Search

CN-121998574-A - Enterprise demand full life cycle management system and method

CN121998574ACN 121998574 ACN121998574 ACN 121998574ACN-121998574-A

Abstract

The invention relates to the technical field of computer software, and discloses a system and a method for managing a full life cycle of enterprise requirements. The architecture asset management subsystem calculates a technology invasiveness index based on the architecture topology map, and the demand orchestration subsystem routes demands to different levels of management and control processes accordingly. And the project implementation management and control subsystem dynamically calculates the maximum parallel implementation project number threshold value and executes standing congestion control according to the agility implementation cooperative ability score fed back by the assessment and evaluation subsystem. The production verification subsystem generates a digital fingerprint based on the requirement identification, the architecture snapshot and the code record, and performs consistency verification before deployment. The invention realizes the closed-loop refined management and risk control of the all-link requirement through risk quantitative diversion, resource self-adaptive adjustment and production fingerprint binding.

Inventors

  • ZHAO YINGYING

Assignees

  • 中国建设银行股份有限公司江苏省分行

Dates

Publication Date
20260508
Application Date
20251215

Claims (10)

  1. 1. An enterprise demand full lifecycle management system, the system operating in a computer network environment including an application server, a database server, and a client terminal, the system comprising: A demand system tally system configured to maintain a demand entity vector and manage full lifecycle state flows of demands from entry, stand, implementation to production using state machine logic; The architecture asset management subsystem is configured to construct and maintain an architecture topological graph comprising a physical subsystem, a logical subsystem, application engineering and component nodes, and record static weight attributes of each node and coupling strength coefficients of directed edges among the nodes; The project implementation management and control subsystem is configured to manage the project arrangement period of the requirement, calculate the maximum parallel implementation project number threshold value of the department in the current period according to the scoring data read from the assessment and evaluation subsystem, and execute congestion control on the project operation of the new requirement according to the maximum parallel implementation project number threshold value; the production verification subsystem is configured to generate production digital fingerprints based on the requirement identification, the architecture configuration snapshot and the code submission record in the requirement production stage, and execute fingerprint consistency verification before deployment; The assessment and evaluation subsystem is configured to calculate a multi-dimensional quantitative score according to the demand circulation records, the overdue records and the verification logs; The assessment and evaluation subsystem is configured to feed the calculated agile implementation coordination ability score back to the project implementation management and control subsystem, and the project implementation management and control subsystem is configured to calculate the maximum parallel implementation project number threshold value of the next period by taking the agile implementation coordination ability score as an input parameter.
  2. 2. The enterprise demand full life cycle management system of claim 1, wherein the architecture asset management subsystem is configured to perform breadth-first search algorithms traversing the architecture topology graph based on an initial set of influencing nodes involved in demand to obtain a technical invasiveness index; The calculation logic of the technical invasiveness index is that for each initial influence node in the initial influence node set, the sum of the static weight value of the initial influence node and the accumulated influence value of all downstream influenced nodes of the initial influence node is calculated; Traversing each downstream affected node which is reachable by the initial affected node through a directed edge, multiplying the static weight value of each downstream affected node by the accumulated coupling strength on a path, and dividing the static weight value by the square value of the topological distance between the downstream affected node and the initial affected node to obtain a calculation result; The demand system tally is configured to compare the technical invasiveness index with a preset low risk threshold and high risk threshold, and route the demand to a fast channel process, a standard technical review process or a strict management process according to the comparison result.
  3. 3. The enterprise demand full lifecycle management system as claimed in claim 2, wherein the routing logic performed by the demand system tally system comprises: Routing to a fast path that does not include a technology review node when the technology invasiveness index is less than the low risk threshold; And when the technical invasiveness index is larger than the high risk threshold, routing to a strict management flow and sending a locking instruction to the architecture asset management subsystem, and limiting the writing permission of the code warehouse corresponding to the initial influence node set.
  4. 4. The enterprise demand full lifecycle management system of claim 1, wherein the project implementation management subsystem is configured to calculate the maximum parallel implementation project number threshold by adjusting a department's base concurrency capacity constant; The adjusting logic comprises the steps of obtaining a agile implementation coordination capacity score of a target department in the last period, calculating a difference value between the agile implementation coordination capacity score and a system reference score and dividing the difference value by a normalization factor, executing hyperbolic tangent function operation on a calculation result, multiplying an output value of the hyperbolic tangent function by an adjusting sensitivity coefficient and adding one to obtain an adjusting coefficient, multiplying the basic concurrency capacity constant by the adjusting coefficient and executing downward rounding operation to obtain the maximum parallel implementation project number threshold value.
