A multi-dimensional and multi-layer index system construction method for complex system design

Abstract

The application discloses a kind of multi-dimensional multi-layer index system construction methods for complex system design, based on RFLP process, index is divided into efficiency layer, function / performance layer and technical layer;In the demand stage, the improved hall system engineering method is used, to cross definition covering the whole life cycle of efficiency index with system dimension, time dimension, knowledge dimension;In the functional and logical architecture design stage, four layers of function / performance index are refined using SPIT method, i.e., social layer-practice layer-information layer-technology layer;At the end of logical architecture and physical architecture stage, the function / performance index corresponding to the minimum logical unit is converted into technical index for physical components, and a two-way mapping between the three-layer index and the architecture unit of each stage of RFLP is established;The method realizes process integration, whole life cycle coverage and closed-loop tracing, and alleviates the problems of index disconnection, single dimension and lack of two-way mapping.

Description

Technical Field

[0001] This invention belongs to the field of model-based systems engineering and complex system design verification technology, and in particular relates to a method for constructing a multi-dimensional and multi-layer index system for complex system design. Background Technology

[0002] As the design of complex systems such as aerospace vehicles, large electromechanical equipment, and intelligent unmanned systems becomes increasingly complex, how to construct a comprehensive indicator system that can reflect the quality of system design and support the evaluation and optimization of solutions in the early stages of design has become a key issue in the field of systems engineering. Model-based systems engineering (MBSE) organizes design information in a unified manner through multi-view models such as requirements, functions, logic, and physical models, providing a model-based foundation for the construction of indicator systems. However, there is still a lack of mature methodological support for establishing a clear two-way mapping relationship between the indicator system and system architecture units.

[0003] In existing technologies, the construction of indicator systems typically adopts the following approaches: First, it constructs an indicator tree based on the analytic hierarchy process or expert experience, relying on subjective judgment to decompose it layer by layer; second, it is based on the Hall system engineering method, determining the indicator dimensions through time, logic, and knowledge dimensions; third, as proposed in patent CN118778945A, "a method and system for supporting the association, fusion, and verification of indicators in complex systems," decomposing the indicators according to the organizational hierarchy of mission-system-system, constructing a multi-level indicator model and traceability matrix, and realizing the association and fusion of indicators through an ontology knowledge model.

[0004] However, existing technologies have obvious shortcomings: the indicator system is disconnected from MBSE's RFLP (Requirements → Functions → Logic → Physical) design process, making it difficult to trace indicators back to specific design units; the horizontal dimension is singular, failing to systematically generate performance indicators from the entire lifecycle and multidisciplinary knowledge dimensions, and also lacking a definition path for non-functional indicators; there is a lack of bidirectional mapping relationship between indicators and architectural units, making it impossible to support a collaborative closed loop of forward decomposition and reverse verification. Summary of the Invention

[0005] Purpose of the invention: The purpose of this invention is to provide a method for constructing a multi-dimensional and multi-layer indicator system for complex system design, which solves the problems in the prior art such as the disconnect between the indicator system and the RFLP design process, the single horizontal dimension and the lack of non-functional indicator definition paths, and the lack of bidirectional mapping relationship between indicators and architectural units. It realizes the deep integration of the indicator system and the system design process, the complete coverage of multi-dimensional indicators throughout the entire life cycle, and the positive mapping and reverse traceability closed loop between indicators and architectural units.

[0006] Technical solution: The method for constructing a multi-dimensional and multi-layered indicator system for complex system design as described in this invention includes the following steps:

[0007] S1: Based on the RFLP process of requirement definition-functional architecture-logical architecture-physical architecture, the indicator system to be built is divided into the performance indicator layer, the functional / performance indicator layer and the technical indicator layer.

[0008] S2: In the requirement definition phase, based on the improved Hall system engineering method, taking the requirement definition node of the system dimension in the improved Hall system engineering method as the axis, the performance indicators covering the entire life cycle of the system are covered by the cross definition of the time dimension and knowledge dimension in the improved Hall system engineering method; wherein, the improved Hall system engineering method is obtained by changing the logical maintenance of the original Hall system engineering method to the system dimension.

[0009] S3: In the functional and logical architecture design phase, based on the SPIT method (Social Layer—Practice Layer—Information Layer—Technology Layer), the functional / performance indicators in the functional / performance indicator layer are refined in four dimensions.

[0010] S4: At the end of the logical architecture design phase and the physical architecture design phase, the functional / performance indicators corresponding to the smallest indivisible logical unit after refinement by the SPIT method are converted into technical indicators for physical components. A bidirectional mapping relationship is established between the performance indicators, the functional / performance indicators, the technical indicators and the architectural units included in each stage of the RFLP process, thereby outputting a multi-dimensional and multi-layer indicator system.

[0011] The method provided by this invention can effectively alleviate the problem of the disconnect between the indicator system and the RFLP design process in the prior art. Step S1 directly maps the indicator layer to each stage of RFLP, realizing the deep integration of the indicator system and the system design process. In response to the shortcomings of the single horizontal dimension and lack of non-functional indicator definition path, step S2 uses the improved Hall system engineering method to construct performance indicators covering the entire life cycle of the system in the requirements definition stage. Step S3 further introduces the SPIT method to refine the functional / performance indicators in four dimensions, thereby supporting the complete coverage of multi-dimensional indicators (including non-functional indicators) throughout the entire life cycle. At the same time, in response to the problem of the lack of bidirectional mapping relationship between indicators and architectural units, step S4 establishes a bidirectional mapping between performance indicators, functional / performance indicators, technical indicators and architectural units in each stage of RFLP, forming a positive mapping and reverse traceability closed loop between indicators and architectural units, improving the completeness and traceability of the indicator system.

[0012] Preferably, the performance indicator layer mentioned in step S1 corresponds to the requirement definition phase in the RFLP process and is used to measure whether the system delivery meets the top-level task requirements of stakeholders. The performance indicators are extracted from the needs of stakeholders and generated based on the full lifecycle use cases.

[0013] The functional / performance metrics layer corresponds to the functional architecture and logical architecture design phases in the RFLP process. It is used to evaluate the system's functional implementation and performance. Functional metrics are used to evaluate the effectiveness of completing a certain function, while performance metrics are used to describe the measurement of physical or functional attributes related to system operation.

