CN-122022152-A - Marginal information identification and decision support method and system based on management line

CN122022152ACN 122022152 ACN122022152 ACN 122022152ACN-122022152-A

Abstract

The invention provides a marginal information identification and decision support method and system based on a management line, which relate to the field of enterprise management digitization and comprise the following steps of acquiring multi-source business data and flow data generated in an enterprise management process, and associating and mapping the multi-source business data and the flow data by combining the management line comprising a longitudinal management path and a transverse cooperative network to obtain association and mapping results; based on the management line, combining the association and mapping results, carrying out key node identification on the multi-source business data and the flow data to determine a target node influencing the management decision, identifying marginal information which changes relative to the historical state, carrying out management influence evaluation on the identified marginal information to judge the influence degree of the identified marginal information on the current management decision, generating corresponding decision support information and outputting the corresponding decision support information to a management decision interface. The method improves the accuracy and efficiency of management decisions, adapts to dynamic business scenes of enterprises and continuously optimizes decision support effects.

Inventors

WANG YANG
WAN SHIXIONG

Assignees

四川莱芈斯信息技术有限公司

Dates

Publication Date: 20260512
Application Date: 20260127

Claims (10)

1. The marginal information identification and decision support method based on the management line is characterized by comprising the following steps of: S1, multi-source business data and flow data generated in an enterprise management process are acquired, and the multi-source business data and the flow data are associated and mapped by combining a management line comprising a longitudinal management path and a transverse cooperative network to obtain an association and mapping result; S2, based on the management line, combining the association and mapping results, carrying out key node identification on the multi-source business data and the flow data, and determining a target node affecting the management decision; S3, at the target node, carrying out marginal change analysis on the real-time multi-source service data and the flow data related to the node state change, and identifying marginal information which changes relative to the historical state; s4, performing management influence assessment on the identified marginal information, and judging the influence degree of the marginal information on the current management decision; and S5, when the influence degree meets the preset decision triggering condition, generating corresponding decision support information and outputting the corresponding decision support information to a management decision interface.
2. The marginal information identification and decision support method based on management line according to claim 1, wherein the steps of obtaining the multi-source business data and the flow data generated in the enterprise management process, and associating and mapping the multi-source business data and the flow data by combining the management line including the longitudinal management path and the transverse cooperative network, and obtaining the association and mapping result comprise the following steps: S11, identifying a longitudinal management path and a transverse cooperative network of an enterprise according to an organization architecture and a business flow of the enterprise, and generating a management line for representing an enterprise management decision and an execution context according to the longitudinal management path and the transverse cooperative network; S12, collecting multi-source service data and flow data generated in an enterprise management process, and obtaining standard multi-source service data and flow data by carrying out data cleaning and standardized cleaning on the multi-source service data and the flow data; S13, according to the management line and the standard multi-source service data and flow data, determining the dynamic association relation and mapping result between the multi-source service data and flow data and the management line nodes by analyzing the causal link between the nodes.
3. The marginal information identification and decision support method based on management line according to claim 2, wherein the steps of identifying the longitudinal management path and the transverse cooperative network of the enterprise according to the organization architecture and the business process of the enterprise, and generating the management line for representing the management decision and the execution context of the enterprise according to the longitudinal management path and the transverse cooperative network, comprises the steps of: S111, acquiring an organization architecture of an enterprise, and identifying a decision layer, a management layer and an execution layer of the enterprise to obtain an identification result; s112, determining decision nodes and execution nodes in the longitudinal management path according to the identification result to form the longitudinal management path; S113, acquiring a business process of an enterprise, and identifying cooperative nodes and cooperative rules by analyzing interaction and cooperative relations among functional departments; S114, connecting nodes scattered in different functional departments according to the service coupling degree and the information dependency relationship based on the cooperative nodes and the cooperative rules to form a transverse cooperative network; And S115, carrying out structural fusion on the longitudinal management path and the transverse cooperative network to obtain a management line for representing enterprise management decisions and execution venues.
4. The management line-based marginal information identification and decision support method according to claim 2, wherein the determining the dynamic association and mapping result between the multi-source service data and the flow data and the management line node by analyzing the causal link between the nodes according to the multi-source service data and the flow data of the management line and the standard comprises the steps of: s131, classifying and aligning standard multi-source service data and flow data according to the node attribute of the management line to form a data-node initial association pool; s132, filtering invalid data irrelevant to the management line node management and control requirements in a data-node initial association pool by using a preset expert experience library to form a node association candidate data set; S133, carrying out structural conversion and feature extraction on the node association candidate data set, and constructing a causal chain by identifying causal association rules between the service data mode and the node state change; S134, constructing and dynamically updating a mapping relation library between the service data mode and the node state according to the causal chain and the real-time data stream, and outputting a dynamic association mapping result.
