Packaging dynamic evaluation system and evaluation method based on entropy weight TOPSIS method

Abstract

The application discloses a packaging dynamic evaluation system and method based on an entropy weight TOPSIS method, and the packaging dynamic evaluation system based on the entropy weight TOPSIS method comprises a data sensing and collecting module, an entropy weight dynamic weighting module, a TOPSIS comprehensive evaluation module, a cross-sensory mapping analysis module and a design parameter reconstruction module which are connected in sequence. The application discloses and verifies a strong coupling relationship between a tactile texture and emotional safety, the Pearson correlation coefficient is not less than 0.95, a nonlinear mapping function is established, a scientific basis is provided for packaging design, and a change from experience driving to data driving of the packaging design is promoted.

Description

Technical Field

[0001] This invention relates to the field of packaging evaluation and design optimization technology, and in particular to a packaging dynamic evaluation system and evaluation method based on the entropy weight TOPSIS method. Background Technology

[0002] As the consumer market continues to mature and upgrade, product packaging plays an increasingly prominent role in marketing communication and user experience. Consumers' evaluation of packaging has expanded from simple functional needs to a comprehensive value judgment that includes brand image, ease of use, and multi-sensory experience. As the first interface between a product and the consumer, packaging not only bears the basic functions of protecting goods and conveying information, but also serves as an important carrier for establishing brand awareness and emotional connection.

[0003] Current research and practice on packaging design evaluation still have significant limitations. A common phenomenon in design practice is an overemphasis on visual presentation and a neglect of interactive experience. Designers often devote most of their energy to color matching, graphic creativity, and layout design, while relatively ignoring design considerations such as material texture, grip comfort, and ease of opening. Research shows that this design tendency results in some products, while possessing strong appeal on shelf displays, failing to provide consumers with a consistently high-quality experience during actual contact and use, and even reducing consumer trust in product quality due to poor tactile feedback.

[0004] Furthermore, in the field of packaging design evaluation, although various evaluation models have been established, a quantitative evaluation system for multi-sensory experiences is still incomplete. Existing evaluation research mainly focuses on the following two dimensions: one is based on Life Cycle Assessment (LCA) theory, emphasizing the environmental impact and economic cost analysis of packaging materials; the other is qualitative analysis based on aesthetics and psychology, focusing on the exploration of visual form. In terms of quantitative evaluation methods, the Analytic Hierarchy Process (AHP) and its improved models are widely used. This method has advantages in handling multi-level indicators, but the determination of its indicator weights mainly relies on expert scoring, which is easily influenced by the evaluator's subjective preferences, knowledge structure, and psychological state, thus limiting the objectivity and universality of the evaluation results.

[0005] Patent document CN115310637A discloses a product packaging design evaluation method based on user experience. This method collects users' subjective evaluations of the packaging through questionnaires and uses the analytic hierarchy process (AHP) to determine the weights of the indicators. However, this method still suffers from problems such as strong subjectivity, limited sample size, and difficulty in capturing subtle perceptual differences. Furthermore, existing technologies fail to effectively reveal the intrinsic mechanism by which consumers build trust in products through tactile feedback, resulting in a lack of scientific basis for packaging design. Therefore, there is an urgent need for an evaluation system and method that can objectively quantify multi-sensory experiences, particularly revealing the tactile-safety perception mapping mechanism, to provide scientific decision support for packaging design. Summary of the Invention

[0006] To address the aforementioned technical problems, this invention provides a packaging dynamic evaluation system based on the entropy weight TOPSIS method with a simple structure, and also provides a packaging dynamic evaluation method based on the entropy weight TOPSIS method.

[0007] The technical solution of the present invention to solve the above-mentioned technical problems is: a packaging dynamic evaluation system based on the entropy weight TOPSIS method, comprising a data sensing and acquisition module, an entropy weight dynamic weighting module, a TOPSIS comprehensive evaluation module, a cross-sensory mapping analysis module, and a design parameter reconstruction module connected in sequence;

[0008] The data perception and acquisition module collects multimodal perception data of users on product packaging and builds a user perception database;

[0009] The entropy weight dynamic weighting module calculates the information entropy value of each perception dimension based on the collected multimodal perception data, dynamically calculates the weight coefficient of each evaluation indicator, and eliminates subjective weighting bias.

[0010] The TOPSIS comprehensive evaluation module constructs a space of positive and negative ideal solutions, calculates the Euclidean distance between the packaging scheme and the ideal solution, and generates a comprehensive closeness score.

[0011] The cross-sensory mapping analysis module calculates the correlation matrix between all evaluation indicators and verifies the coupling strength of the preset core coupling indicator pairs.

[0012] The design parameter refactoring module refactors parameters when the packaging design does not meet the requirements.