[0014] The technical specification layer corresponds to the physical architecture design phase in the RFLP process and is used to describe the technical specifications and parameters required for physical components to achieve functional / performance indicators.

[0015] This preferred approach effectively alleviates the misalignment between the indicator system and the system design process by explicitly mapping the performance indicator layer, functional / performance indicator layer, and technical indicator layer to the requirement definition, functional and logical architecture design, and physical architecture design phases in the RFLP process, respectively. Specifically, the performance indicator layer, extracted from stakeholder needs and generated based on full lifecycle use cases, helps enhance the indicator's coverage of top-level task requirements and full lifecycle scenarios; the functional / performance indicator layer distinguishes between the effectiveness of function completion and operational attributes, facilitating the refinement of evaluation dimensions; and the technical indicator layer focuses on the technical specifications and parameters of physical components, providing clearer quantitative evidence for the transition from function to implementation. Overall, this hierarchical mapping further supports the connection between the indicator system and the core activities of each stage of RFLP.

[0016] Preferably, the improved Hall system engineering method described in step S2 includes:

[0017] The system dimensions are set to include five stages: requirements definition, functional architecture, logical architecture, physical architecture, and verification and validation.

[0018] The time dimension is further refined into the planning stage, design stage, production stage, installation stage, operation stage, and recycling stage;

[0019] The knowledge dimension is set as including five basic areas: economics, management, law, technology, and environmental protection.

[0020] Using the requirement definition stage of the system dimension as the axis, the intersection of any life cycle stage of the time dimension and any domain of the knowledge dimension is taken to generate an effectiveness index corresponding to the selected life cycle stage and domain, and the effectiveness index is directly mapped to the needs of stakeholders.

[0021] This preferred approach redefines and organizes the Hall system engineering methodology across system, time, and knowledge dimensions. Using the requirements definition phase of the system dimension as a pivot, it generates performance indicators at the intersection of each lifecycle stage of the time dimension and each domain of the knowledge dimension, directly mapping them to stakeholder needs. This effectively alleviates the problems of incomplete coverage and insufficient cross-domain considerations of performance indicators in traditional methods. Specifically, the correspondence between the system dimension and the RFLP process and verification phase helps strengthen the connection between the indicator system and the system design mainline; the detailed time dimension, from planning to recovery, provides a foundation for covering the differentiated indicator requirements of each stage; and the introduction of economic, management, legal, technical, and environmental fields in the knowledge dimension facilitates the generation of non-functional and multi-dimensional constraint indicators. Overall, this method improves the systematicity and traceability of performance indicators across the entire lifecycle and multiple professional domains.

[0022] Preferably, the granularity of the time dimension and the knowledge dimension is adjusted according to the characteristics of the system: for short-cycle systems, some stages in the time dimension are merged, and for specific domains, domains in the knowledge dimension are added; the economic domain in the knowledge dimension is further decomposed into sub-domains of cost control, investment return range, and investment return cycle.

[0023] This preferred approach enhances the adaptability of the indicator construction method to different types of systems and specific application scenarios by merging stages in the time dimension and expanding and decomposing sub-domains in the knowledge dimension. Specifically, merging certain stages in the time dimension for short-cycle systems simplifies the indicator generation process and better aligns with the actual lifecycle characteristics of such systems. Adding domains to the knowledge dimension for specific domains, and further decomposing the economic domain into sub-domains such as cost control, investment return range, and investment return cycle, refines the dimensions for generating performance indicators, supporting more accurate mapping of stakeholder needs. Overall, this approach alleviates the shortcomings of fixed-granularity methods in dealing with system diversity and detailed economic analysis, improving the flexibility and domain-specificity of indicator system construction.

[0024] Preferably, the performance indicators covering the entire lifecycle of the system as defined in step S2 through the intersection of the time dimension and the knowledge dimension in the system dimension include:

[0025] Cost control indicators are generated through the intersection of the planning stage and the economic field.

[0026] Integrability indicators are generated through the intersection of the installation phase and the management phase;

[0027] Safety compliance indicators are generated through the intersection of the operational phase and the legal field.

[0028] Reliability indicators are generated through the intersection of the operational phase and the technical field.

[0029] The recycling stage intersects with the environmental protection field to generate recyclability indicators;

[0030] The performance indicators are extracted from the original stakeholder needs and associated with specific scenarios at each stage of the entire lifecycle.

[0031] This preferred solution generates performance indicators by specifically defining the intersections of each stage in the time dimension and each domain in the knowledge dimension. For example, the intersection of the planning stage and the economic domain forms cost control indicators; the intersection of the installation stage and the management domain forms integrability indicators; the intersection of the operation stage and the legal and technical domains forms safety compliance indicators and reliability indicators, respectively; and the intersection of the recycling stage and the environmental protection domain forms recyclability indicators. This effectively alleviates the problem of performance indicators being disconnected from the specific scenarios throughout the system's entire lifecycle. Simultaneously, this solution directly links the indicator generation process with the original stakeholder needs and the actual scenarios of the corresponding lifecycle stages, helping to enhance the traceability of indicator sources and their relevance under constraints at different stages and domains, making the performance indicators more aligned with the core concerns of the system throughout its entire lifecycle, from planning to recycling.

[0032] Preferably, the SPIT method based on the social layer—practice layer—information layer—technology layer, as described in step S3, refines the functional / performance indicators in four dimensions, including:

[0033] Social-level indicators are used to assess whether the system solution meets the needs of different stakeholders, and to make the reliability, compliance, and testability of the top-level performance indicators explicit, as well as to establish connections with specific stakeholders;

[0034] Practice-level metrics are used to evaluate the implementation of functions and performance, including performance metrics and the correctness of model building, as well as the refinement of non-functional metrics.

[0035] Information layer metrics are used to evaluate the consistency, integrity, and traceability of model information and data transmission processes;

[0036] Technical layer metrics are used to evaluate the technical specification compliance when mapping logical components to physical components and to describe the quantifiable parameters of physical devices.

