Ship human factor index evaluation method and system based on index weight dynamic allocation

Abstract

The application provides a ship human factor index evaluation method and system based on index weight dynamic distribution, the ship human factor index evaluation method based on index weight dynamic distribution comprises the following steps: selecting a corresponding target evaluation example index system from an extensible index library according to a target evaluation task; an index weight vector of the target evaluation example index system is adjusted by user operation to obtain a dynamic weight vector; an adaptive comprehensive evaluation model is obtained according to the target evaluation example index system; after receiving index layer data, a comprehensive evaluation result is obtained according to the dynamic weight vector and the comprehensive evaluation model. After the weight of the evaluation example index system is adjusted, the index layer data is comprehensively evaluated by the comprehensive evaluation model, which helps to solve the technical problem that there is no ship human factor evaluation index evaluation method for different research stages and different task scenes in the prior art.

Description

Technical fields: This invention relates to the field of human factors engineering technology, and in particular to a method and system for evaluating human factors indicators of ships based on dynamic allocation of indicator weights. Background technology: As a typical complex human-machine system operating under long-term isolation and high-load conditions, the level of human factors engineering (human-machine-environment fit) on ships directly affects operational safety, efficiency, and personnel health. Scientifically constructing an evaluation index system is fundamental to human factors engineering assessment. However, existing technologies have the following prominent shortcomings: The indicator system is static and fixed: existing evaluation indicator systems are mostly constructed one-time based on specific stages or typical scenarios, and once the indicators and weights are determined, they are difficult to adjust. They cannot adapt to the changes in focus at different stages of the ship's entire life cycle (design, development, testing, and use), nor can they respond to the differentiated requirements of indicator weights for different mission scenarios (cruise, surveillance, combat, and damage control).

[0001] Rigid weighting methods: Often employing a single, static weighting method (such as fixed expert scoring or simple AHP), lacking a dynamic adjustment mechanism that combines objective data with subjective experience. Weights cannot be iteratively optimized as evaluation data accumulates, the task context changes, or new evaluation cases are introduced, leading to a disconnect between evaluation results and actual effectiveness.

[0002] Insufficient engineering implementation and continuity: Static systems are difficult to update indicators and adjust weights in engineering practice, resulting in the evaluation system being disconnected from the actual engineering development process, and failing to form a continuous evaluation data chain that is traceable and comparable throughout the entire life cycle of the equipment.

[0003] There is an urgent need for a method and system for evaluating ship human factors indicators based on dynamic allocation of indicator weights. This would help solve the technical problem of the lack of a method for evaluating ship human factors indicators in different development stages and mission scenarios in the existing technology. Summary of the Invention: In one embodiment, the present invention provides a method for evaluating ship human factors indicators based on dynamic allocation of indicator weights. By adjusting the weights of the indicator system in the evaluation embodiment, a comprehensive evaluation model is performed on the indicator layer data, which helps to solve the technical problem of the lack of a method for evaluating ship human factors indicators under different development stages and mission scenarios in the prior art.

[0004] The target evaluation instance index system is selected from the scalable index library according to the target evaluation task. The scalable index library has at least one evaluation instance index system that matches different development stages and task scenarios, as well as index weight vectors. The target evaluation instance index system is the evaluation instance index system that matches the target evaluation task. The indicator weight vector of the target evaluation embodiment indicator system is dynamically adjusted by user operation to obtain the indicator weight vector. A comprehensive evaluation model is obtained based on the indicator system of the aforementioned target evaluation implementation example; After receiving the indicator layer data, the comprehensive evaluation result is obtained based on the dynamic weight vector and the comprehensive evaluation model.

[0005] In one embodiment, after receiving the index layer data and obtaining the comprehensive evaluation result based on the dynamic weight vector and the comprehensive evaluation model, the method further includes: The comprehensive evaluation results, the dynamic weight vector, and the comprehensive evaluation model from this assessment will be saved as case studies.

[0006] In one embodiment, the indicator weight vector of the target evaluation embodiment indicator system is dynamically adjusted by user operations to obtain the weight vector. After multiple users independently adjust the weights, the divergence value between the weight vector of each user and the average weight vector of all users is calculated. If all divergence values ​​are less than a preset threshold, the average weight is adopted as the final result; otherwise, the divergence values ​​are fed back to the users so that they can make the next round of adjustments until all divergence values ​​are less than the preset threshold.

[0007] In one embodiment, if the target evaluation embodiment index system includes both quantifiable indicators and qualitative language evaluation indicators, then the fuzzy comprehensive evaluation model is preferred; if the evaluation focuses on performance shortcomings, then the geometric comprehensive evaluation model is used as an auxiliary reference.