[0013] The aforementioned packaging dynamic evaluation system based on the entropy weight TOPSIS method includes a data perception and acquisition module comprising a visual perception acquisition unit, a tactile perception acquisition unit, a functional perception acquisition unit, an emotional perception acquisition unit, and an environmental standardization control unit.

[0014] The visual perception acquisition unit collects visual evaluation data of users on the color, pattern and layout of the packaging. The visual evaluation data includes one or more of the following: questionnaire scores, image recording data and eye-tracking data.

[0015] The tactile sensing acquisition unit collects tactile evaluation data when the user holds the packaging. The tactile evaluation data includes one or more of the following: material tactile score, pressure distribution data, surface roughness data, friction coefficient data, and temperature change data.

[0016] The functional perception and acquisition unit collects functional experience data when the user opens the packaging. The functional experience data includes one or more of the following: opening convenience score, opening time, opening force, and number of action steps.

[0017] The emotion perception acquisition unit collects users' emotional evaluation data on packaging safety and environmental protection. The emotional evaluation data includes one or more of questionnaire scores and physiological signal data.

[0018] The environmental standardization control unit constructs a test environment with constant temperature, humidity, and illuminance to ensure the consistency and comparability of data acquisition.

[0019] The aforementioned dynamic evaluation system for packaging based on the entropy weight TOPSIS method, in its entropy weight dynamic weighting module, considers evaluation indicators from the dimensions of visual representation, tactile interaction, functional experience, and emotional mapping. The visual representation dimension includes visual attractiveness. and class recognition Two evaluation metrics: the tactile interaction dimension includes tactile texture. and grip comfort Two evaluation metrics: the functional experience dimension includes ease of activation. One evaluation metric, the emotion mapping dimension, includes emotional security. and environmental awareness Two evaluation indicators; the first The tactile texture of each packaging solution The calculation method is as follows:

[0020] ;

[0021] in, For the number of tests, For the first Packaging scheme #1 The surface friction coefficient obtained from this test and then normalized. For the first Packaging scheme #1 The normalized temperature comfort level after this test For the first Packaging scheme #1 The normalized perceived hardness value from this test. This is the weighting coefficient for the surface friction coefficient. This is a weighting factor for temperature comfort. This is the weighting coefficient for the perceived hardness value, and ;

[0022] Apart from tactile texture, other evaluation indicators are calculated according to a unified framework of "sub-indicator collection—directional consistency—normalization—intra-dimensional weighting," that is: Let the... The first packaging scheme The first evaluation indicator Sub-indicators are ,like As a benefit-oriented indicator, according to Normalize, For the normalized first The first packaging scheme The first evaluation indicator Sub-indicators, For the first The first evaluation indicator The minimum value of each sub-indicator For the first The first evaluation indicator The maximum value of each sub-indicator; if As a cost-type indicator, according to Perform normalization; if If the evaluation score is already on a 1-10 scale, then take it directly. ;

[0023] No. Packaging scheme #1 Output value of each evaluation indicator for:

[0024] ;

[0025] in, For the first The number of sub-indicators under each evaluation indicator For the first Within the evaluation indicators, the first Weights within the dimensions of each sub-indicator.

[0026] In the aforementioned dynamic evaluation system for packaging based on the entropy-weighted TOPSIS method, the formula for calculating the information entropy value in the dynamic entropy weighting module is as follows:

[0027] ;

[0028] in, For the first The information entropy value of each evaluation indicator For the total number of packaging options, It is the natural logarithm function. , for Zhan Di The proportion of the standardized values ​​of all packaging schemes to the total value of each evaluation indicator. For the first The packaging scheme is in the first The standardized values ​​of each evaluation indicator For the index of packaging schemes, For the index of evaluation indicators; when hour, Process as 0.

[0029] The aforementioned packaging dynamic evaluation system based on the entropy weight TOPSIS method includes a TOPSIS comprehensive evaluation module comprising a standardization processing unit, a weighted decision matrix construction unit, an ideal solution space construction unit, and a distance calculation unit.

[0030] The standardization unit uses the range transformation method to standardize the original data. The standardization formula is as follows:

[0031] ;

[0032] in, For the first The packaging scheme is in the first The standardized values ​​of each evaluation indicator For the index of packaging schemes, For the index of evaluation indicators, For the first The packaging scheme is in the first The original observed values ​​of each evaluation indicator and The first The maximum and minimum values ​​of each evaluation indicator;

[0033] The weighted decision matrix construction unit multiplies the standardized value by the entropy weight result to construct the weighted standardized decision matrix. Weighted standardized decision matrix The Middle Line 1 The elements of the column are , For the index of packaging schemes, , For the total number of packaging options, For the index of evaluation indicators, , The total number of evaluation indicators, , For the first The entropy weight coefficient of the first evaluation indicator; Coefficient of difference of each evaluation indicator , No. Entropy weight coefficient of each evaluation indicator The calculation formula is:

[0034] ;

