Interactive mechanical product design change method based on user operation behavior driving

Abstract

The application is an interactive mechanical product design change method based on user operation behavior driving, including the following steps: studying user operation behavior and product behavior under different function scenarios by means of dynamic element analysis method, recording user operation behavior of users using products, deducing corresponding parts from unreasonable parts in user operation behavior to obtain design change source; supplementing the reasonable part of user operation behavior into object process methodology (OPM) as a behavior layer, then combining with FBS model to obtain FBS-OPM product structure correlation model containing user operation behavior; considering user information from three directions of change demand, product structure correlation model and three-party composite importance, introducing user operation behavior factor by combining dynamic element analysis method, FBS, OPM and intelligent optimization algorithm, realizing change propagation path analysis of interactive mechanical products, so as to obtain change propagation path with stronger interaction and more in line with user demand.

Description

Technical Field

[0001] This invention relates to the field of change path propagation technology, and in particular to an interactive mechanical product design change method driven by user operation behavior. Background Technology

[0002] Product design is shifting from a "technology-driven" to a "value-oriented" approach, with a greater emphasis on human-centered design principles and personalized, flexible, and sustainable innovation. Against this backdrop, interactive mechanical products, as a crucial medium connecting humans and industrial systems, have become key targets for human-machine collaborative design and intelligent manufacturing. These products achieve efficient matching between user intent and mechanical response through real-time interaction, leveraging human creativity and judgment while enhancing the automation and intelligence of the system. However, with the diversification of interaction needs and the increasing complexity of usage scenarios, companies need to continuously optimize interactive performance during product iteration to adapt to user behavior and feedback. Engineering and design changes become inevitable in this process. How to ensure technical performance while simultaneously prioritizing user experience has become a pressing issue in the field of mechanical product design.

[0003] Design changes refer to modifications or adjustments made to the original design during product development to adapt to new market conditions, solve problems discovered after product release, or respond to user feedback. These modifications involve significant alterations to the product structure, components, interfaces, or functions to ensure product upgrades and iterations. Existing research on product design changes largely relies on subjective questionnaires, user model construction, and measurement using sensory perception tools to obtain user needs. For example, Wang Xiangbing et al. (Wang Xiangbing, Du Quanbin, Zhou Hang, et al. Research on Modular Design Method of Flexible Product Platform Driven by Customer Needs [J]. Mechanical Design, 2020, 37(09):100-111.DOI:10.13841 / j.cnki.jxsj.2020.09.016.) established a user needs model and set of user needs to obtain design change requirements. They used the AHP method to quantify the characteristics of the needs, but the establishment of the set and the acquisition of relevant weight data still depended on subjective scoring by experts. Therefore, the process of uncovering user needs in the field of design changes is still highly subjective, ambiguous, and uncertain. It is difficult to provide clear target and change requirements solely through subjective analysis methods and verbal expressions in questionnaires. The challenge at this stage lies in how to uncover potential user needs while avoiding interference from subjective factors. Therefore, this invention introduces user operational behavior to address these challenges. Specifically, by collecting and analyzing behavioral data from actual operational processes, the difficulties and strategies exposed by users during use are transformed into quantifiable and traceable evidence, thereby achieving a representation from "subjective feelings" to "objective needs."

[0004] Furthermore, existing research on design change behavior mostly focuses on product behavior rather than user operation behavior, and places more emphasis on Function-Behavior-Structure (FBS) research. For example, Li Congdong et al. (Li Congdong, Zhang Zhiwei, Cao Cejun, et al. Multi-objective optimization of propagation path of complex product design change[J]. Computer Integrated Manufacturing Systems, 2021, 27(03):842-856.DOI:10.13196 / j.cims.2021.03.016.) introduced the FBS model and multiple network theory to express the design change propagation path, so as to effectively improve the prediction accuracy of the design change propagation path. Wang Qiyang et al. (Wang Qiyang, Deng Yimin. Research on the Probability of Component Change in Mechanical Product Redesign Based on Bayesian Network [J]. Mechanical Design and Research, 2023, 39(02):1-4+11.DOI:10.13952 / j.cnki.jofmdr.2023.0046.) believe that the behavioral correlation of mechanical products is related to "flow". They construct an FBS model with the help of energy flow, information flow and material flow to realize the correlation modeling of complex mechanical products. However, the above behaviors are all product behaviors rather than user operation behaviors. The complex correlation between user operation behaviors and product functions and structures is also different from the product behavior modeling method. New methods need to be constructed to introduce user operation behaviors.

[0005] Therefore, this invention objectively obtains user target needs and change requirements by introducing user operation behavior, constructs a correlation model between user operation behavior and product FBS model, and transforms discrete and complex operation behavior data into structured information that can be used for design decisions, thereby supporting the accurate generation and optimization of design change solutions. Summary of the Invention

[0006] To address the shortcomings of existing technologies, the technical problem this invention aims to solve is to propose an interactive mechanical product design change method driven by user operation behavior. This method considers user information from three directions: change requirements, product structure correlation model, and three-party composite importance. It combines dynamic element analysis, FBS, OPM, and intelligent optimization algorithms (such as ant colony algorithm) to introduce user operation behavior factors, thereby realizing change propagation path analysis for interactive mechanical products and obtaining a change propagation path that is more interactive and better meets user needs.