[0037] This preferred approach refines functional / performance indicators layer by layer through four dimensions: social, practical, informational, and technical. This effectively alleviates the shortcomings of traditional methods in defining paths for non-functional indicators and multi-dimensional constraint indicators. Specifically, the social layer explicitly defines reliability, compliance, and testability from top-level performance indicators and links them to specific stakeholders, enhancing the traceability of indicators to diverse needs. The practical layer introduces model correctness and refinement of non-functional indicators while evaluating functional implementation and performance, supporting model-driven indicator evaluation. The informational layer focuses on the consistency, completeness, and traceability of model information and data transmission, improving the reliability of indicator transmission across different stages. The technical layer focuses on the technical specification compliance and quantitative parameters when mapping logical components to physical components, providing clearer indicator criteria for the connection between the design and implementation phases. Overall, this four-layer refinement method improves the structure and comprehensive coverage of functional / performance indicators in the design process of complex systems.

[0038] Preferably, the detailed non-functional indicators in the practice layer indicators include the following eight categories: usability indicators, installability indicators, maintainability indicators, safety indicators, reliability indicators, economic cost indicators, implementability indicators, and environmental protection requirement indicators;

[0039] The reliability index is further decomposed into sub-indicators such as battery cell cycle life, controller failure rate, and connector contact reliability; the maintainability index is further decomposed into sub-indicators such as fault detection coverage, redundancy switching success rate, and ground-based on-orbit fault location time.

[0040] This preferred approach effectively alleviates the problem of insufficient granularity of non-functional indicators in traditional methods, which makes them difficult to directly apply to subsequent technical indicator conversion, by clearly defining eight categories of non-functional indicators at the practical level and further decomposing reliability and maintainability indicators into sub-indicators. Specifically, the reliability indicator is decomposed into sub-indicators such as battery cell cycle life, controller failure rate, and connector contact reliability, which helps to implement the system's reliability requirements onto specific components or features. The maintainability indicator is decomposed into sub-indicators such as fault detection coverage, redundancy switching success rate, and ground-based on-orbit fault location time, which can support multi-dimensional maintainability evaluation during the system operation and maintenance phase. Overall, this refinement and decomposition method enhances the operability and quantifiable basis of non-functional indicators in the design process of complex systems, providing clearer input for subsequent mapping to physical component technical indicators.

[0041] Preferably, the information layer indicators include one or more of the following indicators:

[0042] The system requires traceability metrics between the various stages of the RFLP process;

[0043] The completeness of the positive decomposition chain from performance indicators to technical indicators;

[0044] Consistency indicators for interdisciplinary parameter exchange;

[0045] The traceability indicators for the construction process of the indicator system record the person who defined the indicator, the definition time, the source of the requirement, the supporting documents, and the change history.

[0046] This preferred solution effectively alleviates the problem of lack of full-process traceability and information consistency assurance in the indicator system in traditional methods by introducing traceability indicators for system requirements across each stage of RFLP, integrity indicators for the positive decomposition chain from performance indicators to technical indicators, consistency indicators for cross-disciplinary parameter exchange, and traceability indicators for the indicator system construction process covering the definer, definition time, requirement source, supporting documents, and change history. Specifically, the first two types of indicators help enhance the integrity and continuity of indicators in the vertical transmission of the RFLP process; the cross-disciplinary parameter exchange consistency indicators facilitate information alignment in multidisciplinary collaborative design; and the construction process traceability indicators provide a clearer record basis for indicator version management and source review. Overall, this solution improves the transparency and auditability of the indicator system in the design process of complex systems.

[0047] Preferably, the conversion of the functional / performance indicators corresponding to the smallest indivisible logical unit after refinement by the SPIT method in step S4 into technical indicators for physical components needs to meet the following principles:

[0048] Each technical indicator must correspond to a specific physical component attribute or parameter and be verifiable through testing or analysis.

[0049] When the logical unit corresponding to a function / performance indicator cannot be further subdivided into smaller independent functional units, the function / performance indicator of the logical unit is converted into the technical indicator of the physical component corresponding to the logical unit.

[0050] The bidirectional mapping relationship includes: a forward mapping relationship, used to record the source relationship between performance indicators and stakeholder requirements, the correspondence between functional / performance indicators and functional modules and logical units, and the correspondence between technical indicators and physical components; and a reverse mapping relationship, used to support bottom-up verification of the satisfaction of technical indicators, functional / performance indicators, and performance indicators starting from physical components.

[0051] This preferred solution effectively alleviates the problems of unclear relationships between indicators and architectural units, and broken traceability and verification paths in traditional methods by clarifying two core principles for converting functional / performance indicators into technical indicators, and establishing a bidirectional mapping relationship including forward and reverse mapping. Specifically, the conversion principle requires that each technical indicator correspond to specific physical component attributes or parameters and possess testability or analyzability, which helps enhance the verifiability and feasibility of the indicators at the implementation level. Using the smallest indivisible logical unit as the conversion node provides clearer boundaries and basis for the transition of indicators from the logical layer to the physical layer. Based on this, forward mapping records the correspondence between indicators and requirement items, functional modules, logical units, and physical components, while reverse mapping supports bottom-up verification of the satisfaction of indicators at each layer from physical components, together forming an indicator traceability and closed-loop verification mechanism that connects the entire process of requirements—function—logic—physical.

[0052] Preferably, the multi-dimensional and multi-layered output index system described in step S4 is visualized by constructing a module definition diagram, a parameter diagram, and designing a structure matrix:

[0053] The module definition diagram is used to define the hierarchical structure and indicator attributes of the system, including: creating a top-level system module and adding performance indicators as value attributes, creating sub-modules and adding corresponding performance indicators as value attributes, and creating physical component modules and adding technical indicator value attributes.

[0054] The parameter diagram is used to describe the mathematical calculation relationship between indicators, including: defining constraint blocks to describe the mathematical relationship between indicators, instantiating the constraint blocks in the parameter diagram, and binding the parameters of the constraint blocks to the value attributes of the corresponding modules in the module definition diagram;

[0055] The design structure matrix is ​​used to visualize the mapping relationship between indicators and architectural units. Indicators at different levels are used as rows of the matrix, and stakeholder needs, functional modules, logical units, and physical components are used as columns of the matrix to discover breakpoints or redundancies in the indicator decomposition chain.