  5. 5. The enterprise demand full lifecycle management system of claim 4, wherein the congestion control performed by the project implementation management subsystem comprises: Counting the number of demands of the current in-implementation state of the target department in real time; if the number of the requirements reaches or exceeds the maximum parallel implementation project number threshold, sending a blocking signal to the requirement overall planning subsystem, freezing a standing operation control and storing new requirements into a buffer area to be processed; And releasing the demand in the buffer to be processed only when the demand number falls below the maximum parallel implementation project number threshold value or a new check period causes the maximum parallel implementation project number threshold value to increase.
  6. 6. The enterprise demand full lifecycle management system of claim 1, wherein the commissioning verification subsystem is configured to generate the commission digital fingerprint using a collision-resistant one-way hash function and a bitwise exclusive-or operation; The logic for generating the production digital fingerprint comprises the steps of obtaining a unique identifier of a requirement, calculating a hash value of architecture configuration snapshot data recorded during implementation, calculating a hash value of a code branch latest submitting record identifier, performing bitwise exclusive OR operation on the unique identifier of the requirement, the hash value of the architecture configuration snapshot data and the hash value of the code branch latest submitting record identifier, and performing the unidirectional hash function calculation again on an exclusive OR operation result to obtain an expected check value; When the deployment instruction is triggered, the production verification subsystem recalculates the fingerprint of the current environment by adopting the same logic and performs binary comparison with the production digital fingerprint.
  7. 7. The enterprise demand full life cycle management system of claim 6, wherein when the recalculated fingerprint is inconsistent with the on-stream digital fingerprint, the on-stream verification subsystem sends a blocking signal to a production environment deployment interface to terminate deployment, generates an exception audit log with a recorded difference term, and sends the exception audit log to the assessment subsystem.
  8. 8. The enterprise demand full life cycle management system of claim 1, wherein the assessment subsystem is configured to calculate a weighted total assessment score comprising a market response capability score, a demand orchestration capability score, a agile implementation collaboration capability score, and a risk improvement rate score.
  9. 9. The enterprise demand full life cycle management system of claim 8, wherein the assessment subsystem, when calculating the agile performance coordination capability score, comprises calculating a planned execution degree sub-term based on a number of overdue records generated by the project performance management and control subsystem, and calculating an execution quality sub-term based on a number of verification failure records generated by the production verification subsystem, wherein the number of overdue records and the number of verification failure records are both used as a reduction parameter, and a zero value is taken when the result after the reduction is less than zero.
  10. 10. An enterprise demand full life cycle management method, characterized in that it is applied to an enterprise demand full life cycle management system as claimed in any one of claims 1-9, comprising the steps of: s1, an architecture asset management subsystem establishes and maintains an architecture topological graph containing node weights and edge coupling strengths; S2, a demand system tally receives business demand entry and analyzes a target node set related to the business demand; S3, the architecture asset management subsystem calculates technology invasiveness indexes based on the distribution of the target node set in the architecture topological graph, and the demand overall subsystem routes demands to different levels of management and control flows according to the technology invasiveness indexes; S4, the project implementation management and control subsystem obtains a agile implementation cooperative capacity score of a department in the previous period, calculates a maximum parallel implementation project number threshold value of the current period according to the agile implementation cooperative capacity score, and performs congestion control on the newly required stand by releasing or storing the stand in a buffer zone based on the maximum parallel implementation project number threshold value; S5, in the project implementation process, the project implementation management and control subsystem monitors the time consumption and generates overdue records; S6, when the demand is put into production, the production verification subsystem generates production digital fingerprints based on the demand identification, the architecture configuration snapshot and the code submitting record, performs fingerprint consistency verification before deployment, performs deployment operation when the production digital fingerprints are consistent with the fingerprints of the current environment, and performs blocking operation when the production digital fingerprints are inconsistent with the fingerprints of the current environment; And S7, the assessment and evaluation subsystem calculates the multi-dimensional capacity scores of all departments according to the circulation data, the overdue records and the fingerprint verification logs, and provides calculation results to the project implementation management and control subsystem for being used as input of the next cycle execution step S4.