[0007] The technical solution adopted by the present invention to solve the aforementioned technical problem is as follows: An interactive mechanical product design change method based on user operation behavior, the design change method includes the following steps: By using dynamic element analysis, we study user operation behavior and product behavior under different functional scenarios, record user operation behavior when using the product, and deduce corresponding parts from unreasonable parts of user operation behavior to obtain the source of design change. The reasonable parts of user operation behavior are added to the Object Process Methodology (OPM) as the behavior layer, and then combined with the FBS model to obtain the FBS-OPM product structure relationship model that includes user operation behavior. Based on the FBS-OPM product structure association model and the PageRank algorithm, the PR value of the product structure is calculated to obtain the product structure importance of each structure. We acquire online comment data, use the TF-IDF algorithm to calculate the importance of each structure from the user's perspective, and use the importance of the structure as the user's attention level. By examining the company's Bill of Materials (BOM), collecting all time and economic cost data for structural changes, and then normalizing and weighting the data, the change costs for each structure of the product can be obtained. Based on user attention to each structure, product structure importance, and change costs, the entropy weight method is applied to calculate the three-way composite importance of each product structure. Calculate the propagation probability of conditional change, and use the propagation probability of conditional change to combine the three-party composite importance to calculate the propagation intensity of change between adjacent structures; With the goal of minimizing the sum of change propagation intensities of adjacent structures along the change propagation path, and taking the design change source as the starting position of the intelligent optimization algorithm, the intelligent optimization algorithm is used to obtain the optimal structural change propagation path to guide product iterative design.

[0008] Furthermore, the process for calculating the importance of the product structure is as follows: The process-based structural correlation strength evaluation criteria are used to assess the correlation strength of each structure in the product and obtain the correlation weights between each structure. B ij And based on this, calculate the transition probability of two adjacent structures; The transition probability is substituted into the PageRank algorithm to calculate the product structure PR value, and the product structure PR value is used as the importance of the product structure.

[0009] Furthermore, the process for handling the propagation intensity of changes in adjacent structures is as follows: Combined with the aforementioned transition probability calculation structure S i To adjacent structure S j The joint change probability during change propagation is used to calculate the conditional change propagation probability. If the propagation probability of conditional change is 0, then structure S i With adjacent structure S j The propagation strength of the change is 0; if the propagation probability of the conditional change is not 0, then the structure S i With adjacent structure S j The change propagation strength is: the negative of the conditional change propagation probability plus 1, then combined with the adjacent structure S. jThe importance of the three components is multiplied.

[0010] Furthermore, the intelligent optimization algorithm is at least one of ant colony optimization, genetic algorithm, or multi-objective particle swarm optimization algorithm.

[0011] Furthermore, the construction process of the FBS-OPM product structure relationship model is as follows: Step S21: Perform functional analysis on the target product, obtain the functional layer, behavioral layer and structural layer, and construct the FBS model of the target product; Step S22: Construct a behavior-structure mapping for the reasonable parts of the target product user operation behavior in the dynamic element analysis table. Then, using the OPM modeling language, apply the concepts in OPM to couple the functional, behavioral, and structural information of the target product. Construct an information association model with the target product user operation behavior in the dynamic element analysis table as the process and the user and product structure as the object. Establish information association models at different functional layers. Then, combine it with the FBS model to construct a target product FBS-OPM product structure association model that includes user operation behavior.

[0012] Furthermore, the target product is a mountain bike, etc.

[0013] Furthermore, in the motion element analysis method, following the principle of motion economy, user operation behavior is divided into reasonable and unreasonable parts, and the unreasonable parts of user operation behavior are mapped to the product structure to identify the source of design change; the principle of motion economy includes the following four points: reducing the number of actions, working with both hands simultaneously, shortening the action distance, and making actions easy.

[0014] Compared with the prior art, the beneficial effects of the present invention are: (1) This invention obtains design change sources based on objective evidence of user operation behavior. User operation behavior directly determines the product interaction efficiency and usage results. This invention introduces the dynamic element analysis method, starting from the efficiency and operation cost of users completing tasks, to identify unreasonable operations such as inefficiency, repetition, rollback, and accidental triggering that occur in the actual use process, and maps such unreasonable operations to product structural elements, realizing the location of design change sources from "behavioral anomalies to structural orientation". In this way, quantifiable and traceable behavioral data replaces subjective analysis, reducing the subjectivity and uncertainty of the requirement mining process.

[0015] (2) Constructing an FBS-OPM product structure relationship model to incorporate user operation behavior. Traditional FBS can use the behavior layer as a bridge to sort out design information and form a product model, but this method lacks the mapping of user operation behavior. This invention complements the advantages of the FBS model and the OPM method, and uses the conceptual coupling function, behavior and structural information in OPM to construct a relationship model with user operation behavior as the process and users and product structure as objects, so as to realize the relationship modeling of user operation behavior.

[0016] (3) Using the composite importance of the three parties' needs to guide the selection of change propagation paths. Existing importance calculations are mostly based on the relationship between components and the cost of change, which can meet the internal design modification task objectives of enterprises. However, the final evaluation subject of interactive mechanical products is the user. Users have different perceptions of the importance of different structural elements, and their evaluation criteria are closely related to the specific use scenario. If the user's perspective is ignored, the change propagation path may deviate from the actual use effect, or even lead to optimization directions contrary to user expectations. To this end, this invention filters, statistically analyzes, and structures the concerns and pain points of users, designers, and decision-makers, calculates the composite importance of the three parties, and uses the composite importance of the three parties as the basis for filtering change propagation paths, thereby improving the consistency and effectiveness of change solutions with the actual needs of users.