[0008] In one embodiment, before the step of selecting the corresponding target evaluation embodiment indicator system from the scalable indicator library according to the target evaluation task, the method further includes: Based on a basic human factors index library, evaluation example index systems are generated by screening and combining indicators according to specific ship types, development stages, and mission scenarios.

[0009] In one embodiment, the basic human factors index library has a basic human factors index set, which has a three-layer hierarchical structure, namely, the target layer, the criterion layer, and the index layer. The target layer includes safety, efficiency, economy, and user-friendliness; The criteria layer includes the relationship between people and hardware, the relationship between people and software, the relationship between people and the environment, and the relationship between people.

[0010] In one embodiment, the human-hardware relationship includes an operational error rate and an alarm response. The efficiency includes task completion time and human workload; The evaluation embodiment index system includes data information on safety, operational error rate, alarm response time, efficiency, task completion time, and workload.

[0011] In one embodiment, the present invention also provides a ship human factors index evaluation system based on dynamic allocation of index weights, the system comprising: An extensible indicator library management module has an extensible indicator library. The extensible indicator library management module is used to store and manage the basic human factors indicator set and supports the generation and retrieval of evaluation instance indicator systems. A multi-source weight dynamic allocation engine is used to dynamically allocate indicator weights to the evaluation instance indicator system. An adaptive evaluation model module is used to adapt the corresponding comprehensive evaluation model from a variety of preset comprehensive evaluation models based on the structural characteristics and data types of the evaluation instance indicator system. An evaluation execution and visualization output module is used to calculate the comprehensive evaluation result from the indicator layer data based on the dynamic weight vector and the comprehensive evaluation model. The evaluation method performed by the system is as follows: The target evaluation instance index system is selected from the scalable index library according to the target evaluation task. The scalable index library has at least one evaluation instance index system that matches different development stages and task scenarios, as well as index weight vectors. The target evaluation instance index system is the evaluation instance index system that matches the target evaluation task. The indicator weight vector of the target evaluation embodiment indicator system is dynamically adjusted by user operation to obtain the indicator weight vector. A comprehensive evaluation model is obtained based on the indicator system of the aforementioned target evaluation implementation example; After receiving the indicator layer data, the comprehensive evaluation result is obtained based on the dynamic weight vector and the comprehensive evaluation model.

[0012] In one embodiment, the multi-source weight dynamic allocation engine includes: A scenario-weight mapping unit is used to store the initial indicator weight vector based on typical development stages and mission scenarios; An interactive weight adjustment and convergence unit is used to provide an interface for loading and allowing users to adjust the initial weight vector, and to perform convergence judgment and control based on the degree of divergence between the user weight vector and the average vector. A case study unit is used to use the comprehensive evaluation results, the dynamic weight vector, and the comprehensive evaluation model as cases to optimize the scenario-weight mapping unit.

[0013] In one embodiment, the evaluation execution and visualization output module can generate views that reflect the current weight distribution of indicators, a bar chart reflecting the scores of each indicator, and a line chart reflecting the trend of the overall score of the same equipment at different stages. Attached image description: Figure 1 This is a schematic diagram of the overall architecture of the system in one embodiment of the present invention; Figure 2 This is a schematic diagram illustrating the generation relationship between the scalable indicator library and the evaluation instance indicator system in another embodiment of the present invention; Figure 3 This is a schematic diagram of the workflow of the multi-source weight dynamic allocation engine in another embodiment of the present invention; Figure 4 This is a schematic diagram of the model recommendation logic of the evaluation model adaptive adaptation module in another embodiment of the present invention; Figure 5 This is a radar chart and a schematic diagram of the overall score generated by the evaluation execution and visualization output module in another embodiment of the present invention; Figure 6 This is a schematic diagram of cross-stage evaluation trend tracking in a data storage and traceability platform according to another embodiment of the present invention. Detailed implementation method: To enable those skilled in the art to better understand the technical solution of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0014] Various embodiments and features of this application are described herein with reference to the accompanying drawings.

[0015] These and other features of this application will become apparent from the following description of preferred forms of embodiments given as non-limiting examples, with reference to the accompanying drawings.

[0016] It should also be understood that although this application has been described with reference to some specific examples, those skilled in the art can certainly implement many other equivalent forms of this application, which have the features described in the claims and are therefore all within the scope of protection defined herein.

[0017] The above and other aspects, features and advantages of this application will become more apparent when taken in conjunction with the accompanying drawings and in view of the following detailed description.