Description

Enterprise demand full life cycle management system and method Technical Field The invention relates to the technical field of computer software, in particular to a system and a method for managing a full life cycle of enterprise requirements. Background With the penetration of enterprise digital transformation, the architecture of a software system is increasingly complex, and the iteration frequency of business requirements is also continuously accelerated. In enterprise-level software development management, how to efficiently and safely manage the whole process from the requirement proposal to the final production is a key for guaranteeing the service continuity and the system stability. When the existing demand management tool performs flow control, the approval path is usually determined only according to the manually filled declaration level or business strip line attribute. This lack of automated analysis of underlying system architecture and coupling relationships between components results in risk assessment relying primarily on subjective judgment of project managers. Because of the lack of objective quantitative basis based on technical view, the management system often adopts a cut-off management mode, so that not only is the low-risk tiny change subjected to excessive approval cost, but also the high-risk change is lack of sufficient verification and management in the implementation process due to the fact that the deep coupling relation is not identified. In terms of scheduling and scheduling of project-implemented resources, conventional systems often employ static quota management or simple estimation based on the number of people. This management does not take into account fluctuations in the actual delivery performance and coordination ability of the implementation team over different periods. When delivery efficiency is reduced or tasks backlog occurs to an implementation team for various reasons, a management system lacking a dynamic feedback mechanism still allows new demands to stand according to the original quota. The open-loop resource allocation method is easy to cause congestion of an implementation channel, so that overdue tasks are continuously piled up, and the delivery quality and response speed of a team are further deteriorated. In addition, in the requirement production deployment link, the existing full life cycle management flow and the automatic operation and maintenance pipeline often have data faults. The verification before deployment mainly focuses on the compiling state of codes or the passing rate of test cases, and lacks technical means for strictly binding the stand requirement identification of a management end, the configuration snapshot of an operating environment and the submitting record of a code warehouse. Because of the lack of the cross-dimension integrity check, the system is difficult to automatically identify unexpected configuration changes or unauthorized code submission in the implementation process, so that the actual content of the production line is inconsistent with the approved requirement range, and the running risk of the production environment is increased. Disclosure of Invention Aiming at the defects of the prior art, the invention provides a full life cycle management system and method for enterprise requirements, and solves the problems of resource waste or risk omission caused by unified management and control modes in the prior art, continuous accumulation of new requirements when delivery capacity is reduced, and easy unauthorized modification or configuration drift in later implementation period. The first aspect of the present invention provides an enterprise demand full lifecycle management system. The system operates in a computer network environment including an application server, a database server, and a client terminal. The system comprises a demand overall planning subsystem, an architecture asset management subsystem, a project implementation management and control subsystem, a production verification subsystem and an assessment and evaluation subsystem. The demand system tally system is configured to maintain a demand entity vector and manage full lifecycle state flows of demands from entry, stand, implementation to production using state machine logic. The architecture asset management subsystem is configured to construct and maintain an architecture topology map comprising physical subsystem, logical subsystem, application engineering and component nodes, and record static weight attributes of each node and coupling strength coefficients of directed edges between nodes. The project implementation management and control subsystem is configured to manage the standing schedule of the demands, calculate a maximum parallel implementation project number threshold value of departments in the current period according to the scoring data read from the assessment and evaluation subsystem, and execute congestion control on the sta