[0035] in, The total number of evaluation indicators, ;

[0036] Ideal solution space building blocks are used to determine positive ideal solutions. and negative ideal solution , , The first in the positive ideal solution One element, , This represents taking the weighted standardized decision matrix. The Middle The maximum value of the column elements; The first in the positive ideal solution One element, , The first of the negative ideal solutions One element, , This represents taking the weighted standardized decision matrix. The Middle The minimum value of a column element. The first in the set of negative ideal solutions One element;

[0037] The distance calculation unit calculates the Euclidean distance from each packaging scheme to the positive ideal solution and the negative ideal solution. The calculation formula is as follows:

[0038] ;

[0039] ;

[0040] in, For the first The distance between each packaging solution and the ideal solution. For the first The distance from each packaging scheme to the negative ideal solution;

[0041] The formula for calculating overall similarity is:

[0042] ;

[0043] in, Indicates the first The overall fit of the packaging solutions.

[0044] The aforementioned packaging dynamic evaluation system based on the entropy weight TOPSIS method includes a cross-sensory mapping analysis module comprising a correlation analysis unit, a coupling strength verification unit, a cross-modal mapping model construction unit, and a perception threshold determination unit.

[0045] The correlation analysis unit calculates the Pearson correlation coefficient matrix between various evaluation indicators based on the collected multimodal sensing data. , ,, For the first The evaluation index and the first Pearson correlation coefficients among the evaluation indicators; correlation analysis units are... As the result of full-scale cross-sensory relationship recognition, it is used to output the correlation strength between various evaluation indicators and to determine the set of strongly coupled indicator pairs. ,gather The indicators in the middle are for meeting and ,in This is a strong coupling screening threshold;

[0046] The coupling strength verification unit extracts the correlation coefficients corresponding to preset core coupling index pairs from the Pearson correlation coefficient matrix and uses them as constraints in addition to the comprehensive evaluation for verification. The preset core coupling index pairs include tactile texture and emotional safety, and their corresponding correlation coefficients are , which represent the Pearson correlation coefficients between tactile texture and emotional safety. The coupling strength verification unit verifies whether the preset coupling threshold is reached. Other correlation coefficients are used as auxiliary diagnostic relationships to identify the linkage characteristics between visual, tactile, functional and emotional indicators. If it is less than the preset coupling threshold, an alarm is triggered and the corresponding packaging scheme is sent to the design parameter reconstruction module.

[0047] The cross-modal mapping model building unit establishes a nonlinear mapping function between the physical parameters of touch and the sense of security. The nonlinear mapping function is as follows:

[0048] ;

[0049] in, Rate your sense of security. For material thickness, It is the natural logarithm function. Surface hardness, This is a weighting factor for material thickness. This is a weighting coefficient for surface hardness. For the fitting parameters, , and The data was obtained by fitting the sample packaging material thickness, surface hardness, and safety perception score data; the target safety perception score was... Surface hardness When given, the material thickness adjustment unit adjusts according to... Calculate the target material thickness and limit it to the range of 0.15mm-0.25mm; if the target safety score is still not met after the thickness reaches the boundary of the range, adjust the surface hardness to Shore A hardness of 50-70.

[0050] The perception threshold determination unit determines the critical perception threshold of tactile texture. When the tactile texture score is lower than the critical perception threshold, the sense of emotional security decreases.

[0051] The aforementioned packaging dynamic evaluation system based on the entropy weight TOPSIS method includes a design parameter reconstruction module comprising a parameter adjustment target determination unit, a candidate parameter generation unit, a parameter constraint unit, and an effect verification unit.

[0052] The parameter adjustment target determination unit determines the type of packaging design parameters to be adjusted based on the evaluation index that the overall closeness does not reach the preset threshold and the core coupling index pair that does not reach the preset coupling threshold.

[0053] The candidate parameter generation unit generates at least one candidate adjustment parameter from among the following: packaging surface tactile parameters, packaging structural morphology parameters, material thickness parameters, and material hardness parameters, based on the type of packaging design parameter to be adjusted.

[0054] The parameter constraint unit filters and constrains candidate adjustment parameters based on packaging materials, packaging structure, processing technology, and safety requirements to generate an optimized packaging solution.

[0055] The effect verification unit retests the evaluation indicators of the optimized packaging scheme, and re-executes the entropy weight TOPSIS comprehensive evaluation and core coupling constraint verification to determine whether the comprehensive closeness of the optimized packaging scheme reaches the preset threshold and whether the correlation coefficient of the preset core coupling indicators reaches the preset coupling threshold.