[0018] Specific embodiments of this application are described below with reference to the accompanying drawings; however, it should be understood that the claimed embodiments are merely examples of this application, which can be implemented in various ways. Well-known and / or repeated functions and structures are not described in detail to ascertain the true intent based on the user's historical operations, and to avoid unnecessary or redundant details that would obscure this application. Therefore, the specific structural and functional details claimed herein are not intended to be limiting, but merely serve as the basis and representative basis for the claims to teach those skilled in the art to use this application in various ways with substantially any suitable detailed structure.

[0019] This specification may use the phrases “in one embodiment,” “in another embodiment,” “in yet another embodiment,” or “in other embodiments,” all of which may refer to one or more of the same or different embodiments according to this application.

[0020] This invention aims to overcome the shortcomings of existing technologies and provide a method and system for constructing a ship human factors assessment index system based on dynamic allocation of index weights and adaptive multi-assessment model. This system enables scenario-based and dynamic configuration of index weights and supports intelligent matching and fusion of assessment models, thereby improving the adaptability, versatility, and continuity of the index system to complex and ever-changing engineering realities.

[0021] Figure 1 This is a schematic diagram of the overall architecture of the system in one embodiment of the present invention; Figure 2 This is a schematic diagram illustrating the generation relationship between the scalable indicator library and the evaluation instance indicator system in another embodiment of the present invention; Figure 3 This is a schematic diagram of the workflow of the multi-source weight dynamic allocation engine in another embodiment of the present invention; Figure 4 This is a schematic diagram of the model recommendation logic of the evaluation model adaptive adaptation module in another embodiment of the present invention; Figure 5 This is a radar chart and a schematic diagram of the overall score generated by the evaluation execution and visualization output module in another embodiment of the present invention; Figure 6 This is a schematic diagram of cross-stage evaluation trend tracking in a data storage and traceability platform according to another embodiment of the present invention. Figures 1 to 6 As shown, in one embodiment, the present invention provides a method for evaluating ship human factors indicators based on dynamic allocation of indicator weights, the method comprising: The target evaluation instance index system is selected from the scalable index library according to the target evaluation task. The scalable index library has at least one evaluation instance index system that matches different development stages and task scenarios, as well as index weight vectors. The target evaluation instance index system is the evaluation instance index system that matches the target evaluation task. The indicator weight vector of the target evaluation embodiment indicator system is dynamically adjusted by user operation to obtain the indicator weight vector. A comprehensive evaluation model is obtained based on the indicator system of the aforementioned target evaluation implementation example; After receiving the indicator layer data, the comprehensive evaluation result is obtained based on the dynamic weight vector and the comprehensive evaluation model.

[0022] In one embodiment, after receiving the index layer data and obtaining the comprehensive evaluation result based on the dynamic weight vector and the comprehensive evaluation model, the method further includes: The comprehensive evaluation results, the dynamic weight vector, and the comprehensive evaluation model from this assessment will be saved as case studies.

[0023] In one embodiment, the indicator weight vector of the target evaluation embodiment indicator system is dynamically adjusted by user operations to obtain the weight vector. After multiple users independently adjust the weights, the divergence value between the weight vector of each user and the average weight vector of all users is calculated. If all divergence values ​​are less than a preset threshold, the average weight is adopted as the final result; otherwise, the divergence values ​​are fed back to the users so that they can make the next round of adjustments until all divergence values ​​are less than the preset threshold.

[0024] In one embodiment, if the target evaluation embodiment index system includes both quantifiable indicators and qualitative language evaluation indicators, then the fuzzy comprehensive evaluation model is preferred; if the evaluation focuses on performance shortcomings, then the geometric comprehensive evaluation model is used as an auxiliary reference.

[0025] In one embodiment, before the step of selecting the corresponding target evaluation embodiment indicator system from the scalable indicator library according to the target evaluation task, the method further includes: Based on a basic human factors index library, evaluation example index systems are generated by screening and combining indicators according to specific ship types, development stages, and mission scenarios.

[0026] In one embodiment, the basic human factors index library has a basic human factors index set, which has a three-layer hierarchical structure, namely, the target layer, the criterion layer, and the index layer. The target layer includes safety, efficiency, economy, and user-friendliness; The criteria layer includes the relationship between people and hardware, the relationship between people and software, the relationship between people and the environment, and the relationship between people.

[0027] In one embodiment, the human-hardware relationship includes an operational error rate and an alarm response. The efficiency includes task completion time and human workload; The evaluation embodiment index system includes data information on safety, operational error rate, alarm response time, efficiency, task completion time, and workload.