[0056] This preferred solution visualizes a multi-dimensional, multi-layered indicator system using one or more of the following methods: module definition diagrams, parametric diagrams, and design structure matrices. This effectively alleviates the problems of unintuitive indicator system structure, unclear relationships between indicators, and difficulty in identifying mapping breakpoints found in traditional methods. Specifically, the module definition diagram, by defining top-level system modules and performance indicators, sub-modules and performance indicators, and physical component modules and technical indicator value attributes, helps to form a hierarchical indicator view corresponding to the RFLP process. The parametric diagram, through constraint blocks, defines the mathematical calculation relationships between indicators and binds them to module value attributes, clearly expressing the derivation and dependency logic of indicators. The design structure matrix, with indicators as rows and stakeholder needs and various architectural units as columns, facilitates the identification of breakpoints or redundancies in the indicator decomposition chain, providing visual support for the integrity and consistency analysis of the indicator system. Overall, this visualization solution improves the clarity of the indicator system's expression and the ease of analysis.

[0057] Beneficial Effects: Compared with existing technologies, this invention has the following significant advantages: 1. This invention maps the indicator layer to the three stages of RFLP (requirement definition, functional / logical architecture, and physical architecture), and establishes a positive mapping and reverse traceability relationship from performance indicators to technical indicators, alleviating the problems of disconnection, single dimension, and lack of bidirectional mapping, and realizing process integration, full life cycle coverage, and closed-loop traceability; 2. Based on the improved Hall system engineering method, with the requirement definition node as the axis, it cross-defines through the time dimension (planning, design, operation, recycling, etc.) and the knowledge dimension (economic, management, technical, environmental protection, etc.), alleviating the problem of single horizontal dimension and inability to support full life cycle assessment; 3. Based on the SPIT method, the practice layer incorporates eight types of non-functional indicators, such as availability, maintainability, reliability, economic cost, and environmental protection requirements, into a unified framework and can be further quantified and decomposed, alleviating the problem of lack of definition path for non-functional indicators; 4. By recording the traceability, decomposition completeness, and change history of the information layer indicators, and using the design structure matrix to visualize the mapping relationship, breakpoints or redundancies can be quickly discovered, improving the verifiability and reusability of the indicator system. Attached Figure Description

[0058] Figure 1 This is a schematic diagram of the method flow of the present invention;

[0059] Figure 2 This is a schematic diagram of the multi-level indicator system defined by the RFLP system analysis process of the present invention;

[0060] Figure 3 This is a three-dimensional structural schematic diagram of the improved Hall system engineering method of the present invention;

[0061] Figure 4 A schematic diagram illustrating the principle of defining performance indicators based on the improved Hall system engineering method of the present invention;

[0062] Figure 5 This is a schematic diagram of the framework of the multidimensional and multi-layer index system construction method of the present invention. Detailed Implementation

[0063] The technical solution of the present invention will be further described below with reference to the accompanying drawings.

[0064] This invention provides a method for constructing a multi-dimensional, multi-layered indicator system for complex system design. Based on the MBSE Hall system engineering method, using the RFLP system analysis process as the vertical layering basis, and the SPIT method as the horizontal dimension refinement tool, it establishes an automated construction process from stakeholder needs input to the output of a complete indicator system. The main steps include:

[0065] Step S1: Metric Hierarchy Based on the RFLP Process. Based on the RFLP process followed in system design, the metric system is divided into three levels: performance metrics, functional / performance metrics, and technical metrics. The performance metrics level corresponds to the requirements definition phase and measures whether the delivered system meets the top-level task requirements of stakeholders. The functional / performance metrics level corresponds to the functional and logical architecture design phases and is used to evaluate the system's functional implementation and performance. The technical metrics level corresponds to the physical architecture design phase and describes the technical specifications and parameters required for physical components to achieve the functional / performance metrics.

[0066] Step S2: Defining Performance Indicators Based on the Improved Hall Systems Engineering Approach. In the requirements definition phase, using the "requirements definition" node of the system dimension as the axis, and combining the time and knowledge dimensions of the Hall systems engineering approach, performance indicators covering the entire system lifecycle are defined. Specifically, the entire system lifecycle is divided into stages such as planning, design, production, installation, operation, and recycling (time dimension), and the knowledge domains involved in each stage are identified, such as economics, management, law, environmental protection, and technology (knowledge dimension). The intersection of the time and knowledge dimensions is used to generate performance indicators for the corresponding stages. These performance indicators are directly mapped to stakeholder needs and used for top-level task satisfaction assessment.

[0067] Step S3: Define and refine functional / performance metrics based on the SPIT method. During the functional and logical architecture design phases, refine the metrics for the social, practical, informational, and technical layers using the SPIT layering method. Specifically, social layer metrics assess whether the system solution meets the needs of different stakeholders; practical layer metrics assess the implementation of functions and performance, including functional and non-functional metrics (availability, installability, maintainability, security, reliability, economic cost, implementability, environmental requirements, etc.); informational layer metrics assess the consistency, integrity, and traceability of model information and data transmission; and technical layer metrics assess the technical specification compliance when mapping logical components to physical components.

[0068] Step S4: Technical Indicator Definition and Indicator System Output. At the end of the logical architecture design phase and the physical architecture design phase, the functional / performance indicators corresponding to the smallest indivisible logical units after SPIT refinement are converted into technical indicators oriented towards physical components. This ultimately forms a complete indicator system, with design task requirements as input and a combination of performance indicators, non-functional and functional indicators, and technical indicators as output. Furthermore, a multi-dimensional, multi-layered indicator system model can be constructed and output using methods such as module definition diagrams, parameter diagrams, and design structure matrices.

[0069] The method of the present invention will be described in detail below with reference to the accompanying drawings and using a satellite energy subsystem as an example.

[0070] Step S1: Metric hierarchy division based on RFLP process. For example... Figure 1 As shown, this invention first clarifies the RFLP process followed in complex system design: Requirements Definition → Functional Architecture Design → Logical Architecture Design → Physical Architecture Design. The first stage of the RFLP method is requirements definition, including defining and managing requirements, and implementing requirement definition, traceability, and verification. The second stage is functional architecture design, mapping corresponding functional points according to design specifications and requirements, generating a description of the capabilities and services provided by the system to meet those requirements. The third stage is logical architecture design, forming abstract units capable of executing system functions without requiring technical implementation, describing the dynamic behavior of the system and the logical relationships between its components. The fourth stage is physical architecture design, defining physical devices with detailed parameters and their relationships, representing a concrete and feasible physical solution for implementing the logical architecture. The RFLP analysis method has been validated in the construction of indicators and system modeling in complex product development.

