forms the erroneous propagation boundary, the identification range of the propagation interruption location is more convergent, and the consistency, stability, and reliability of the interruption determination and the prediction of the business entity's situation are improved.

[0073] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A business entity situation prediction method based on digital twins, characterized in that: include, Obtain contract performance events, distinguish supply relationships according to the transaction direction from the supplier to the purchaser, arrange the contract performance events under each supply relationship into performance event segments, and establish a business digital twin based on the stage changes recorded in the performance event segments; In the business digital twin, calculate the bidirectional causal entropy of adjacent business entities and determine the dominant direction of causal entropy. Select the mutation causal entropy from the bidirectional causal entropy corresponding to the dominant direction of causal entropy, and record the mutation transmission termination position where the mutation causal entropy stops propagating as the candidate propagation break point. Identify the missing performance stage between the contract performance events on both sides of the candidate propagation break and mark it as an intermediate performance event gap. Extract the intermediate contract performance event fragment that is in the same stage as the intermediate performance event gap and in the same direction of causal entropy dominance from the performance event segment. Fill the intermediate contract performance event fragment into the intermediate performance event gap. Preserve the range of the order of the positions of the participants in the bidirectional causal entropy calculation on both sides of the candidate propagation break before filling it in as the causal entropy verification window. Update the performance status in the causal entropy verification window using the contract performance events after filling in the gap and recalculate the bidirectional causal entropy on both sides of the candidate propagation break. Delete the candidate propagation break where the mutation causal entropy resumes transmission and solidify the candidate propagation break where the mutation causal entropy still stops transmission as the propagation blocking position. Based on the location of the propagation blockade, the changes in the performance status in the business digital twin are deduced along the direction of causal entropy dominance, and the transmission of causal entropy of the mutation is cut off at the propagation blockade location to form the business entity situation prediction result.

2. The business entity situation prediction method based on digital twin as described in claim 1, characterized in that: The supplier is the business entity that provides goods to the purchasing entity in accordance with the contract in the supply relationship; The procuring entity is a business entity that purchases goods from a supplier in accordance with the contract in a supply relationship; The business entity mentioned above is an operating entity that participates in contract performance and is determined as either the supplier or the purchaser based on the direction of the transaction; The contract performance events include purchase orders, supply confirmation, production completion, goods shipment, goods receipt, performance acceptance, invoice issuance, and payment. The supply relationship is a transactional relationship formed when a supplier provides goods to a purchaser in accordance with a contract.

3. The business entity situation prediction method based on digital twin as described in claim 2, characterized in that: The arrangement of contract performance events under each supply relationship into performance event segments is specifically as follows: Read the contract number, order number, event type, and occurrence time recorded in the contract performance event record, and group contract performance events with the same contract number and order number under the same supply relationship into the same performance event set; The contract performance events in the performance event set are arranged according to the time of occurrence. Contract performance events with the same time of occurrence are then arranged in the order of performance: purchase order, supply confirmation, production completion, goods shipment, goods receipt, performance acceptance, invoice issuance and payment, to determine the chronological order of each contract performance event. By sequentially connecting the contract performance events according to their order of appearance, a performance event segment is formed.

4. The business entity situation prediction method based on digital twin as described in claim 3, characterized in that: The establishment of the business digital twin specifically refers to... The supplier and purchaser are mapped to digital twin business entities respectively, and adjacent digital twin business entities are connected by supply relationships to form a digital twin entity chain; At the stage change position of the performance event segment, the performance status of adjacent digital twin business entities before and after the stage change position is recorded respectively, forming a performance stage boundary state piece; Based on their sequential order, the boundary status slices of the performance phase are written into the main chain of the digital twin, thus completing the establishment of the business digital twin.

5. The business entity situation prediction method based on digital twin as described in claim 4, characterized in that: The term "candidate propagation breakpoint" is specifically defined as follows: Arrange the performance status of the digital twin business entities at both ends of the same supply relationship according to their sequential positions to form a dual-end performance status sequence; In the business digital twin, the preceding and subsequent performance states in the two-end performance state sequence are paired bidirectionally to measure the uncertainty reduction from the supplier to the purchaser and the uncertainty reduction from the purchaser to the supplier, thus forming a bidirectional causal entropy. By comparing the uncertainty reduction of bidirectional causal entropy in two transmission directions, the transmission direction in which the uncertainty reduction continuously exceeds the opposite direction is defined as the causal entropy dominant direction. The transmission direction is the direction of the change in performance status between adjacent digital twin business entities along the supply entity to the purchasing entity and along the purchasing entity to the supply entity. Select the bidirectional causal entropy that is higher than both the preceding and following positions from the bidirectional causal entropy corresponding to the direction of causal entropy dominance to form the mutation causal entropy; The transmission of mutation causal entropy is traced along the digital twin entity chain. The business entity connection position where the transmission of mutation causal entropy stops in the next supply relationship is determined as the mutation transmission termination position and output as a candidate propagation break point.