[0028] In one embodiment, the present invention also provides a ship human factors index evaluation system based on dynamic allocation of index weights, the system comprising: An extensible indicator library management module has an extensible indicator library. The extensible indicator library management module is used to store and manage the basic human factors indicator set and supports the generation and retrieval of evaluation instance indicator systems. A multi-source weight dynamic allocation engine is used to dynamically allocate indicator weights to the evaluation instance indicator system. An adaptive evaluation model module is used to adapt the corresponding comprehensive evaluation model from a variety of preset comprehensive evaluation models based on the structural characteristics and data types of the evaluation instance indicator system. An evaluation execution and visualization output module is used to calculate the comprehensive evaluation result from the indicator layer data based on the dynamic weight vector and the comprehensive evaluation model. The evaluation method performed by the system is as follows: The target evaluation instance index system is selected from the scalable index library according to the target evaluation task. The scalable index library has at least one evaluation instance index system that matches different development stages and task scenarios, as well as index weight vectors. The target evaluation instance index system is the evaluation instance index system that matches the target evaluation task. The indicator weight vector of the target evaluation embodiment indicator system is dynamically adjusted by user operation to obtain the indicator weight vector. A comprehensive evaluation model is obtained based on the indicator system of the aforementioned target evaluation implementation example; After receiving the indicator layer data, the comprehensive evaluation result is obtained based on the dynamic weight vector and the comprehensive evaluation model.

[0029] In one embodiment, the multi-source weight dynamic allocation engine includes: A scenario-weight mapping unit is used to store the initial indicator weight vector based on typical development stages and mission scenarios; An interactive weight adjustment and convergence unit is used to provide an interface for loading and allowing users to adjust the initial weight vector, and to perform convergence judgment and control based on the degree of divergence between the user weight vector and the average vector. A case study unit is used to use the comprehensive evaluation results, the dynamic weight vector, and the comprehensive evaluation model as cases to optimize the scenario-weight mapping unit.

[0030] In one embodiment, the evaluation execution and visualization output module can generate views that reflect the current weight distribution of indicators, a bar chart reflecting the scores of each indicator, and a line chart reflecting the trend of the overall score of the same equipment at different stages.

[0031] Based on the above system, an extensible indicator library management module can be used to store and manage basic human factors indicator sets, and support the generation of assessment instance indicator systems according to specific assessment objects and stages. A multi-source weight dynamic allocation engine is used to dynamically allocate indicator weights to the evaluation instance indicator system; the engine includes: The scenario-weight mapping unit stores the initial indicator weight vector based on typical development stages and typical task scenarios; An interactive weight adjustment and convergence unit provides an interface for loading and allows experts to adjust the initial weights, and performs convergence judgment and control based on the degree of divergence between the expert weight vector and the average vector. A case learning unit is used to store the converged final weights and their evaluation context as new cases to optimize the mapping unit; The evaluation model adaptive adaptation module is used to recommend an appropriate evaluation model from a variety of preset comprehensive evaluation models based on the structural characteristics and data types of the evaluation instance indicator system. The evaluation execution and visualization output module is used to perform comprehensive calculations and visualization of the input indicator data based on the assigned weights and the adapted evaluation model. The data storage and traceability platform is used to store evaluation data throughout the entire process, supporting traceability and comparative analysis across time and tasks.

[0032] In the interactive weight adjustment and convergence unit, the degree of divergence is quantified by calculating the gray similarity correlation between each expert's weight vector and the average weight vector, and a divergence threshold ξ is preset; when the degree of divergence is ≥ ξ, a feedback mechanism is triggered to require experts to readjust their weights; when the degree of divergence is < ξ, the weights are determined to converge.

[0033] The multi-source weight dynamic allocation engine also integrates subjective weighting algorithms, objective weighting algorithms, and a combination of subjective and objective weighting algorithms. It also supports calling the objective weighting algorithm to generate objective weights based on the indicator sample data of this evaluation during the interactive adjustment process, which can then be used as a reference for expert adjustments.

[0034] The comprehensive evaluation model preset in the adaptive adaptation module of the evaluation model includes at least the following: linear weighted comprehensive method, geometric comprehensive method, ideal point method (TOPSIS), fuzzy comprehensive evaluation method and radar chart feature value evaluation method; the recommendation is based on the rule engine, and the rules at least consider the independence between indicators, data type (precise value / interval value / linguistic value), and evaluation preference (whether to focus on the weakest link effect).