[0017] (4) The present invention targets interactive mechanical products and user operation behavior, which is significantly different from traditional product behavior (interaction behavior between basic functional modules). This application clarifies the difference between user operation behavior and product behavior, adopts dynamic element analysis theory and combines FBS model and OPM method to replace subjective analysis with quantifiable and traceable behavioral data. Using OPM modeling language, the basis for introducing user operation behavior into the FBS model framework is given. Taking user operation behavior as the process and user and product structure as the object, following the principle of process change object state association construction, user operation behavior is introduced into product structure analysis to construct FBS-OPM product structure association model. This model is more accurate and provides more reliable input conditions and decision basis for subsequent change propagation path evaluation and optimal structure change propagation path extraction. Attached Figure Description

[0018] To more clearly illustrate the technical solution of the present invention, a mountain bike is selected as an implementation example. The drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only preferred embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0019] Figure 1This is a schematic diagram of the process framework of one embodiment of the interactive mechanical product design change method driven by user operation behavior according to the present invention.

[0020] Figure 2 This is a schematic diagram of the FBS mapping process for a mountain bike according to an embodiment of the present invention.

[0021] Figure 3 This is a schematic diagram of the FBS-OPM product structure relationship model for mountain bikes according to an embodiment of the present invention.

[0022] Figure 4 This is a schematic diagram of the overall structural modification scheme for the mountain bike product of this invention.

[0023] Figure 5 This is a schematic diagram illustrating the design principle of the mountain bike product of this invention. Detailed Implementation

[0024] The present invention will be further explained below with reference to the embodiments and accompanying drawings, but this is not intended to limit the scope of protection of this application.

[0025] Introduction to the OPM method: Object Process Methodology (OPM) follows the principle of minimal ontology, enabling conceptual modeling of any system in any domain using objects, processes, and relationships, along with corresponding scaling and expansion mechanisms. Its principle is that processes transform object states. From a systems perspective, interactive mechanical products, where humans exist in physical form, can trigger or influence product behavior through user actions. Therefore, user actions can be viewed as processes that change object states, and OPM can incorporate these actions into the Functional Business Model (FBS). On one hand, OPM can expand the behavioral layer based on the interaction characteristics between users and products, incorporating user actions. On the other hand, FBS can logically retrieve design-related information, simplifying the process and reducing the complexity of the OPM modeling process, thus clarifying the design structure.

[0026] Unlike Unified Modeling Language (UML) and System Modeling Language (SysML), OPM provides a unified dual-mode view to comprehensively describe all aspects of a system. It expresses the desired objectives using both OPD diagrams and OPL language, representing the function, behavior, and structure of any system. This satisfies the cognitive "dual-mode assumption," facilitating better understanding of the model by both modelers and users. OPD is a single type of diagram in OPM notation, based on both graphical and natural language to uniformly describe the system's function, behavior, and structure. Because it lacks a defined perspective, the model can be defined and read from multiple dimensions, including descriptions of spatial dimensions, physical properties, and process information, enabling visual modeling. OPL is the textual description mode of OPM, complementing the OPD diagram. By providing a structured natural language description of the OPD diagram, it gains support and understanding from stakeholders and provides a foundation for automatically generating design applications, establishing the basic framework for system description programs.

[0027] Table 1 Definition of Basic Concepts of OPM

[0028] The key attribute of an object is its existence within a certain timeframe. It could be a simple piece of material, a sentence, or a browsing history, or it could be a complex organization, the human brain, or a galaxy. A process, on the other hand, is a continuous state, a series of actions and changes, a method of operation, a state of functioning, or a state of growth; it is often seen as the thing that triggers "transformation." A state belongs to an object and cannot exist independently; it only has meaning when combined with the object it belongs to. Objects possess states in any space and time, while processes have the ability to change the state of an object. Transformation requires a certain amount of time to change something or the state of something; it belongs to the process and is the meaning of the process's existence. In the concrete expression of object-process-method, each element has a fixed mode of expression, including both graphics and language. In graphic representations, objects are represented by rectangles, processes by ellipses, states by rounded rectangles, and connections representing transformations are represented by arrows. Based on the different inherent attributes of each element, they can be further divided: subject links refer to the support links of human-responsible processes; means links refer to the support links of tools required by the process; unidirectional structural links refer to general links of unidirectional transmission relationships; and influence links refer to the transformation links of the process affecting the object.

[0029] This invention fully considers users' dynamic interactive behavior during the process of obtaining change requests. It utilizes dynamic element analysis to obtain the design requirements that need to be changed, and optimizes the product design accordingly. To place users and products in the same dimension for research, the OPM method is introduced as a system platform and combined with FBS to jointly express user operation behavior and product behavior. This not only clarifies the optimization goals, but also helps to analyze the impact of changes caused by user operation behavior on adjacent systems, playing a bridging role.