5. The method for identifying and supporting decision-making based on marginal information of management line according to claim 4, wherein said step of performing structural transformation and feature extraction on the node association candidate data set and constructing a causal link by identifying causal association rules between traffic data patterns and node state changes comprises the steps of: S1331, converting business events and flow states in the node association candidate data set into standardized transactions taking a preset time window as a unit, and extracting characteristics related to node decision for each standardized transaction to form a structured transaction set; S1332, respectively calculating the support and attention indexes of the combination of the service data mode and the node state change according to the structured transaction set, and when the preset constraint condition is met, setting the inclusion condition frequent item set of the combination of the service data mode and the node state change; S1333, constructing a causal chain with a causal direction based on the conditional frequent item set.
6. The management-line-based marginal information identification and decision support method of claim 5 wherein constructing a causal link having a causal direction based on the set of conditional frequent items comprises the steps of: s13331, generating candidate association rules which take the business data mode as a front piece and change the node state as a rear piece based on the condition frequent item set; s13332, calculating the confidence coefficient of each candidate association rule, and screening candidate association rules with the confidence coefficient larger than a preset minimum confidence coefficient threshold value; s13333, carrying out time sequence analysis on the screened candidate association rule, and verifying whether the occurrence time of the former event is earlier than that of the latter event, if so, determining the candidate association rule as a causal chain with a causal direction, otherwise, eliminating the candidate association rule.
7. The marginal information identification and decision support method based on management line according to claim 1, wherein the identifying key nodes of the multi-source business data and the flow data based on the management line and combining the association and mapping result, determining the target nodes affecting the management decision comprises the following steps: s21, acquiring marginal index data of each node in the management line based on the association and mapping results, and constructing a marginal information time sequence of each management line node according to the marginal index data; s22, calculating the marginal effect of each management line node according to the marginal information time sequence of each management line node, and identifying key decision points to form a marginal effect analysis report; s23, determining a target node which has an influence on the current management decision from the key decision points based on the marginal effect analysis report by combining a preset decision influence threshold and a target node screening rule.
8. The method for identifying and supporting marginal information based on management lines according to claim 7, wherein the calculating the marginal effect of each management line node according to the marginal information time sequence of each management line node, and identifying the key decision point, and forming the marginal effect analysis report comprises the following steps: s221, dividing the management line nodes into an intervention group node and a comparison group node according to the marginal information time sequence of each management line node and in combination with the hierarchical attribute of the management line node; S222, checking whether the intervention group node meets a preset parallel trend assumption in a time window before marginal change occurs, and adjusting the intervention group node which does not meet the parallel trend until the intervention group node meets the preset parallel trend assumption; s223, for the intervention group nodes meeting the assumption of parallel trend, combining with the matched comparison group nodes, constructing a counter fact time sequence of the intervention group nodes without marginal change; S224, according to the anti-fact time sequence of the intervention group node without marginal change, extracting an actual observation result of the intervention group node after the marginal change occurs, and calculating a difference value of the actual observation result and a corresponding time point value of the anti-fact time sequence to obtain a marginal effect instantaneous value of each time point; S225, calculating an average treatment effect based on the marginal effect instantaneous value, and taking the average treatment effect as an initial marginal effect estimated value of the intervention group node; S226, calculating the statistical significance of the final marginal effect estimated value, screening intervention group nodes meeting a preset judgment standard, determining the screened intervention group nodes as key decision points, and generating a marginal effect analysis report containing the size and the direction of the marginal effect through integration.
9. The management line-based marginal information identification and decision support method according to claim 8, wherein the dividing the management line nodes into the intervention group nodes and the comparison group nodes according to the marginal information time sequence of each management line node in combination with the hierarchical attribute of the management line nodes comprises the following steps: S2211, extracting the hierarchical attribute of each management line node according to the longitudinal management path and the transverse cooperative network of the management line; s2212, identifying the marginal change trend of each node in a preset time window based on the marginal information time sequence of each node, and detecting whether a marginal change event meeting a preset standard exists or not; S2213, marking the management line nodes with detected marginal change events as intervention candidate nodes, and taking the rest management line nodes with no detected significant change as contrast nodes.
10. A management line-based marginal information identification and decision support system for implementing the management line-based marginal information identification and decision support method according to any one of claims 1 to 9, characterized in that the system comprises: The association mapping module is used for acquiring multi-source service data and flow data generated in the enterprise management process, and associating and mapping the multi-source service data and the flow data by combining a management line comprising a longitudinal management path and a transverse cooperative network to obtain association and mapping results; The target node determining module is used for carrying out key node identification on the multi-source business data and the flow data based on the management line and combining the association and the mapping result, and determining target nodes influencing the management decision; the marginal information identification module is used for carrying out marginal change analysis on the real-time multi-source service data and the flow data related to the node state change at the target node and identifying the marginal information which changes relative to the historical state; The influence degree judging module is used for carrying out management influence evaluation on the identified marginal information and judging the influence degree of the marginal information on the current management decision; And the decision support output module is used for generating corresponding decision support information and outputting the corresponding decision support information to the management decision interface when the marginal information meets the preset decision triggering condition.