[0071] like Figure 2 As shown, based on this, the present invention divides the indicator system into three levels, corresponding to different stages of the RFLP process:

[0072] Performance metrics layer: Output during the requirements definition phase, directly reflecting the top-level task requirements of stakeholders. Performance metrics refer to indicators that can directly reflect the system's task completion status and clarify whether the system delivery meets the needs of stakeholders. They are usually extracted from stakeholder needs and supported by performance metrics such as cost, delivery, lifespan, and operating cycle based on the full lifecycle use cases drawn during the requirements definition process.

[0073] Functional / Performance Metrics Layer: Produced during the functional and logical architecture design phases, these metrics are used to evaluate the implementation and performance of system functions. Once subsystems are defined, performance metrics are also divided according to the overall requirements, generating functional and performance requirement metrics for multiple subsystems. Functional metrics generally refer to the effectiveness of completing a certain function; performance metrics are descriptions that support the achievement of performance metrics, representing the physical or functional attributes related to system operation.

[0074] Technical Specifications Layer: Produced during the physical architecture design phase, this layer describes the technical specifications required for physical components to achieve their functional / performance metrics. When a logical unit cannot be further subdivided, its functional and performance metrics are decomposed into technical specifications oriented towards the physical unit. These technical specifications are typically used to measure the feasibility of physical components in achieving the functional and performance metrics.

[0075] This hierarchical division creates a one-to-one correspondence between the indicator system and the design process, ensuring that each indicator has a clear design stage to which it belongs.

[0076] Step S2: Define the performance index based on the improved Hall system engineering method. For example... Figure 3 As shown, this invention introduces the Hall systems engineering method and improves it according to the entire lifecycle process of complex system design, thereby generating performance indicators. The Hall systems engineering method is a way to solve problems in the planning, organization, and management of large-scale engineering systems. It mainly constructs solutions from three dimensions: time, logic, and knowledge. Based on the characteristics of complex system design, this invention changes the logical maintenance in the original Hall systems engineering to a system dimension, which includes five stages: "requirements definition—functional architecture—logical architecture—physical architecture—verification and validation." Verification and validation is the verification and implementation stage of the indicators, which will not be discussed here. Furthermore, this invention combines the MBSE method, collectively referring to the design and development stages of the time dimension in the original Hall systems engineering method as the R&D stage, and adjusting the final update stage to the obsolescence stage, to adapt to the full lifecycle management of complex systems.

[0077] In this invention, the time dimension is further refined into the planning stage, design stage, production stage, installation stage, operation stage, and recycling / disposal stage. The knowledge dimension, based on the five fundamental areas of economics, management, law, technology, and environmental protection, can be further refined according to system requirements. For example, the economic field can be broken down into sub-fields such as cost control, investor scope, and investment return cycle. Figure 4 As shown, the performance indicators are defined using a cross-dimensional approach of "knowledge dimension × time dimension": taking the "demand definition" stage of the system dimension as the axis, the intersection of a certain lifecycle stage of the time dimension and a certain domain of the knowledge dimension is used to generate the corresponding performance indicators. Specifically, for the satellite energy subsystem:

[0078] Planning Phase × Economic Knowledge Dimension: Cost Control Indicator – “The total development cost of the satellite energy subsystem shall not exceed RMB 8 million.” The “energy subsystem development cost” indicator directly reflects the economic feasibility requirements of the planning phase. If it exceeds the budget, the plan needs to be adjusted or more economical component selections should be adopted.

[0079] Installation Phase × Management Knowledge Dimension: Integrability Indicator – “Matching deviation of the energy subsystem and the overall satellite mechanical interface ≤ 0.5mm”. The “interface alignment deviation” indicator reflects the degree of matching between the energy subsystem and the overall satellite structure during the installation phase. Excessive deviation may lead to damage to electrical connectors or stress concentration.

[0080] Operational Phase × Legal Knowledge Dimension: Safety Compliance Indicator – “The energy subsystem complies with relevant international regulations on satellite orbital debris mitigation.” The “compliance” indicator ensures that the system will not generate uncontrolled debris or interfere with the orbital environment during its on-orbit operation.

[0081] Operational Phase × Technical Knowledge Dimension: Reliability Indicator – “Battery on-orbit design life not less than 8 years.” The “design life” indicator reflects the long-term operational capability of the energy subsystem; if it falls below the requirement, the battery quality level needs to be improved or redundancy design needs to be added.

[0082] Recycling Phase × Environmental Knowledge Dimension: Recyclability Indicator – “Battery material recyclability rate is no less than 95%”. The “recyclability rate” indicator ensures that after the satellite’s lifespan ends, the battery materials will not cause persistent environmental pollution, meeting the requirements of sustainable development.

[0083] The aforementioned performance indicators are extracted from the original stakeholder needs and associated with specific scenarios at each stage of the entire lifecycle, ensuring the systematic nature and completeness of the indicators. For example, the performance indicator of "energy supply reliability" can be decomposed into two sub-indicators: "energy supply reliability in the sunlit area" and "energy supply reliability in the shaded area." The reliability in the shaded area is further reflected in functional / performance indicators such as "battery performance," "charge and discharge management performance," and "power distribution performance."

[0084] Step S3: Definition and Dimensional Refinement of Functional / Performance Indicators Based on the SPIT Method. In the functional and logical architecture design phases, this invention employs the SPIT method to refine functional / performance indicators across four dimensions. The SPIT method decomposes extremely complex systems into four abstract layers: social, practical, informational, and technical. This helps to understand the common characteristics of the system and guide the detailed construction of the indicator system. Specifically, the social layer focuses on creating a shared and understandable model, determining the system's behavior and actions from top to bottom, and includes the concepts of people or organizations; the practical layer focuses on providing end-to-end practices that deliver strategies and achieve results, belonging to the system's process layer and including the system's principles; the informational layer focuses on providing the right information to the right people at the right time and ensuring they can understand and trust it; and the technical layer focuses on methods that support system development, are compatible, usable, updatable, and reusable, including equipment, facilities, methods, and implementation logic. In the functional and logical architecture design phases, this invention employs the SPIT method to refine functional / performance indicators across four dimensions, with each layer focusing on different key indicators.