[0030] The described motion element analysis method involves observing human actions, recording and analyzing these actions using motion element symbols, recording user behavior when using the product, identifying unreasonable actions, mapping them to product structural elements, and then making improvements. Reasonable actions are added to the product structure relationship model as behavioral layers, while unreasonable actions are used to deduce corresponding components, thereby directly identifying the design change source (i.e., the change source part), achieving the location of the design change source from "behavioral anomaly to structural orientation."

[0031] In the change propagation path solving stage, the difference between interactive mechanical products and traditional products in the structural design of each part lies in the fact that users have their own evaluation criteria for the design of each part of the product. New indicators need to be introduced to evaluate user satisfaction with each part of the design, increasing the user's voice in the design process. Therefore, this invention constructs a tripartite composite importance index from the perspectives of users, designers, and decision-makers as an evaluation indicator, striving to find a change propagation path acceptable to all three parties during the change propagation path search process for interactive mechanical products, minimizing the impact of changes.

[0032] Example 1 The specific process of the user-driven interactive mechanical product design change method in this embodiment is as follows: Step S1: Analyze user operation behavior using the motion element analysis method to obtain a user operation behavior motion element analysis table. Analyze the user operation behavior in the motion element analysis table and classify the user operation behavior according to the motion economy principle, dividing it into two categories: unreasonable operation behavior and reasonable operation behavior. Map the unreasonable (such as wasteful) parts of the user operation behavior to the product structure to determine the source of design change. The motion economy principle includes the following four points: reduce the number of actions, work with both hands simultaneously, shorten the action distance, and make actions easy.

[0033] Step S2: Incorporate the reasonable parts of user operations as the behavior layer into the Object Process Methodology (OPM). Utilize the concepts in the OPM to couple functional, behavioral, and structural information, constructing a model with user operations as the process and users and product structures as objects. Following the principle of process transformation and object state association, this model realizes the association between user operations and product structures. Then, combined with the FBS model representing the relationship between product functions, product behaviors, and product structures, an FBS-OPM product structure association model containing user operations is constructed.

[0034] Step S3 (User Perspective): Obtain online comment data, perform user comment analysis, use the TF-IDF algorithm to calculate the importance of each structure from the user's perspective, use the importance of the structure as the user's attention, and build a user attention table D for all structures.

[0035] Step S4 (Designer's Perspective): Calculate the product structure PR value based on the FBS-OPM product structure relationship model and the PageRank algorithm to obtain the product structure importance of each structure.

[0036] Step S5 (Decision Maker's Perspective): Review the company's Bill of Materials (BOM) and collect all structural change time and economic cost data. After normalization and weighting, obtain the change costs for each structure of the target product. The change costs for all structures of the product constitute the change cost table C.

[0037] Step S6: Standardize the user attention, product structure importance, and change costs of each structure, apply the entropy weight method to calculate the index weights, and obtain the three-way composite importance of each product structure.

[0038] Step S7: Calculate the conditional change propagation probability, and combine the conditional change propagation probability with the three-party composite importance to obtain the change propagation intensity CPI of adjacent structures.

[0039] Step S8: Set the target change amount T for the design change source and the change structure absorption amount F for all structures of the product.

[0040] Step S9: Based on the design change source determined in Step S1, set the initial change structure number as the starting point of the ant colony algorithm, initialize the parameters, and use the ant colony algorithm to solve for the optimal structure change propagation path to guide product iterative design optimization.

[0041] The probability of conditional change propagation refers to the likelihood that a change instruction, once initiated, will propagate from one structure to neighboring structures. Its value is related to the degree of correlation between the behavioral processes of the structures; the higher the correlation, the easier it is for propagation to occur.

[0042] When a propagation event occurs, it means that the structure of the upstream node has changed, and the structure of the downstream node has also changed. Therefore, it can be considered as the probability of event B occurring given that event A has occurred, and the probability of propagation due to conditional change. P ij It can be represented as: , Representation structure S i Structure S under the condition of change i and adjacent structure S j The probability of simultaneous changes.

[0043] When structure S i To adjacent structure S j When a change propagates, it can be assumed that both changes occur simultaneously, and the joint change probability of the two can be calculated: , in, Representation structure S i , S j The probability of simultaneous changes is called the probability of joint changes. P ( S i ), P ( S j ) respectively represent structure S i Adjacent structures S j The probability of a change occurring needs to be obtained from the historical change database. P ( S i ∩ S j The larger the value, the easier it is for product changes to spread through this path.

[0044] Structure S i Select adjacent structure S j transition probability Through structure S i Weights associated with adjacent structures (nodes) B ij and structure S i The sum of the association weights with all adjacent downstream structures The ratio is determined, specifically: , in, k For adjacent downstream structure indexes, m This represents the total number of adjacent downstream structures. B ijThe determination is based on the process-based structural correlation strength evaluation criteria, as shown in Table 2: Table 2. Process-based structural correlation strength evaluation criteria

[0045] It is worth noting that there may be more than one process connecting the two structures, and it is necessary to sum the association weights of all processes between the two structures. Assuming the structures... S i With adjacent structures S j There are n The processes are connected, and the association weights of these processes are respectively used as... B ij1 , B ij2 , B ij3 ,... , B ijn To represent, then the structure S i With adjacent structures S j Association weight between B ij It can be calculated using the following formula: , in, This represents the association weight value of the r-th process.