[0056] The packaging dynamic evaluation method based on the entropy weight TOPSIS method includes the following steps:

[0057] Step 1, Multimodal Perception Data Acquisition: In a standardized environment, collect multimodal perception data from multiple target users on several packaging schemes;

[0058] Step 2, Dynamic Entropy Weighting: Based on multimodal sensing data, calculate the information entropy value of each sensing dimension and dynamically determine the weight coefficients of the evaluation indicators;

[0059] Step 3, Comprehensive Evaluation: Construct a space of positive and negative ideal solutions, calculate the comprehensive closeness of each packaging scheme, and rank them;

[0060] Step 4, cross-sensory mapping analysis: calculate the correlation matrix between all evaluation indicators and verify the coupling strength of the preset core coupling indicator pairs;

[0061] Step 5, Design Packaging Scheme Verification: Verify the new packaging design from multiple dimensions to determine whether it meets user experience requirements;

[0062] Step 6, Design Parameter Reconstruction: If the new packaging design does not meet the requirements, the parameter reconstruction module will be activated to generate an optimized packaging design.

[0063] In the above-mentioned dynamic evaluation method for packaging based on the entropy weight TOPSIS method, step five, the specific process of designing and validating the packaging scheme, is as follows:

[0064] First, the new packaging design scheme is input into the data perception and acquisition module to obtain perception data of the new packaging design scheme on 7 evaluation indicators;

[0065] Then, the overall fit of the new packaging design is calculated using the entropy weight dynamic weighting module and the TOPSIS comprehensive evaluation module.

[0066] Next, the overall fit of the new packaging design scheme will be compared with the preset threshold. For comparison, the overall similarity is used to characterize the packaging scheme's performance. to The overall performance across all evaluation metrics; simultaneously, using pre-set core coupling metrics as baseline constraints, to verify tactile texture. With emotional security Pearson correlation coefficient Does it meet the requirement of being greater than or equal to the preset coupling threshold? If the overall relevance of the new packaging design is less than the preset threshold... or tactile texture With emotional security Pearson correlation coefficient Less than the preset coupling threshold If the new packaging design does not meet user experience requirements, the parameters will be refactored.

[0067] Finally, if the requirements are still not met after three consecutive parameter reconstructions, an exception report will be generated and a redesign prompt report will be output.

[0068] The beneficial effects of this invention are as follows:

[0069] 1. This invention constructs a high-precision multimodal sensing data acquisition system with a sampling frequency of not less than 120Hz and a data accuracy of not less than 0.01N, realizing the objective quantification of user experience and breaking through the limitations of traditional design methods that emphasize vision over touch.

[0070] 2. This invention proposes a dynamic weight calculation method based on information entropy, which scientifically quantifies the core position of tactile texture and emotional security in consumption decisions. The sum of the weights of the two is not less than 0.35 of the total weight, thus avoiding the arbitrariness of subjective weighting by experts.

[0071] 3. This invention reveals and verifies the strong coupling relationship between tactile texture and emotional security, with a Pearson correlation coefficient of not less than 0.95. It establishes a nonlinear mapping function, providing a scientific basis for packaging design and promoting the transformation of packaging design from experience-driven to data-driven.

[0072] 4. This invention designs a closed-loop verification and parameter reconstruction mechanism. It uses multi-dimensional verification to determine whether the design scheme meets user experience requirements. If it does not, it performs parameter reconstruction, which improves the fit between the design scheme and the user's real needs and avoids the risk of market failure caused by design deviation. Attached Figure Description

[0073] Figure 1 This is a structural block diagram of the packaging dynamic evaluation system of the present invention.

[0074] Figure 2 This is a flowchart of the packaging dynamic evaluation method of the present invention.

[0075] Figure 3 This diagram illustrates the coupling relationship between tactile texture and emotional security. Detailed Implementation

[0076] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0077] like Figure 1 As shown, the packaging dynamic evaluation system based on the entropy weight TOPSIS method includes a data sensing and acquisition module, an entropy weight dynamic weighting module, a TOPSIS comprehensive evaluation module, a cross-sensory mapping analysis module, and a design parameter reconstruction module connected in sequence.

[0078] The data sensing and acquisition module collects multimodal perception data of users on product packaging and builds a user perception database; the sampling frequency of the data sensing and acquisition module is not less than 120Hz and the data accuracy is not less than 0.01N.

[0079] The entropy weight dynamic weighting module calculates the information entropy value of each perception dimension based on the collected multimodal perception data, dynamically calculates the weight coefficient of each evaluation indicator, and eliminates subjective weighting bias.

[0080] The TOPSIS comprehensive evaluation module constructs a space of positive and negative ideal solutions, calculates the Euclidean distance between the packaging scheme and the ideal solution, and generates a comprehensive closeness score.

[0081] The cross-sensory mapping analysis module calculates the correlation matrix between all evaluation indicators and verifies that the coupling strength of the preset core coupling indicator pairs is not less than 0.95.

[0082] The design parameter refactoring module refactors parameters when the packaging design does not meet the requirements.