[0035] The adaptive evaluation model module supports multi-model fusion evaluation, which means that for the same evaluation data, multiple recommendation models are used in parallel for calculation, and the results of each model are weighted and fused to output the final comprehensive evaluation value.

[0036] The views generated by the visualization output module include at least: an indicator weight radar chart reflecting the current weight distribution, a bar chart reflecting the scores of each indicator, and a line chart reflecting the trend of the overall score of the same equipment at different stages.

[0037] The basic human factors indicator set in the extensible indicator library management module is organized in a three-tiered hierarchical structure of "target layer - criterion layer - indicator layer". The target layer includes safety, efficiency, economy, and human-friendliness; the criterion layer covers at least the dimensions of human-hardware relationship, human-software relationship, human-environment relationship, and human-human relationship.

[0038] The method is as follows: A. For the target assessment task, select and customize indicators from an extensible indicator library to generate an assessment example indicator system; B. Based on the development stage and mission scenario associated with the assessment task, the initial weights are obtained from the scenario-weight mapping unit, and the final dynamic weight vector is obtained through the interactive adjustment and convergence unit. C. The evaluation model adaptive adaptation module recommends a suitable comprehensive evaluation model based on the evaluation instance index system. D. Input or collect indicator layer data, use the dynamic weight vector and the adapted evaluation model to calculate, obtain the comprehensive evaluation result and output it visually; E. Save the complete context, weight vector, model, and results of this evaluation as a case study for use in optimizing subsequent weight recommendations in the system.

[0039] The interactive adjustment and convergence process described in step B specifically includes: inviting multiple experts to independently adjust the weights, calculating the degree of divergence between each expert's weight vector and the average weight vector of all experts; if all divergences are less than a preset threshold, then the average weight is adopted as the final result; otherwise, the divergence information is fed back to the experts, and a new round of adjustment is initiated until convergence.

[0040] In step C, if the evaluation instance index system includes both quantifiable indicators and qualitative language evaluation indicators, the fuzzy comprehensive evaluation model is preferred; if the evaluation focuses particularly on performance shortcomings, the geometric comprehensive evaluation model is also recommended as an auxiliary reference.

[0041] To achieve the above objectives, this invention proposes a system for constructing a ship human factors assessment index system based on dynamic allocation of index weights, comprising: The scalable indicator library management module stores and manages a multi-layered, multi-dimensional set of basic indicators for ship human factors engineering. This indicator set covers at least the dimensions of "human-hardware," "human-software," "human-environment," and "human-human" relationships, and is organized in a hierarchical structure of "target layer - criterion layer - indicator layer." It allows users to select, combine, and customize extended indicators from the basic indicator library based on specific evaluation objects (such as specific ship models or systems) and evaluation stages (such as scheme design, land-based commissioning, and sea trials), generating specific evaluation instance indicator systems.

[0042] Multi-source dynamic weight allocation engine: This engine integrates multiple weight allocation algorithms, including subjective weighting methods (such as the improved Analytic Hierarchy Process (AHP) and the Delphi method), objective weighting methods (such as entropy weighting and principal component analysis), and a combination of subjective and objective weighting methods. Its core features are: Scenario-Weight Mapping Library: A pre-established mapping relationship library between "typical development stages - typical task scenarios" and indicator weight vectors. This library is built based on historical evaluation data, expert knowledge, and task analysis.

[0043] Interactive dynamic adjustment mechanism: During specific evaluations, the engine first loads initial weights from the mapping library based on the "stage-scenario" associated with the current evaluation instance. Subsequently, through an interactive weight adjustment interface, evaluation experts are allowed to fine-tune the weights based on the specificities of the current task. During the adjustment process, the system calculates the divergence between the current expert weight vector and the group average weight vector in real time (e.g., based on gray similarity correlation). If the divergence exceeds a set threshold, a feedback mechanism is triggered, prompting experts to re-examine the weights or initiate a new round of expert consultation until the weights converge.

[0044] Case learning and weight iteration: After the evaluation is completed, the system stores the final weight vector adopted in this evaluation and the corresponding evaluation context (stage, scenario, object) as a new case in the mapping library, which is used to optimize the initial weight recommendation for similar scenarios in the future, so as to realize the continuous learning and dynamic evolution of weights.

[0045] The evaluation model adaptive adaptation module incorporates multiple comprehensive evaluation model algorithms, including at least linear weighted synthesis, geometric synthesis, ideal point method (TOPSIS), fuzzy comprehensive evaluation, and radar chart area-perimeter feature value evaluation. Based on the structural characteristics of the current evaluation instance's indicator system (e.g., the independence between indicators), the data type of indicator values ​​(exact values, interval values, verbal comments), and evaluation preferences (whether inter-indicator compensation is allowed, whether to focus on the weakest link effect), the module recommends one or more of the most suitable evaluation models through a rule engine or lightweight machine learning model. It also supports weighted fusion of results from multiple models to improve the robustness and interpretability of the evaluation.