[0085] (1) Social-level indicators are mainly used to assess whether the overall solution meets the needs of different stakeholders, and are a further development of performance indicators at the functional / performance level. For the satellite energy subsystem, social-level indicators can be defined as: whether the energy subsystem meets the owner's (satellite overall unit) requirements for energy supply for the orbital mission (e.g., zero power outages), whether it meets the operator's (satellite on-orbit management team) requirements for telemetry data observability (e.g., no fewer than 4 battery SOC monitoring points), whether it meets the regulatory agency's (China National Space Administration, International Telecommunication Union) requirements for battery pack passivation (e.g., battery discharge to a safe voltage after the mission), and whether it meets the development unit's (power subsystem supplier) requirements for production and testing convenience (e.g., standardized test interfaces). These social-level indicators make explicit the contents of the top-level performance indicators such as "reliability," "compliance," and "testability," and establish a connection with specific stakeholders.

[0086] (2) The practical layer indicators evaluate the implementation of functions and performance, which includes two core aspects. The first aspect is performance indicators, which evaluate whether the functions have been implemented and the extent of implementation. For example, whether the satellite energy subsystem can complete the expected functions such as power generation, energy storage, power management, and power distribution, as well as the accuracy and efficiency of these functions. The second aspect is the correctness indicators of model construction, which ensure that the existing model architecture is constructed in the correct way. For example, completeness indicators characterize whether the model construction is complete, and correctness indicators characterize whether the model construction method is correct.

[0087] Furthermore, the practical layer also includes a refinement of non-functional indicators. Combining the non-functional indicators of software development with the characteristics of complex systems, non-functional indicators mainly include eight categories: availability (e.g., "discharge depth in the shaded area does not exceed 40%"), installability (e.g., "battery pack installation gap is not less than 10mm"), maintainability (e.g., "fault switching time ≤ 30s"), safety (e.g., "automatic reset capability to cope with single-event lockout"), reliability (e.g., "battery on-orbit design life is not less than 8 years" and "annual degradation rate of solar cell array output power ≤ 2%)), economic cost (e.g., "single unit cost ≤ 800,000 yuan"), feasibility (e.g., "batch production yield of monocrystalline silicon solar cells ≥ 95%)"), and environmental protection requirements (e.g., "battery materials are mercury-free and cadmium-free"). These non-functional indicators are decomposed in a layer-by-layer refinement manner. For example, the "reliability" indicator can be further decomposed into sub-indicators such as battery cell cycle life, controller failure rate, and connector contact reliability; the "maintainability" indicator can be further decomposed into fault detection coverage, redundancy switching success rate, and ground-based location time for on-orbit faults.

[0088] (3) Information layer indicators assess the consistency of data, information, and knowledge, focusing on consistency and completeness in the information transmission process. For satellite energy subsystems, information layer indicators can be defined as: the traceability of energy demand across different stages of RFLP (e.g., whether the total dark-state power demand ≤3kW can be traced back to the power capacity design in the functional architecture, then to the power allocation of each power module in the logical architecture, and finally to the selection of controller power switch capacity in the physical architecture), the completeness of the positive decomposition chain from performance indicators to technical indicators (e.g., whether there is a complete decomposition path for "discharge depth ≤40%" that ultimately leads to specific technical parameters such as "battery capacity 80Ah" and "power consumption 300W in shadow mode"), the consistency of cross-disciplinary parameter exchange (e.g., whether the power parameters exchanged between the energy subsystem and the attitude control subsystem are consistent), and the traceability of the indicator system construction process (e.g., recording the person who defined the indicator, the definition time, the source of the demand, the supporting documents, and the change history). The core role of information layer indicators is to ensure that, in the process of multidisciplinary collaborative design of complex systems, indicator information will not cause design conflicts due to transmission distortion.

[0089] (4) Technical layer indicators assess whether the R&D process conforms to technical specifications. This is a dimensional mapping of technical indicators, focusing on the technical specification compliance when mapping logical components to physical components. Technical layer indicators typically describe quantifiable parameters of physical equipment, such as the accuracy of the battery controller's charge / discharge cutoff voltage setting, the shunt regulator's response time, and the DC / DC module's input voltage range. Through the SPIT four-layer refinement, functional / performance indicators are no longer just a simple list of performance values, but a comprehensive indicator system that includes stakeholder dimensions, functional and non-functional dimensions, information consistency dimensions, and technical compliance dimensions, providing complete input for subsequent technical indicator definitions.

[0090] Step S4: Technical Indicator Definition and Indicator System Output. At the end of the logical architecture design phase and the physical architecture design phase, the functional / performance indicators corresponding to the smallest indivisible logical unit after SPIT refinement are converted into technical indicators for physical components. Technical indicators describe the quantifiable parameters of physical devices and are used to measure the feasibility of physical components implementing the functional and performance indicators. The definition of technical indicators must meet the following principles: each technical indicator must correspond to a specific physical component attribute or parameter and can be verified through testing or analysis; the decomposition of technical indicators should follow the boundary of the "smallest logical unit," that is, when the logical unit corresponding to the functional / performance indicator cannot be further subdivided into smaller independent functional units, the functional / performance indicator of that logical unit is converted into the technical indicator of the physical component corresponding to that logical unit.

[0091] Taking the satellite energy subsystem as an example, the conversion process from functional / performance indicators to technical indicators is as follows. The functional / performance indicator "maximum depth of discharge of the battery ≤ 40%" corresponds to the "depth of discharge calculation logic unit" and "energy management logic unit" in the logical architecture. Converting it into physical component technical indicators includes: battery rated capacity (≥ 80Ah, corresponding to the design selection parameters of the physical component "battery pack"), battery cell operating voltage platform (3.6V, corresponding to the electrochemical system parameters of the "battery cell"), maximum energy consumption in the shadow area (determined by the sum of the power consumption and duration of each working mode, corresponding to the power parameters of each load), battery charging and discharging efficiency (≥ 90%, corresponding to the efficiency parameters of the "battery" and "charge and discharge controller"), and battery discharge cut-off voltage (accuracy requirements of the set value, corresponding to the control parameters of the "battery management unit"). For example, the functional / performance indicator "battery on-orbit life ≥ 8 years" is converted into: battery cell cycle life (≥ 2000 cycles at 80% DOD depth, corresponding to the cycle life parameter of "battery cell"), battery calendar life (≥ 10 years, including ground storage and on-orbit operation, corresponding to the aging characteristic parameter of "battery"), battery self-discharge rate (≤ 3% / month, corresponding to the storage performance parameter of "battery"), and battery operating temperature range (-10℃~45℃, corresponding to the environmental adaptability parameter of "battery").