[0083] The data perception and acquisition module includes a visual perception acquisition unit, a tactile perception acquisition unit, a functional perception acquisition unit, an emotion perception acquisition unit, and an environmental standardization control unit.

[0084] The visual perception acquisition unit collects visual evaluation data of users on the color, pattern, and layout of the packaging through a camera (24 million pixels) and an eye tracker (sampling rate 120Hz). The visual evaluation data includes one or more of the following: questionnaire scores, image recording data, and eye tracking data.

[0085] The tactile sensing acquisition unit collects tactile evaluation data when a user holds the packaging through a pressure sensor array (accuracy not less than 0.01N), a surface roughness detector, and a temperature sensor (accuracy ±0.1℃). The tactile evaluation data includes one or more of the following: material tactile score, pressure distribution data, surface roughness data, friction coefficient data, and temperature change data, with a sampling frequency of not less than 100Hz.

[0086] The functional perception acquisition unit collects functional experience data of the user when opening the package through an optical motion capture system. The functional experience data includes one or more of the following: opening convenience score, opening time, opening force, and number of action steps.

[0087] The emotion perception acquisition unit collects users' emotional evaluation data on packaging safety and environmental protection through questionnaires and physiological signal sensors. The emotional evaluation data includes one or more of questionnaire scores and physiological signal data.

[0088] The environmental standardization control unit constructs a test environment with constant temperature (25±0.5℃), constant humidity (50±2%), and constant illuminance (500±10 lux) to ensure the consistency and comparability of data collection.

[0089] The data sensing and acquisition module is equipped with a pressure sensor array and a standardized testing environment, enabling it to acquire multimodal sensory data from users in simulated real-world usage scenarios. This avoids the recall bias and inaccurate statements common in traditional questionnaires. The module also acquires material surface friction coefficient, temperature change, and hardness sensing data through the tactile sensing acquisition unit, providing a scientific basis for the quantification of tactile texture indicators. With a sampling frequency of no less than 120Hz and a data accuracy of no less than 0.01N, the module ensures the reliability and accuracy of data acquisition.

[0090] In the aforementioned entropy weight dynamic weighting module, the evaluation indicators are considered from the dimensions of visual representation, tactile interaction, functional experience, and emotional mapping. The visual representation dimension includes visual attractiveness. and class recognition Two evaluation metrics: the tactile interaction dimension includes tactile texture. and grip comfort Two evaluation metrics: the functional experience dimension includes ease of activation. One evaluation metric, the emotion mapping dimension, includes emotional security. and environmental awareness Two evaluation indicators; the first The tactile texture of each packaging solution The calculation method is as follows:

[0091] ;

[0092] in, For the number of tests, For the first Packaging scheme #1 The surface friction coefficient obtained from this test and then normalized. For the first Packaging scheme #1 The normalized temperature comfort level after this test For the first Packaging scheme #1 The normalized perceived hardness value from this test. This is the weighting coefficient for the surface friction coefficient. This is a weighting factor for temperature comfort. This is the weighting coefficient for the perceived hardness value, and ;

[0093] Apart from tactile texture, other evaluation indicators are calculated according to a unified framework of "sub-indicator collection—directional consistency—normalization—intra-dimensional weighting," that is: Let the... The first packaging scheme The first evaluation indicator Sub-indicators are ,like As a benefit-oriented indicator, according to Normalize, For the normalized first The first packaging scheme The first evaluation indicator Sub-indicators, For the first The first evaluation indicator The minimum value of each sub-indicator For the first The first evaluation indicator The maximum value of each sub-indicator; if As a cost-type indicator, according to Perform normalization; if If the evaluation score is already on a 1-10 scale, then take it directly. ;

[0094] No. Packaging scheme #1 Output value of each evaluation indicator for:

[0095] ;

[0096] in, For the first The number of sub-indicators under each evaluation indicator For the first Within the evaluation indicators, the first Weights within the dimensions of each sub-indicator.

[0097] The dynamic entropy weighting module uses information entropy theory to automatically determine indicator weights based on the dispersion of data, avoiding interference from human factors. In particular, the module innovatively ensures that the sum of the weight coefficients for tactile texture and emotional security is no less than 0.35 of the total weights, reflecting the core position of these two dimensions in consumer decision-making. The formula for calculating the information entropy value is:

[0098] ;

[0099] in, For the first The information entropy value of each evaluation indicator For the total number of packaging options, It is the natural logarithm function. , for Zhan Di The proportion of the standardized values ​​of all packaging schemes to the total value of each evaluation indicator. For the first The packaging scheme is in the first The standardized values ​​of each evaluation indicator For the index of packaging schemes, For the index of evaluation indicators; when hour, Process as 0.

[0100] The TOPSIS comprehensive evaluation module includes a standardization processing unit, a weighted decision matrix construction unit, an ideal solution space construction unit, and a distance calculation unit.