[0046] Finally, the propagation intensity of changes in adjacent structures is calculated based on the propagation probability of conditional changes. P ij ≠0, then , in, , and These represent the weights of user attention, product structure importance, and change cost for each structure, respectively, determined using the entropy weight method. ; Indicates the importance of the three-party composite; D j , PR j and C j These represent the user attention, product structure importance, and change cost of the j-th structure, respectively. Representing structure S i With adjacent structure S j The intensity of the change in transmission; like P ij= 0, then structure S i With adjacent structure Sj The change propagation intensity is 0.

[0047] Probability of conditional change propagation P ij The larger the value, the more likely the change is to occur along this route. To minimize the impact of changes, the conditional change propagation probability value needs to be as high as possible, and the three-way composite importance of adjacent structures needs to be as low as possible. P ij and D j , PR j , C j The opposite variable. This application aims to positiveize a negative variable by using (1- P ij ( ) to participate in the construction of the formula for calculating the intensity of change propagation.

[0048] Furthermore, the process of constructing the FBS-OPM product structure relationship model is as follows: Step S21: Perform functional analysis on the target product, obtain the functional layer, behavioral layer and structural layer, and construct the FBS model of the target product; Step S22: Construct a behavior-structure mapping for the reasonable parts of the target product user operation behavior in the dynamic element analysis table. Then, using the OPM modeling language, apply the concepts in OPM to couple the functional, behavioral, and structural information of the target product. Construct an information association model with the target product user operation behavior in the dynamic element analysis table as the process and the user and product structure as the objects at different functional layers. Then, combine it with FBS to construct a target product FBS-OPM product structure association model that includes user operation behavior.

[0049] The three-party stakeholder perspectives in this invention include user attention calculation from the user's perspective, product structure importance calculation from the designer's perspective, and change cost calculation from the decision-maker's perspective.

[0050] Example 2 This embodiment is based on a user-driven interactive mechanical product design change method, including the following steps: Step S1: Use dynamic element analysis to study user operation behavior and product behavior under different functional scenarios, record user operation behavior when using the product, and use the unreasonable parts in the user operation behavior to deduce the corresponding parts and obtain the source of design change. Step S11: Determine the target product and the specific functions to be optimized; Step S12: Determine the target population for the survey and limit the scope of the research users; Step S13: Issue a command to the user to use a certain function and observe the user's operation process and behavior; Step S14: Record data and store user operation behavior in the form of dynamic elements in the dynamic element analysis table; Step S15: Analyze user operation behavior in the action element analysis table, classify user operation behavior according to the principle of action economy, divide user operation behavior into reasonable and unreasonable parts, and map the unreasonable parts of user operation behavior to the product structure to determine the source of design change; the principle of action economy includes the following four points: reduce the number of actions, work with both hands at the same time, shorten the action distance, and make actions easy.

[0051] Step S2: Combining the Function-Behavior-Structure (FBS) model and the Object-Process Methodology (OPM), reasonable parts of user operations are added to the OPM as the behavior layer. Using the dual-channel modeling languages ​​OPL and OPD, correlation analysis is performed on each static structure from the perspectives of attributes and constraints. Dynamic correlations at each level are analyzed through energy flow, information flow, and material flow information to construct the FBS-OPM product structure correlation model. OPL and OPD are the OPM modeling tools.

[0052] Step S3: Conduct a three-party composite importance assessment of the product based on the perspectives of three stakeholders (users, designers, and decision-makers), and introduce the probability of conditional change propagation to calculate the intensity of change propagation.

[0053] The calculation of the tripartite composite importance comprises three modules: first, user attention calculation based on the TF-IDF algorithm from the user's perspective; second, product structure importance calculation based on the PageRank algorithm and the FBS-OPM product structure relationship model from the designer's perspective; and third, change cost calculation from the decision-maker's perspective. After acquiring data from these three modules, the entropy weight method is used to calculate the tripartite composite importance for subsequent design guidance.

[0054] Step S4: Use the ant colony algorithm to find the optimal propagation path for structural changes.

[0055] Step S41: Set the change structure absorption capacity to represent the ability of a certain structure in the product design to absorb changes and the impact of the changes.

[0056] Step S42: Set the target change amount T to represent the size of the change task.

[0057] Step S43: With the goal of minimizing the sum of change propagation intensities of adjacent structures along the change propagation path, and taking the design change source as the starting position of the ant colony algorithm, the ant colony algorithm is used to solve the change propagation path to obtain the optimal structural change propagation path and guide the product iterative design.

[0058] Example 3 This invention addresses the lack of user-related factors in the design change propagation process of existing interactive mechanical products. It comprehensively incorporates user information from three aspects: obtaining change requests, establishing a product structure correlation model, and calculating the importance of the three parties involved, thus optimizing the change propagation path analysis process. In this embodiment, a mountain bike is used as the research object. First, the dynamic element analysis method is used to analyze user operational behavior under different functional scenarios, constructing a dynamic element analysis table. Unreasonable parts of user operational behavior are mapped to the product structure, thereby determining the starting point of the design change, i.e., the location of the design change source. Subsequently, the FBS model is used to analyze the internal structural relationships of the product. Combined with the reasonable parts of user operational behavior from the dynamic element analysis results, and using OPM, an FBS-OPM product structure correlation model is constructed, enabling unified modeling and correlation expression of user operational behavior and product structure information.