[0046] The evaluation execution and visualization output module is used to import or collect raw data (objective measurement data or subjective scores) from the indicator layer in real time. Based on the weights output by the dynamic allocation engine and the evaluation model selected by the adaptation module, it calculates the scores of each level of indicators and the overall system evaluation value. This module provides rich visualization outputs, including but not limited to: weight distribution radar charts, indicator score bar charts, comprehensive evaluation trend charts (for cross-stage tracking), and multi-scheme comparison views based on radar charts.

[0047] Data storage and traceability platform: Used to persistently store the indicator system, weight version, raw data, evaluation model, calculation process, and results of all evaluation instances. Supports multi-dimensional data retrieval and comparative analysis based on timeline, equipment model, and mission scenario, ensuring the complete traceability of the evaluation history and providing a data foundation for the continuous evaluation of equipment human factors performance.

[0048] This invention also provides a method for constructing and evaluating an indicator system based on the above system, comprising the following steps: S1: Based on the objectives of this assessment (such as assessing the human factors effectiveness of the XX ship's command system during the design phase), select and customize an assessment example indicator system from the scalable indicator library.

[0049] S2: The multi-source weight dynamic allocation engine loads initial weights from the mapping library based on the "development stage" and "mission scenario" associated with the evaluation instance, and corrects them through an interactive adjustment mechanism combined with expert opinions, outputting the final determined dynamic weight vector.

[0050] S3: The evaluation model adaptive adaptation module analyzes the current indicator system and data characteristics, and recommends and determines the applicable comprehensive evaluation model.

[0051] S4: Collect or input indicator layer data, and the evaluation execution module uses the determined weights and models to perform calculations and generate comprehensive evaluation results.

[0052] S5: Store all elements of this evaluation (indicator set, weights, model, data, results) into the system as new cases to complete knowledge iteration.

[0053] The beneficial effects of this invention are: Compared with the prior art, the present invention has the following significant advantages: Strong dynamic adaptability: By combining "scenario-weight mapping" with "interactive adjustment", the indicator weights are dynamically and accurately configured according to the evaluation stage, task scenario, and expert knowledge, overcoming the fundamental defect of poor adaptability of static weights.

[0054] It has good engineering practicality: It provides the flexibility to quickly build an instance system from the basic library, as well as an intuitive weight adjustment interface, which greatly reduces the technical threshold and cost of adjusting and iterating the indicator system in engineering practice, and promotes the implementation of evaluation work.

[0055] High level of intelligence: The introduction of an adaptive matching mechanism for evaluation models avoids blind selection of models and improves the scientific nature and efficiency of evaluation; the system has self-optimization capabilities through case learning.

[0056] Highly versatile and sustainable: The constructed indicator library and weight mapping library can cover multiple types of ship equipment and the entire life cycle, supporting horizontal comparison of evaluation results between different systems, as well as vertical performance tracking of the same system across stages, forming a highly sustainable evaluation data asset.

[0057] Scientific consensus-building: By using the quantification of disagreement to control the weight adjustment process, expert opinions are effectively converged on the basis of full expression, thereby improving the scientific nature and acceptance of the weight allocation results.

[0058] The specific working principle and steps are as follows: Example 1: System Architecture and Workflow like Figure 1 As shown, the system adopts a layered modular design. Users define evaluation instances through the configuration layer (corresponding to module 1). The core calculation layer (corresponding to modules 2 and 3) completes dynamic weight allocation and model adaptation. The execution and display layer (corresponding to module 4) performs calculations and visualization. All process data is accumulated in the data layer (corresponding to module 5), forming a closed loop.

[0059] Example 2: Dynamic Weight Allocation Take the assessment of "the human factors interface of damage control during the sea trial phase of a new type of destroyer" as an example.

[0060] Step S2-1: The system identifies the scenarios of "trial phase" and "damage control mission" and retrieves the preset initial weight vector W0 from the mapping library (which may emphasize the weight of indicators such as "emergency operation accessibility" and "information display intuitiveness").

[0061] Step S2-2: Three experts (user, designer, and human factors expert) view W0 through the interactive interface and make fine adjustments based on specific issues observed during this test run (such as unclear valve markings) to generate individual weight vectors W1, W2, and W3.