[0101] The standardization unit uses the range transformation method to standardize the original data. The standardization formula is as follows:

[0102] ;

[0103] in, For the first The packaging scheme is in the first The standardized values ​​of each evaluation indicator For the index of packaging schemes, For the index of evaluation indicators, For the first The packaging scheme is in the first The original observed values ​​of each evaluation indicator and The first The maximum and minimum values ​​of each evaluation indicator;

[0104] The weighted decision matrix construction unit multiplies the standardized value by the entropy weight result to construct the weighted standardized decision matrix. Weighted standardized decision matrix The Middle Line number The elements of the column are , For the index of packaging schemes, , For the total number of packaging options, For the index of evaluation indicators, , The total number of evaluation indicators, , For the first The entropy weight coefficient of the first evaluation indicator; Coefficient of difference of each evaluation indicator , No. Entropy weight coefficient of each evaluation indicator The calculation formula is:

[0105] ;

[0106] in, The total number of evaluation indicators, ;

[0107] Ideal solution space building blocks are used to determine positive ideal solutions. and negative ideal solution , , The first in the positive ideal solution One element, , This represents taking the weighted standardized decision matrix. The Middle The maximum value of the column elements; The first in the positive ideal solution One element, , The first of the negative ideal solutions One element, , This represents taking the weighted standardized decision matrix. The Middle The minimum value of a column element. The first in the set of negative ideal solutions One element;

[0108] The distance calculation unit calculates the Euclidean distance from each packaging scheme to the positive ideal solution and the negative ideal solution. The calculation formula is as follows:

[0109] ;

[0110] ;

[0111] in, For the first The distance between each packaging solution and the ideal solution. For the first The distance from each packaging scheme to the negative ideal solution;

[0112] The formula for calculating overall similarity is:

[0113] ;

[0114] in, This represents the overall similarity of the i-th packaging scheme. The value range is [0,1]. The higher the value, the closer the overall experience of the packaging solution is to the ideal state. Experimental results show that tactile texture... With emotional security Pearson correlation coefficient The value exceeded 0.95, verifying the strong coupling relationship between tactile texture and emotional security.

[0115] The TOPSIS comprehensive evaluation module can objectively quantify the differences in user experience among different packaging solutions, providing a scientific basis for design decisions.

[0116] The cross-sensory mapping analysis module includes a correlation analysis unit, a coupling strength verification unit, a cross-modal mapping model construction unit, and a perception threshold determination unit;

[0117] The correlation analysis unit calculates the Pearson correlation coefficient matrix between various evaluation indicators based on the collected multimodal sensing data. , ,, For the first The evaluation index and the first Pearson correlation coefficients among the evaluation indicators; correlation analysis units are... As the result of full-scale cross-sensory relationship recognition, it is used to output the correlation strength between various evaluation indicators and to determine the set of strongly coupled indicator pairs. ,gather The indicators in the middle are for meeting and ,in This is a strong coupling screening threshold;

[0118] The coupling strength verification unit extracts the correlation coefficients corresponding to preset core coupling index pairs from the Pearson correlation coefficient matrix and uses them as constraints in addition to the comprehensive evaluation for verification. The preset core coupling index pairs include tactile texture and emotional safety, and their corresponding correlation coefficients are , which represent the Pearson correlation coefficients between tactile texture and emotional safety. The coupling strength verification unit verifies whether the preset coupling threshold is reached. Other correlation coefficients are used as auxiliary diagnostic relationships to identify the linkage characteristics between visual, tactile, functional and emotional indicators. If it is less than the preset coupling threshold, an alarm is triggered and the corresponding packaging scheme is sent to the design parameter reconstruction module.

[0119] The cross-modal mapping model building unit establishes a nonlinear mapping function between the physical parameters of touch and the sense of security. The nonlinear mapping function is as follows:

[0120] ;

[0121] in, Rate your sense of security. For material thickness, It is the natural logarithm function. For surface hardness, This is a weighting factor for material thickness. This is a weighting coefficient for surface hardness. For fitting parameters, , and The data was obtained by fitting the sample packaging material thickness, surface hardness, and safety perception score data; the target safety perception score was... Surface hardness When given, the material thickness adjustment unit adjusts according to Calculate the target material thickness and limit it to the range of 0.15mm-0.25mm; if the target safety score is still not met after the thickness reaches the boundary of the range, adjust the surface hardness to Shore A hardness of 50-70.

[0122] The perception threshold determination unit determines the critical perception threshold of tactile texture. When the tactile texture score is lower than the critical perception threshold by 3.5 points, the emotional sense of security will decrease significantly, with a decrease of no less than 2.5 points.

[0123] The design parameter reconstruction module includes a parameter adjustment target determination unit, a candidate parameter generation unit, a parameter constraint unit, and an effect verification unit;

[0124] The parameter adjustment target determination unit determines the type of packaging design parameters to be adjusted based on the evaluation index that the overall closeness does not reach the preset threshold and the core coupling index pair that does not reach the preset coupling threshold.