6. The business entity situation prediction method based on digital twin as described in claim 5, characterized in that: The marker is defined as an intermediate performance event gap, specifically... According to the order of stage changes recorded in the performance event segment, arrange the performance stages between the contract performance events before and after the candidate propagation break point to form the performance stages to be verified; The contract performance events recorded on both sides of the candidate propagation gap are checked one by one during the pending performance stage. The performance stages occupied by existing contract performance events are removed, and the performance stages not occupied by contract performance events are retained. The retained performance stages are marked as intermediate performance event gaps.

7. The business entity situation prediction method based on digital twin as described in claim 6, characterized in that: Specifically, filling the gap in intermediate performance events with fragments of intermediate contract performance events involves... The interception interval is defined by the performance stage corresponding to the gap in the intermediate performance event. The contract performance event at the beginning of the interception interval in the performance event segment is recorded as the interception start event, and the contract performance event at the end of the interception interval is recorded as the interception termination event. Between the start and end events of the interception, the business entity direction of each contract performance event is marked according to the transaction direction from the supplier to the purchaser. Contract performance events whose business entity direction is opposite to the direction of causal entropy dominance are eliminated, and the remaining contract performance events are arranged into candidate performance event segments. The contract performance events in the candidate performance event segment are projected to the corresponding positions of the intermediate performance event gaps, and the contract performance events that continuously cover the missing performance stages are extracted to form intermediate contract performance event segments; According to the order of the missing performance stages within the intermediate performance event gaps, the contract performance events in the intermediate contract performance event segments are filled into the intermediate performance event gaps one by one.

8. The business entity situation prediction method based on digital twin as described in claim 7, characterized in that: The recalculation of the bidirectional causal entropy on both sides of the candidate propagation breakpoint specifically involves: Before the intermediate contract performance event fragment is added, the first and last positions in the order of participating in the bidirectional causal entropy calculation on both sides of the candidate propagation breakpoint are selected respectively to determine the front and back boundaries of the causal entropy verification window. After filling the intermediate contract performance event fragment into the intermediate performance event gap, keep the front and back boundaries of the causal entropy verification window unchanged, and classify the filled contract performance events into the causal entropy verification window according to the newly determined order of positions. Update the performance stage boundary state pieces corresponding to the filled positions, and rearrange the supplier's performance status and the purchaser's performance status in the causal entropy verification window to form an updated double-end performance status sequence. From the updated two-way performance state sequence, select the valid performance state combination without missing or conflicting markers, redetermine the conditional probability, and measure the uncertainty reduction from the supplier to the purchaser and the uncertainty reduction from the purchaser to the supplier respectively to form the recalculated two-way causal entropy.

9. The business entity situation prediction method based on digital twin as described in claim 8, characterized in that: The formation of the business entity situation prediction result is specifically as follows: Taking the performance status of the business entity at the starting position of the mutation causal entropy transmission as the starting point of transmission, the performance status change corresponding to the mutation causal entropy is transmitted level by level according to the direction of causal entropy dominance, forming a performance status transmission trajectory. The starting position of the mutation causal entropy transmission is the position of the business entity where the mutation causal entropy first appears along the direction of causal entropy dominance and begins to be transmitted to the next supply relationship. Write the performance status transmission trajectory into the business digital twin, and rewrite the performance stage of each business entity before the propagation blockage point in turn to form the performance status segment before the blockage. At the propagation blocking point, the change in the performance status corresponding to the mutation causal entropy is truncated, and the performance stage of the business entity that has not been rewritten after the propagation blocking point is arranged as the post-blocking performance status segment. The business entity status prediction result is formed by combining the pre-blocking and post-blocking performance status segments according to the sequential positions of the digital twin entity chain.

10. A business entity situation prediction system based on digital twins, based on the business entity situation prediction method based on digital twins according to any one of claims 1 to 9, characterized in that: include, The acquisition module is used to acquire contract performance events, distinguish supply relationships according to the transaction direction from the supplier to the purchaser, arrange the contract performance events under each supply relationship into performance event segments, and establish a business digital twin based on the stage changes recorded in the performance event segments. The filtering module is used to calculate the bidirectional causal entropy of adjacent business entities in the business digital twin, determine the dominant direction of causal entropy, filter out the mutation causal entropy from the bidirectional causal entropy corresponding to the dominant direction of causal entropy, and record the mutation transmission termination position where the mutation causal entropy stops propagating as the candidate propagation breakpoint. The interception module is used to identify the missing performance stage between the contract performance events on both sides of the candidate propagation break and mark it as an intermediate performance event gap. It intercepts the intermediate contract performance event fragments that are in the same stage and in the same direction of causal entropy dominance as the intermediate performance event gap from the performance event segment. It fills the intermediate contract performance event fragments into the intermediate performance event gap, retains the range of the order of the positions of the participants in the bidirectional causal entropy calculation on both sides of the candidate propagation break before filling in the fragments as the causal entropy verification window, updates the performance status in the causal entropy verification window with the contract performance events after filling in the fragments, and recalculates the bidirectional causal entropy on both sides of the candidate propagation break. It deletes the candidate propagation break where the mutation causal entropy resumes transmission and fixes the candidate propagation break where the mutation causal entropy still stops transmission as the propagation blocking position. The prediction module is used to infer the changes in the performance status of the business digital twin based on the propagation blockage location and along the direction of causal entropy dominance, and to cut off the transmission of sudden causal entropy at the propagation blockage location to form the business entity situation prediction result.

Images