[0092] Based on the technical indicator definitions, a two-way mapping relationship between indicators and architectural units is established. The forward mapping records: which stakeholder requirements the performance indicators originate from (e.g., "battery discharge depth ≤ 40%" originates from the "Satellite Energy System Requirements" in the mission requirements document and the shadowed area condition analysis results in the orbital analysis report); which functional modules and logical units correspond to the functional / performance indicators (e.g., "maximum battery discharge depth ≤ 40%" corresponds to the "battery capacity design function" in the functional architecture and the "discharge depth calculation logic unit" in the logical architecture); and which physical components correspond to the technical indicators (e.g., "battery capacity ≥ 80Ah" corresponds to the "battery pack module" in the physical architecture, and "MPPT tracking accuracy ≥ 99%" corresponds to the "MPPT controller module"). The reverse mapping supports upward verification starting from physical components: if a certain technical indicator (e.g., the measured battery capacity is 82Ah) meets the requirements, then it verifies upward whether the supported functional / performance indicators (maximum battery discharge depth ≤ 40%) can be met; if the functional / performance indicators are met, then it further verifies upward whether the supported performance indicators (energy supply reliability) can be met. When a certain technical indicator is not met, the affected higher-level indicator and stakeholder needs can be quickly located through the reverse mapping chain, thereby guiding design corrections.

[0093] Finally, as Figure 5As shown, the multi-dimensional and multi-layer indicator system definition method proposed in this invention divides the indicator levels based on the RFLP process, expands the dimensions of the indicator system based on the improved Hall system engineering method and SPIT method, and forms a complete indicator system with design task requirements as input and a combination of performance indicators, non-functional and functional indicators, and technical indicators as output.

[0094] Furthermore, the multi-dimensional, multi-layered indicator system can be visualized by constructing module definition diagrams, parameter diagrams, and designing structure matrices. Module definition diagrams are used to define the hierarchical structure and indicator attributes of the system. In the modeling tool, a top-level system module, "Satellite Energy Subsystem," is first created. Performance indicators are added as value attributes to this module, such as adding a "Depth of Discharge" value attribute with its type set to Real and unit %, and adding a "System Mass" value attribute with unit set to kg. Then, sub-modules such as "Solar Cell Array," "Energy Storage Battery," "Power Controller," and "Power Distribution Unit" are created, adding corresponding performance indicators as value attributes to each sub-module. For example, "Rated Capacity," "Depth of Discharge Limit," and "Cycle Life" value attributes are added to the "Energy Storage Battery" module. Further finer-grained physical component modules such as "Battery Cell," "DC / DC Module," and "MPPT Controller" are created, adding technical indicator value attributes to these modules. Through the module definition diagram, a strict correspondence can be formed between the hierarchical structure of the indicators and the composition structure of the system.

[0095] Parametric graphs are used to describe the mathematical relationships between indicators and are a core tool connecting the indicator decomposition chain. When constructing a parametric graph, constraint blocks are first defined to describe the mathematical relationships between indicators. For example, a constraint block "DoDConstraint" is defined, containing parameters E_shadow (shadow area power consumption), C_bat (battery capacity), DoD (depth of discharge), and η (discharge efficiency), with the constraint expression DoD = E_shadow / (C_bat × η), used to verify whether the depth of discharge meets the ≤40% limit. Then, the constraint block is instantiated in the parametric graph, and the parameters of the constraint block are bound to the value attributes of the corresponding modules in the module definition graph. For example, C_bat is bound to the "rated capacity" attribute of the "energy storage battery" module, E_shadow is bound to the sum of the products of power consumption and shadow area duration in each operating mode (implemented through nested constraint blocks), and DoD is bound to the "depth of discharge" attribute of the top-level "satellite energy subsystem" module. Through the parametric graph, the computational relationships between indicators are explicitly modeled, supporting subsequent automatic solution verification.

[0096] The Design Structure Matrix (DSM) is used to visualize the mapping relationship between metrics and architectural units. When constructing a DSM, metrics at different levels are used as rows (or columns) of the matrix, and structural elements such as stakeholder needs, functional modules, logical units, and physical components are used as columns (or rows). Through the DSM, breakpoints (e.g., a performance metric lacking a corresponding functional module) or redundancies (e.g., a technical metric mapping to multiple unrelated logical units) in the metric decomposition chain can be clearly identified. Furthermore, the DSM supports simple change impact analysis: when a metric changes, the DSM can quickly identify the affected system components and stakeholder needs.

Claims

1. A method for constructing a multi-dimensional, multi-layered index system for complex system design, characterized in that, Includes the following steps: S1: Based on the RFLP process of requirement definition-functional architecture-logical architecture-physical architecture, the indicator system to be built is divided into the performance indicator layer, the functional / performance indicator layer and the technical indicator layer. S2: In the requirement definition phase, based on the improved Hall system engineering method, taking the requirement definition node of the system dimension in the improved Hall system engineering method as the axis, the performance indicators covering the entire life cycle of the system are covered by the cross definition of the time dimension and knowledge dimension in the improved Hall system engineering method; wherein, the improved Hall system engineering method is obtained by changing the logical maintenance of the original Hall system engineering method to the system dimension. S3: In the functional and logical architecture design phase, based on the SPIT method (Social Layer—Practice Layer—Information Layer—Technology Layer), the functional / performance indicators in the functional / performance indicator layer are refined in four dimensions. S4: At the end of the logical architecture design phase and the physical architecture design phase, the functional / performance indicators corresponding to the smallest indivisible logical unit after refinement by the SPIT method are converted into technical indicators for physical components. A bidirectional mapping relationship is established between the performance indicators, the functional / performance indicators, the technical indicators and the architectural units included in each stage of the RFLP process, thereby outputting a multi-dimensional and multi-layer indicator system.