[0059] Next, user online review data analysis was conducted: keywords were extracted from the text of online review data using the TF-IDF algorithm, and weights were calculated to obtain the importance of each structure from the user's perspective. The importance of the structure was used as the user attention value, and a user attention table was established. Simultaneously, based on the FBS-OPM product structure correlation model, the PageRank algorithm was used to calculate the product structure PR value of each structure. This product structure PR value was used as the product structure importance from the designer's perspective, providing basic data for subsequent comprehensive structural evaluation. Furthermore, the company's Bill of Materials (BOM) was reviewed, and all structural change time and economic cost data were collected. Extreme value normalization was performed on these two data sets. In this embodiment, with the guidance of company experts, a weight ratio of 1:1 was established, and the change cost of a single structure was calculated by weighted summation. The change costs of all structures were compiled into a change cost table, thus obtaining the change costs of each structure of the target product from the decision-maker's perspective.

[0060] Then, the entropy weight method is used to assign weights to three types of indicators: user attention, product structure importance, and change cost, and the three-way composite importance is calculated.

[0061] After obtaining the three-way composite importance, further analysis is conducted from the perspective of change propagation: First, the conditional change propagation probability between each structure is calculated, followed by the change propagation intensity (CPI), which characterizes the transmission capability of the change's impact. Second, the target change amount and the change absorption amount of each structure are set. After setting the initial change structure number and initializing relevant parameters, the ant colony algorithm is used for path optimization, ultimately obtaining the optimal structural change propagation path for the mountain bike product. The optimal structural change propagation path is the path with the minimum sum of change propagation intensities of adjacent structures along the change path.

[0062] The method of this invention effectively realizes human-machine collaboration and information fusion in the design change process by constructing an FBS-OPM product structure relationship model that includes user operation behavior and combining the three-party composite importance. It can significantly improve the efficiency and accuracy of design changes for interactive mechanical products, optimize the change propagation path, shorten the product iteration cycle, thereby improving the user experience and expanding the application scope of change propagation research.

[0063] More specifically, Step S11: Determine the target product as a mountain bike and identify the specific functions to be optimized.

[0064] Step S12: Determine the target population for the survey and limit the scope of research users.

[0065] Step S13: Issue a command to the user to use a certain function and observe the user's operation process and behavior.

[0066] Step S14: Record data, store user operation behavior in the form of dynamic elements, and obtain the user operation behavior dynamic element analysis table.

[0067] Step S15: Organize and analyze motion data to identify design change sources. This process follows the principles of motion economy, including the following four points: reducing the number of movements, using both hands simultaneously, shortening movement distances, and simplifying movements.

[0068] In this embodiment, based on the principle of motion economy, the unreasonable user operation behavior in the product design is mainly concentrated in the upshifting and downshifting operations. The product design places the gear shift lever on the right hand, causing all upshifting and downshifting operations to be concentrated on the right hand, resulting in a significant time gap for the left hand and wasted action, thus leading to a lack of coordination between the left and right hand operations. To balance the workload of both hands, it was decided to use the gear shift lever as the source of design changes for improvement. The goal is to reduce the number of gear shifting operations and balance the workload of both hands, thereby improving the product's functional efficiency.

[0069] Step S21: Perform functional analysis on the mountain bike to obtain the functional layer, behavioral layer, and structural layer, and construct the target product FBS model.

[0070] Step S22: Analyze the reasonable parts of the mountain bike user operation behavior in the dynamic element analysis table in the form of behavior-structure mapping to obtain the product structure corresponding to the user operation behavior. Combined with the FBS model obtained in step S21, construct the FBS-OPM product structure correlation model of mountain bikes that includes user operation behavior.

[0071] Based on the identified design change source as the gearshift lever, a gear-related functional analysis was conducted on the mountain bike. The Functional Breakdown Structure (FBS) model was used to map these functions to the behavioral and structural layers, establishing their interrelationships. The analysis revealed that this mountain bike product comprises eight functions: gear shifting, passenger carrying, riding, steering, warning, braking, seat adjustment, and shock absorption. These functions are further decomposed into 57 product behaviors, involving 77 parts. Figure 2 The relevant functional FBS display includes upshifting, downshifting, and steering. The behavior of product testers using mountain bikes was observed, and their operational behaviors using different functions in all scenarios were analyzed and statistically summarized. The effective dynamic element analysis content was compiled to obtain the structure corresponding to user operations. This structure, along with the behavioral and structural layers obtained from the FBS model mapping, was then incorporated into the OPM to form a mountain bike FBS-OPM product structure correlation model incorporating user operations (see [link]). Figure 3 ).

[0072] Table 3 is a summary of the names and part numbers of mountain bike parts.

[0073] Step S31: Crawl online review data from mountain bike sales networks, and analyze the importance of the structure from the user's perspective using the TF-IDF algorithm to obtain the user attention to each structure of the mountain bike.

[0074] Step S32: Based on the FBS-OPM product structure relationship model for mountain bikes, and combined with the PageRank algorithm, calculate the product structure importance of each structure in the product structure relationship model.

[0075] Step S33: View the company's Bill of Materials (BOM) and collect all structural change time cost and economic cost data. Perform extreme value normalization on the two data, and calculate the change cost of a single structure by weighted summation, thereby obtaining the change cost of each structure of the target product.