[0062] Step S2-3: The system calculates the grey similarity correlation degree divergence d between W1, W2, W3 and the average vector W_avg. If d = 0.15, which is less than the preset threshold ξ = 0.2, the system determines that the opinions have converged and adopts W_avg as the final weight W_final. If d > 0.2, the system will prompt "Expert opinions are highly divergent" and display the indicator with the greatest divergence, guiding experts to conduct a second round of discussion and adjustment until d meets the requirements.

[0063] Step S2-4: Store (sea trial phase, damage control mission, a certain type of destroyer, W_final) as a new case in the mapping library.

[0064] Example 3: Evaluation Model Adaptation Instance (Continued from the previous example)

[0065] The evaluation example indicator system includes both "average operation time" (precise numerical value) and "error rate" (precise numerical value), as well as "interface layout satisfaction" (verbal rating: excellent, good, average, poor). There is a certain correlation between the indicators (such as operation time and error rate).

[0066] Step S3: Adaptive module analysis revealed that: ① the data contains both precise and linguistic values; ② the indicators are not completely independent; ③ the assessment aims to take into account both the overall level and the shortcomings.

[0067] Model Recommendation: Fuzzy comprehensive evaluation is recommended for processing language evaluations, and quantitative indicators should also be converted into fuzzy membership degrees. Simultaneously, considering the need to address the weakest link effect, geometric comprehensive evaluation can be used in parallel for auxiliary calculations. The final comprehensive evaluation result can be a weighted fusion output based primarily on the fuzzy evaluation result, with the geometric method result used as a reference.

[0068] Example 4: System Verification Method To verify the effectiveness and superiority of this system, the following methods can be used: Dynamic adaptability verification: Evaluation cases of the same equipment at different development stages (such as design review and model testing) are selected, and the differences in evaluation results using the dynamic weight of this invention and using fixed weight are compared to analyze whether the dynamic weight more sensitively reflects the stage characteristics and problem improvement.

[0069] Engineering convenience verification: Invite engineers from different backgrounds to use this system to build an evaluation system for the same new scenario, record their operation time and number of adjustments, and compare it with traditional methods (such as reorganizing experts to hold a meeting from scratch to score).

[0070] Model fit validation: For the same evaluation dataset, the system recommendation model and the random selection model are evaluated respectively, and the consistency between the results and the expert qualitative evaluation is compared to verify the effectiveness of the recommendation model.

[0071] like Figure 2 As shown, the evaluation example index system is generated from the basic human factors index library based on different development stages, different mission scenarios, and different objects (specific ship models). The basic human factors index library includes human factors elements such as safety, efficiency, economy, and human comfort. Safety includes human-hardware safety, human-software safety, human-environment safety, and the next level of human-human coordination safety. Human-hardware safety includes operational error rate and alarm response time, while efficiency includes task completion time and workload. Finally, the evaluation example indicator system forms data structures for safety and operational error rate, safety and alarm response time, efficiency and task completion time, and efficiency and workload. This has led to the formation of specific structures for human factors data in different ship signals, development stages, and mission scenarios, realizing the reshaping of the data structure from a basic human factors indicator library to an evaluation example indicator system.

[0072] The above embodiments are merely exemplary embodiments of the present invention and are not intended to limit the present invention. The scope of protection of the present invention is defined by the claims. Those skilled in the art can make various modifications or equivalent substitutions to the present invention within its spirit and scope of protection, and such modifications or equivalent substitutions should also be considered to fall within the scope of protection of the present invention.

Claims

1. A method for evaluating ship human factors indicators based on dynamic allocation of indicator weights, characterized in that, The ship human factors index evaluation method based on dynamic allocation of index weights includes: The target evaluation instance indicator system is selected from the scalable indicator library according to the target evaluation task. The scalable indicator library has at least one evaluation instance indicator system that matches different development stages and task scenarios, as well as indicator weight vectors. The target evaluation instance indicator system is the evaluation instance indicator system that matches the target evaluation task. The indicator weight vector of the target evaluation embodiment indicator system is dynamically adjusted by user operation to obtain the indicator weight vector. A comprehensive evaluation model is obtained based on the indicator system of the aforementioned target evaluation implementation example; After receiving the indicator layer data, the comprehensive evaluation result is obtained based on the dynamic weight vector and the comprehensive evaluation model.

2. The method for evaluating ship human factors indicators based on dynamic allocation of indicator weights according to claim 1, characterized in that, After receiving the indicator layer data and obtaining the comprehensive evaluation result based on the dynamic weight vector and the comprehensive evaluation model, the method further includes: The comprehensive evaluation results, the dynamic weight vector, and the comprehensive evaluation model from this assessment will be saved as case studies.