[0125] The candidate parameter generation unit generates at least one candidate adjustment parameter from among the following: packaging surface tactile parameters, packaging structural morphology parameters, material thickness parameters, and material hardness parameters, based on the type of packaging design parameter to be adjusted.

[0126] The parameter constraint unit filters and constrains candidate adjustment parameters based on packaging materials, packaging structure, processing technology, and safety requirements to generate an optimized packaging solution.

[0127] The effect verification unit retests the evaluation indicators of the optimized packaging scheme, and re-executes the entropy weight TOPSIS comprehensive evaluation and core coupling constraint verification to determine whether the comprehensive closeness of the optimized packaging scheme reaches the preset threshold and whether the correlation coefficient of the preset core coupling indicators reaches the preset coupling threshold.

[0128] The design parameter refactoring module refactors design solutions that do not meet user experience requirements, generating optimized solutions and providing differentiated optimization strategies for different types of packaging. Through parameter refactoring, the system can significantly improve the user experience of packaging.

[0129] like Figure 2 As shown, the packaging dynamic evaluation method based on the entropy weight TOPSIS method includes the following steps:

[0130] Step 1, Multimodal Perception Data Acquisition: In a standardized environment, multimodal perception data of 180 target users on 28 packaging schemes are collected to build a user perception database.

[0131] Step 2, Dynamic Entropy Weighting: Based on multimodal sensing data, calculate the information entropy value of each sensing dimension and dynamically determine the weight coefficients of the 7 evaluation indicators.

[0132] Taking food packaging as an example, the tactile texture is calculated. The weight is 0.1945, emotional security. The weight of is 0.1732, and the sum of the two is 0.3677, which meets the requirement of not less than 0.35. Visual appeal The weight is 0.1258, environmental perception The weight is 0.1483, and the grip comfort is... The weight is 0.1165, enabling convenience. The weight is 0.1179, and the class recognition rate is... The weight is 0.1238.

[0133] Visual appeal and class recognition Two evaluation metrics: the tactile interaction dimension includes tactile texture. and grip comfort Two evaluation metrics: the functional experience dimension includes ease of activation. One evaluation metric, the emotion mapping dimension, includes emotional security. and environmental awareness .

[0134] Step 3, Comprehensive Evaluation: Construct a space of positive and negative ideal solutions, calculate the comprehensive closeness of each packaging scheme, and rank them.

[0135] Taking five typical food packaging types as examples: high-white glass bottles (overall similarity score of 0.8764), metal aluminum cans (overall similarity score of 0.7419), PET transparent bottles (overall similarity score of 0.5240), aseptic Tetra Pak cartons (overall similarity score of 0.4181), and composite film bags (overall similarity score of 0.1127). This ranking accurately reflects the differences in quality among different packaging types across multiple sensory experience dimensions.

[0136] Step 4, cross-sensory mapping analysis: calculate the correlation matrix between all evaluation indicators and verify the coupling strength of the preset core coupling indicator pairs.

[0137] like Figure 3 As shown, Figure 3This diagram illustrates the coupling relationship between tactile texture and emotional security. The horizontal axis represents tactile texture scores, and the vertical axis represents emotional security scores. Scatter points indicate tactile texture and emotional security scores for different packaging options. The fitting trend line E represents the changing relationship between tactile texture and emotional security. The calculated Pearson correlation coefficient between tactile texture and emotional security is 0.99, exceeding the preset threshold of 0.95, indicating a strong coupling relationship between the two. The vertical dashed line A represents the critical threshold line for tactile texture, corresponding to a tactile texture score of 3.5. When the tactile texture score is below 3.5, the emotional security score decreases significantly. The horizontal dashed line B represents the emotional security guarantee line, corresponding to an emotional security score of 6.0. When the emotional security score is below 6.0, the corresponding packaging option is deemed to have failed to meet the requirements for emotional security guarantee. C represents the low-tactile-texture packaging sample points, and D represents the high-tactile-texture packaging sample points, used to illustrate the differences in emotional security scores under different tactile-texture levels.

[0138] Step 5, Design Packaging Scheme Verification: Verify the new packaging design from multiple dimensions to determine whether it meets user experience requirements.

[0139] The specific process for validating the packaging design scheme is as follows:

[0140] First, the new packaging design scheme is input into the data perception and acquisition module to obtain perception data of the new packaging design scheme on 7 evaluation indicators;

[0141] Then, the overall fit of the new packaging design is calculated using the entropy weight dynamic weighting module and the TOPSIS comprehensive evaluation module.