[0076] Step S34: Standardize user attention, product structure importance, and change cost, and apply the entropy weight method to calculate the weights of the three indicators, user attention, product structure importance, and change cost, to obtain the three-way composite importance. Step S35: Calculate the propagation probability of condition change, and calculate the propagation intensity of change based on the perspective of the three stakeholders by combining the three-party composite importance obtained in step S34.

[0077] Calculating user attention from the user's perspective: First, online review data was crawled from the company's sales channels. It was known that the product's online sales page contained 5139 user reviews. Using Octoparse, positive and negative review sets were constructed, and 4456 valid reviews were selected after initial screening. Next, text preprocessing was performed. The Jieba word segmentation library was used to break long sentences into word combinations. A stop word list was loaded and cross-referenced with the source data to remove words with positive comparison results from the source data, thus eliminating invalid words that would affect subsequent data mining and analysis. Then, the processed text underwent data processing: word frequency statistical analysis was performed, and the importance of each term was calculated based on the TF-IDF algorithm, ultimately yielding the user attention value. TF-IDF (Term Frequency-Inverse Document Frequency) is used to evaluate the importance of a word in a document or corpus, using importance as the measure of user attention.

[0078] Calculating the importance of product structures from a designer's perspective: First, based on the FBS-OPM product structure correlation model for mountain bikes, the correlations between product structures are obtained. Then, the correlation strength of each structure is evaluated using the process-based structural correlation strength evaluation criteria to obtain the correlation weights between each structure. B ij This forms a correlation weight matrix, and the transition probability of two adjacent structures is calculated based on this matrix. p ij The sum of the transition probabilities for each row is 1. A damping factor is introduced to avoid the occurrence of loop nodes. β , β The value is set to 0.85, and the convergence threshold is set to 1e-8. After setting, the PageRank value calculation begins: , in, , These represent the product structure PR values ​​of the i-th and j-th structures, respectively; N is the number of structures for the target product. Product structure PR value is used as the measure of product structure importance.

[0079] The user attention, product structure importance, and change cost are standardized, and the entropy weight method is applied to calculate the weights of the indicators. The resulting weights for user attention, product structure importance, and change cost are as follows: =0.31, =0.30, =0.39. The weighted sum of user attention, product structure importance, and change cost yields a composite importance score. This score, combined with the conditional change propagation probability, ultimately determines the change propagation strength of adjacent structures. The optimal structural change propagation path is the path that minimizes the sum of the change propagation strengths of adjacent structures along that path. The objective function of the ant colony algorithm is the total length of the traversed paths.

[0080] Step S41: Establish the change structure absorption capacity to represent the change absorption capacity of a certain structure in the mountain bike design, and to absorb the impact of the change.

[0081] Step S42: Input the source node for the change and set the target change amount T to represent the size of the change task.

[0082] The target change amount and the amount absorbed by the changed structure can be determined using a change database and engineer's corrective analysis.

[0083] Step S43: Use the ant colony algorithm to solve the change propagation path, obtain the optimal structural change propagation path for the mountain bike, and guide the design.

[0084] Using the gearshift lever as the change source, an optimal structural change propagation path search was performed. First, an experienced design and manufacturing engineer set the initial change amount to 145 and iterated based on the structural relationships. Data shows that the algorithm converged after the fifth iteration, with a final change propagation strength of 1.247182. This solution is the optimal solution for the mountain bike change propagation path analysis starting from structure 49, with the output path being 49-24-23-22-26-25-27.

[0085] According to the optimal structural change propagation path, when the design change originates from the shifter, it affects seven parts, with the change task stopping at the front brake cable. After discussing the proposed change with the designer, it aligns with their analysis. Changing the shifter's position affects the right handlebar grip; to maintain consistency between the left and right grips, the left grip also needs modification. Changing the grips modifies the handlebar structure, which in turn affects the user's braking action, necessitating a change in the brake lever position. To ensure braking performance, changing the brake lever position means the brake cable length also needs adjustment.

[0086] The final schematic diagram of the mountain bike structural modification scheme guided by the change propagation path is as follows: Figure 4 As shown in the diagram, the design principle of the mountain bike modification is illustrated below. Figure 5As shown. The specific principle of this modification is as follows: the shifting function is adjusted, the shift lever structure is removed, and the shifting function is integrated into the left and right handlebar grips. During user operation, the right hand rotates the right handlebar grip clockwise to achieve the original upshifting function, and the left hand rotates the left handlebar grip clockwise to achieve the downshifting function. To implement this functional change, the handlebar grips, handlebars, front brake lever, and front brake lever base need to be adjusted accordingly.

[0087] In this solution, the user's method of shifting gears differs significantly from the original product. The modified gear shifting operation eliminates the need for finger movement, thus avoiding action delays caused by gear shifting. Furthermore, the gear shifting function is split, rationally distributing the original operation between the left and right hands, allowing each hand to handle upshifting and downshifting separately. This reduces the workload on the right hand and effectively prevents gear shifting errors caused by right-hand misoperation.