3. The method for evaluating ship human factors indicators based on dynamic allocation of indicator weights according to claim 2, characterized in that, The indicator weight vector of the target evaluation embodiment indicator system is dynamically adjusted by user operations. After multiple users independently adjust the weights, the divergence value between each user's weight vector and the average weight vector of all users is calculated. If all divergence values ​​are less than a preset threshold, the average weight is adopted as the final result; otherwise, the divergence values ​​are fed back to the users so that they can make the next round of adjustments until all divergence values ​​are less than the preset threshold.

4. The method for evaluating ship human factors indicators based on dynamic allocation of indicator weights according to claim 3, characterized in that, If the target evaluation implementation indicator system includes both quantifiable indicators and qualitative language evaluation indicators, then the fuzzy comprehensive evaluation model shall be used first; if the evaluation focuses on performance shortcomings, then the geometric comprehensive evaluation model shall be used as an auxiliary reference.

5. The method for evaluating ship human factors indicators based on dynamic allocation of indicator weights according to claim 4, characterized in that, Before the step of selecting the corresponding target evaluation embodiment indicator system from the expandable indicator library according to the target evaluation task, the method further includes: Based on a basic human factors index library, evaluation example index systems are generated by screening and combining indicators according to specific ship types, development stages, and mission scenarios.

6. The method for evaluating ship human factors indicators based on dynamic allocation of indicator weights according to claim 5, characterized in that, The basic human factors indicator library has a basic human factors indicator set, which has a three-layer hierarchical structure, namely the target layer, the criterion layer, and the indicator layer. The target layer includes safety, efficiency, economy, and user-friendliness; The criteria layer includes the relationship between people and hardware, the relationship between people and software, the relationship between people and the environment, and the relationship between people.

7. The method for evaluating ship human factors indicators based on dynamic allocation of indicator weights according to claim 6, characterized in that, When the relationship between the human and hardware has an operational error rate and alarm response; The efficiency includes task completion time and human workload; The evaluation embodiment index system includes data information on safety, operational error rate, alarm response time, efficiency, task completion time, and workload.

8. A ship human factors index evaluation system based on dynamic allocation of index weights, characterized in that, The system includes: An extensible indicator library management module has an extensible indicator library. The extensible indicator library management module is used to store and manage the basic human factors indicator set and supports the generation and retrieval of evaluation instance indicator systems. A multi-source weight dynamic allocation engine is used to dynamically allocate indicator weights to the evaluation instance indicator system. An adaptive evaluation model module is used to adapt the corresponding comprehensive evaluation model from a variety of preset comprehensive evaluation models based on the structural characteristics and data types of the evaluation instance indicator system. An evaluation execution and visualization output module is used to calculate the comprehensive evaluation result from the indicator layer data based on the dynamic weight vector and the comprehensive evaluation model. The evaluation method performed by the system is as follows: The target evaluation instance indicator system is selected from the scalable indicator library according to the target evaluation task. The scalable indicator library has at least one evaluation instance indicator system that matches different development stages and task scenarios, as well as indicator weight vectors. The target evaluation instance indicator system is the evaluation instance indicator system that matches the target evaluation task. The indicator weight vector of the target evaluation embodiment indicator system is dynamically adjusted by user operation to obtain the indicator weight vector. A comprehensive evaluation model is obtained based on the indicator system of the aforementioned target evaluation implementation example; After receiving the indicator layer data, the comprehensive evaluation result is obtained based on the dynamic weight vector and the comprehensive evaluation model.

9. The ship human factors index evaluation system based on dynamic allocation of index weights according to claim 8, characterized in that, The multi-source weight dynamic allocation engine includes: A scenario-weight mapping unit is used to store the initial indicator weight vector based on typical development stages and mission scenarios; An interactive weight adjustment and convergence unit is used to provide an interface for loading and allowing users to adjust the initial weight vector, and to perform convergence judgment and control based on the degree of divergence between the user's weight vector and the average vector. A case study unit is used to use the comprehensive evaluation results, the dynamic weight vector, and the comprehensive evaluation model as cases to optimize the scenario-weight mapping unit.

10. The ship human factors index evaluation system based on dynamic allocation of index weights according to claim 9, characterized in that, The evaluation execution and visualization output module can generate views that reflect the current weight distribution of indicators, a bar chart reflecting the scores of each indicator, and a line chart reflecting the trend of the overall score of the same equipment at different stages.

Images