Description

Marginal information identification and decision support method and system based on management line Technical Field The invention relates to the field of enterprise management digitization, in particular to a marginal information identification and decision support method and system based on management lines. Background In the modern enterprise management process, along with the advancement of digital transformation, enterprises can generate massive multi-source business data and flow data in various links of purchasing, production, sales, management and control and the like, and the data are core bases for supporting enterprise management decisions. At present, management systems such as OA, ERP, BI widely applied by enterprises mainly bear functions of data recording, statistics summarization and basic report output, can only meet basic requirements of information retention and later disk duplication in daily management of enterprises, has obvious technical shortboards in the aspects of supporting fine and prospective management decisions, and is difficult to adapt to management and control requirements after large-scale development of enterprises. The conventional enterprise management system mainly solves the problems of information recording and statistics, and has the defects that (1) management data are scattered, key management nodes are difficult to focus, the corresponding relation between the data and the management nodes is fuzzy, massive data are invalid and redundant, the massive data are difficult to be converted into effective basis for supporting node management and control and decision making, (2) after report results are biased, risks cannot be identified in advance, early warning is difficult to be sent out in a risk sprouting stage, a manager can only passively deal with the problems which occur, and prospective decision cannot be realized, and (3) the manager needs to manually judge which information really affects decision, and lacks quantitative marginal effect analysis and scientific screening standard, so that the decision efficiency is low, management and control errors are easy to be caused by subjective deviation, and key information which has substantial influence on the decision cannot be quickly locked. For the problems in the related art, no effective solution has been proposed at present. Disclosure of Invention In view of the above, the present invention provides a marginal information identification and decision support method and system based on management lines to solve the above-mentioned problems. In order to solve the problems, the invention adopts the following specific technical scheme: according to an aspect of the present invention, there is provided a management line-based marginal information identification and decision support method including the steps of: S1, multi-source business data and flow data generated in an enterprise management process are acquired, and the multi-source business data and the flow data are associated and mapped by combining a management line comprising a longitudinal management path and a transverse cooperative network to obtain an association and mapping result; S2, based on the management line, combining the association and mapping results, carrying out key node identification on the multi-source business data and the flow data, and determining a target node affecting the management decision; S3, at the target node, carrying out marginal change analysis on the real-time multi-source service data and the flow data related to the node state change, and identifying marginal information which changes relative to the historical state; s4, performing management influence assessment on the identified marginal information, and judging the influence degree of the marginal information on the current management decision; and S5, when the influence degree meets the preset decision triggering condition, generating corresponding decision support information and outputting the corresponding decision support information to a management decision interface. Preferably, the step of obtaining the multi-source service data and the flow data generated in the enterprise management process, and associating and mapping the multi-source service data and the flow data by combining a management line including a longitudinal management path and a transverse cooperative network, so as to obtain an association and mapping result includes the following steps: S11, identifying a longitudinal management path and a transverse cooperative network of an enterprise according to an organization architecture and a business flow of the enterprise, and generating a management line for representing an enterprise management decision and an execution context according to the longitudinal management path and the transverse cooperative network; S12, collecting multi-source service data and flow data generated in an enterprise management process, and obtaining standard multi-sou