2. The method according to claim 1, characterized in that, The performance indicator layer mentioned in step S1 corresponds to the requirement definition phase in the RFLP process. It is used to measure whether the system delivery meets the top-level task requirements of stakeholders. The performance indicators are extracted from the needs of stakeholders and generated based on the full lifecycle use cases. The functional / performance metrics layer corresponds to the functional architecture and logical architecture design phases in the RFLP process. It is used to evaluate the system's functional implementation and performance. Functional metrics are used to evaluate the effectiveness of completing a certain function, while performance metrics are used to describe the measurement of physical or functional attributes related to system operation. The technical specification layer corresponds to the physical architecture design phase in the RFLP process and is used to describe the technical specifications and parameters required for physical components to achieve functional / performance indicators.

3. The method according to claim 1, characterized in that, The improved Hall system engineering method described in step S2 includes: The system dimensions are set to include five stages: requirements definition, functional architecture, logical architecture, physical architecture, and verification and validation. The time dimension is further refined into the planning stage, design stage, production stage, installation stage, operation stage, and recycling stage; The knowledge dimension is set as including five basic areas: economics, management, law, technology, and environmental protection. Using the requirement definition stage of the system dimension as the axis, the intersection of any life cycle stage of the time dimension and any domain of the knowledge dimension is taken to generate an effectiveness index corresponding to the selected life cycle stage and domain, and the effectiveness index is directly mapped to the needs of stakeholders.

4. The method according to claim 3, characterized in that, The granularity of the time dimension and the knowledge dimension is adjusted according to the characteristics of the system: for short-cycle systems, some stages in the time dimension are merged, and for specific domains, domains in the knowledge dimension are added; the economic domain in the knowledge dimension is further decomposed into sub-domains of cost control, investment return range, and investment return cycle.

5. The method according to claim 3, characterized in that, Step S2 describes defining performance indicators covering the entire lifecycle of the system through the intersection of the time dimension and the knowledge dimension within the system dimension. These indicators include: Cost control indicators are generated through the intersection of the planning stage and the economic field. Integrability indicators are generated by the intersection of the installation phase and the management phase; Safety compliance indicators are generated through the intersection of the operational phase and the legal field. Reliability indicators are generated through the intersection of the operational phase and the technical field. The recycling stage intersects with the environmental protection field to generate recyclability indicators; The performance indicators are extracted from the original stakeholder needs and associated with specific scenarios at each stage of the entire lifecycle.

6. The method according to claim 1, characterized in that, Step S3, based on the SPIT method (social layer—practice layer—information layer—technology layer), refines functional / performance indicators across four dimensions, including: Social-level indicators are used to assess whether the system solution meets the needs of different stakeholders, and to make the reliability, compliance, and testability of the top-level performance indicators explicit, as well as to establish connections with specific stakeholders; Practice-level metrics are used to evaluate the implementation of functions and performance, including performance metrics and the correctness of model building, as well as the refinement of non-functional metrics. Information layer metrics are used to evaluate the consistency, integrity, and traceability of model information and data transmission processes; Technical layer metrics are used to evaluate the technical specification compliance when mapping logical components to physical components and to describe the quantifiable parameters of physical devices.

7. The method according to claim 6, characterized in that, The detailed non-functional indicators in the practice layer indicators include the following eight categories: usability indicators, installability indicators, maintainability indicators, safety indicators, reliability indicators, economic cost indicators, implementability indicators, and environmental protection requirement indicators. The reliability index is further decomposed into sub-indicators such as battery cell cycle life, controller failure rate, and connector contact reliability; the maintainability index is further decomposed into sub-indicators such as fault detection coverage, redundancy switching success rate, and ground-based on-orbit fault location time.

8. The method according to claim 6, characterized in that, The information layer metrics include one or more of the following metrics: The system requires traceability metrics between the various stages of the RFLP process; The completeness of the positive decomposition chain from performance indicators to technical indicators; Consistency indicators for interdisciplinary parameter exchange; The traceability indicators for the construction process of the indicator system record the person who defined the indicator, the definition time, the source of the requirement, the supporting documents, and the change history.

9. The method according to claim 1, characterized in that, Step S4, which involves converting the functional / performance metrics corresponding to the smallest indivisible logical unit refined by the SPIT method into technical metrics oriented towards physical components, needs to meet the following principles: Each technical indicator must correspond to a specific physical component attribute or parameter and be verifiable through testing or analysis. When the logical unit corresponding to a function / performance indicator cannot be further subdivided into smaller independent functional units, the function / performance indicator of the logical unit is converted into the technical indicator of the physical component corresponding to the logical unit. The bidirectional mapping relationship includes: a forward mapping relationship, used to record the source relationship between performance indicators and stakeholder requirements, the correspondence between functional / performance indicators and functional modules and logical units, and the correspondence between technical indicators and physical components; and a reverse mapping relationship, used to support bottom-up verification of the satisfaction of technical indicators, functional / performance indicators, and performance indicators starting from physical components.

10. The method according to claim 1, characterized in that, The multi-dimensional and multi-layered output index system described in step S4 is visualized by constructing a module definition diagram, a parameter diagram, and designing a structure matrix in one or more ways: The module definition diagram is used to define the hierarchical structure and indicator attributes of the system, including: creating a top-level system module and adding performance indicators as value attributes, creating sub-modules and adding corresponding performance indicators as value attributes, and creating physical component modules and adding technical indicator value attributes. The parameter diagram is used to describe the mathematical calculation relationship between indicators, including: defining constraint blocks to describe the mathematical relationship between indicators, instantiating the constraint blocks in the parameter diagram, and binding the parameters of the constraint blocks to the value attributes of the corresponding modules in the module definition diagram; The design structure matrix is ​​used to visualize the mapping relationship between indicators and architectural units. Indicators at different levels are used as rows of the matrix, and stakeholder needs, functional modules, logical units, and physical components are used as columns of the matrix to discover breakpoints or redundancies in the indicator decomposition chain.

Images