[0088] This invention provides a novel solution for design changes in interactive mechanical products by introducing user operation behavior analysis and constructing an FBS-OPM product structure correlation model. First, compared to traditional methods that rely on subjective analysis for design requirements acquisition, this invention objectively identifies unreasonable behaviors during user operations through dynamic element analysis, thereby accurately locating the source of design changes. This method not only improves the accuracy and traceability of requirements mining but also significantly reduces the subjectivity and uncertainty in the requirements mining process, ensuring the scientific and practical nature of design changes. Second, by constructing the FBS-OPM product structure correlation model, this invention achieves a deep integration of user operation behavior and product structure. Traditional FBS models typically focus only on the product behavior layer, neglecting the complex relationship between user operation behavior and product design. This invention innovatively combines the OPM method, using user operation behavior as the process and users and product structure as objects, following the principle of process-transformation-object-state correlation construction, introducing user operation behavior into product structure analysis, and achieving multi-dimensional modeling of design information. This innovation not only enhances the accuracy of product design decisions but also provides a more objective and reliable basis for subsequent change path selection. Finally, it integrates the perspectives of users, designers, and decision-makers to optimize the design change path, comprehensively considering the concerns and pain points of different stakeholders, enhancing the applicability of the design change plan and user experience, making the design iteration of interactive mechanical products more in line with market demands, and improving the product's competitiveness.

[0089] In practical applications, taking mountain bikes as an example, by analyzing user behavior, structural relationships, and change costs, the optimal change path can be accurately selected across multiple dimensions, thereby improving product usability and user experience. This method can significantly shorten product iteration cycles and accelerate the rapid updating and optimization of mechanical products, demonstrating significant application value and promotional significance.

[0090] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

[0091] Any aspects not covered in this invention are applicable to existing technologies.

Claims

1. A user-driven interactive mechanical product design change method, characterized in that, The design change method includes the following steps: By using dynamic element analysis, we study user operation behavior and product behavior under different functional scenarios, record user operation behavior when using the product, and deduce corresponding parts from unreasonable parts of user operation behavior to obtain the source of design change. The reasonable parts of user operation behavior are added to the Object Process Methodology (OPM) as the behavior layer, and then combined with the FBS model to obtain the FBS-OPM product structure relationship model that includes user operation behavior. The product structure PR value is calculated based on the FBS-OPM product structure association model and the PageRank algorithm to obtain the product structure importance of each structure. We acquire online comment data, use the TF-IDF algorithm to calculate the importance of each structure from the user's perspective, and use the importance of the structure as the user's attention level. By examining the company's Bill of Materials (BOM), collecting all time and economic cost data for structural changes, and then normalizing and weighting the data, the change costs for each structure of the product can be obtained. Based on user attention to each structure, product structure importance, and change costs, the entropy weight method is applied to calculate the three-way composite importance of each product structure. Calculate the propagation probability of conditional change, and use the propagation probability of conditional change to combine the three-party composite importance to calculate the propagation intensity of change between adjacent structures; With the goal of minimizing the sum of change propagation intensities of adjacent structures along the change propagation path, and taking the design change source as the starting position of the intelligent optimization algorithm, the intelligent optimization algorithm is used to obtain the optimal structural change propagation path to guide product iterative design.

2. The design change method according to claim 1, characterized in that, The process for calculating the importance of the product structure is as follows: The process-based structural correlation strength evaluation criteria are used to assess the correlation strength of each structure in a product and obtain the correlation weights between each structure. B ij And based on this, calculate the transition probability of two adjacent structures; The transition probability is substituted into the PageRank algorithm to calculate the product structure PR value, and the product structure PR value is used as the importance of the product structure.

3. The design change method according to claim 2, characterized in that, The process for handling the propagation intensity of changes in adjacent structures is as follows: Combined with the aforementioned transition probability calculation structure S i To adjacent structure S j The joint change probability during change propagation is used to calculate the conditional change propagation probability based on the joint change probability. If the propagation probability of conditional change is 0, then structure S i With adjacent structure S j The propagation strength of the change is 0; if the propagation probability of the conditional change is not 0, then the structure S i With adjacent structure S j The change propagation strength is: the negative of the conditional change propagation probability plus 1, then combined with the adjacent structure S. j The importance of the three components is multiplied.

4. The design change method according to claim 1, characterized in that, The intelligent optimization algorithm is at least one of ant colony optimization, genetic algorithm, or multi-objective particle swarm optimization algorithm.

5. The design change method according to claim 1, characterized in that, The process of constructing the FBS-OPM product structure relationship model is as follows: Step S21: Perform functional analysis on the target product, obtain the functional layer, behavioral layer and structural layer, and construct the FBS model of the target product; Step S22: Construct a behavior-structure mapping for the reasonable parts of the target product user operation behavior in the dynamic element analysis table. Then, using the OPM modeling language, apply the concepts in OPM to couple the functional, behavioral, and structural information of the target product. Construct an information association model with the target product user operation behavior in the dynamic element analysis table as the process and the user and product structure as the object. Establish information association models at different functional layers. Then, combine it with the FBS model to construct a target product FBS-OPM product structure association model that includes user operation behavior.

6. The design change method according to claim 5, characterized in that, The target product is a mountain bike.

7. The design change method according to claim 6, characterized in that, In motion analysis, following the principle of motion economy, user actions are divided into reasonable and unreasonable parts, and the unreasonable parts of user actions are mapped to the product structure to identify the source of design changes. The principle of motion economy includes the following four points: reducing the number of actions, working with both hands simultaneously, shortening the distance of actions, and making actions easy.

Images