[0142] Next, the overall relevance of the new packaging design will be compared with the preset threshold. For comparison, the overall similarity is used to characterize the packaging scheme's performance. to Overall performance across all evaluation indicators; preset thresholds Dynamically adjust based on product positioning: High-end products Mid-range products economical products ;

[0143] Using preset core coupling indicators as baseline constraints, the tactile texture was verified. With emotional security Pearson correlation coefficient Does it meet the requirement of being greater than or equal to 0.95? If the overall similarity of the new packaging design is less than the preset threshold... or tactile texture With emotional security Pearson correlation coefficient If the value is less than 0.95, the new packaging design is deemed not to meet user experience requirements, and the parameters need to be refactored.

[0144] For example, for an environmentally friendly flexible packaging, its overall proximity score is calculated to be 0.428, the preset threshold τ=0.5 (mid-range product), and the correlation coefficient between tactile texture and emotional security is 0.92. Since 0.92<0.95, it is determined that the design scheme does not meet the user experience requirements and the parameters need to be reconstructed.

[0145] Finally, if the requirements are still not met after three consecutive parameter reconstructions, an exception report will be generated, and a redesign prompt report will be output.

[0146] Step Six, Design Parameter Reconstruction: If the new packaging design does not meet the requirements, the parameter reconstruction module is activated to generate an optimized packaging solution. Optimization is performed on this environmentally friendly flexible packaging: 1) Surface micro-nano texture reconstruction: adding a matte frosted coating with a surface roughness of 4.8μm; 2) Structural morphology optimization: introducing an inward concave curve (85°) conforming to the curvature of the human hand's web, and pre-installing an inflatable spinal channel on the gripping side; 3) Material thickness adjustment: increasing the local thickness to 0.20mm and improving the surface hardness to Shore A hardness 60. After optimization, the overall closeness is recalculated to 0.612, 0.612>0.5, and the correlation coefficient between tactile texture and emotional safety is 0.97, 0.97≥0.95, meeting user experience requirements.

Claims

1. A packaging dynamic evaluation system based on the entropy weight TOPSIS method, characterized in that, It includes a data sensing and acquisition module, an entropy weight dynamic weighting module, a TOPSIS comprehensive evaluation module, a cross-sensory mapping analysis module, and a design parameter reconstruction module, which are connected in sequence. The data perception and acquisition module collects multimodal perception data of users on product packaging and builds a user perception database; The entropy weight dynamic weighting module calculates the information entropy value of each perception dimension based on the collected multimodal perception data, dynamically calculates the weight coefficient of each evaluation indicator, and eliminates subjective weighting bias. The TOPSIS comprehensive evaluation module constructs a space of positive and negative ideal solutions, calculates the Euclidean distance between the packaging scheme and the ideal solution, and generates a comprehensive closeness score. The cross-sensory mapping analysis module calculates the correlation matrix between all evaluation indicators and verifies the coupling strength of the preset core coupling indicator pairs. The design parameter refactoring module refactors parameters when the packaging design does not meet the requirements.

2. The packaging dynamic evaluation system based on the entropy weight TOPSIS method according to claim 1, characterized in that, The data perception and acquisition module includes a visual perception acquisition unit, a tactile perception acquisition unit, a functional perception acquisition unit, an emotion perception acquisition unit, and an environmental standardization control unit. The visual perception acquisition unit collects visual evaluation data of users on the color, pattern and layout of the packaging. The visual evaluation data includes one or more of the following: questionnaire scores, image recording data and eye-tracking data. The tactile sensing acquisition unit collects tactile evaluation data when the user holds the packaging. The tactile evaluation data includes one or more of the following: material tactile score, pressure distribution data, surface roughness data, friction coefficient data, and temperature change data. The functional perception and acquisition unit collects functional experience data when the user opens the packaging. The functional experience data includes one or more of the following: opening convenience score, opening time, opening force, and number of action steps. The emotion perception acquisition unit collects users' emotional evaluation data on packaging safety and environmental protection. The emotional evaluation data includes one or more of questionnaire scores and physiological signal data. The environmental standardization control unit constructs a test environment with constant temperature, humidity, and illuminance to ensure the consistency and comparability of data acquisition.

3. The packaging dynamic evaluation system based on the entropy weight TOPSIS method according to claim 2, characterized in that, In the aforementioned entropy weight dynamic weighting module, the evaluation indicators are considered from the dimensions of visual representation, tactile interaction, functional experience, and emotional mapping. The visual representation dimension includes visual attractiveness. and class recognition Two evaluation metrics: the tactile interaction dimension includes tactile texture. and grip comfort Two evaluation metrics: the functional experience dimension includes ease of activation. One evaluation metric, the emotion mapping dimension, includes emotional security. and environmental awareness Two evaluation indicators; the first The tactile texture of each packaging solution The calculation method is as follows: ； in, For the number of tests, For the first Packaging scheme #1 The surface friction coefficient obtained from this test and then normalized. For the first Packaging scheme #1 The normalized temperature comfort level after this test For the first Packaging scheme #1 The normalized perceived hardness value from this test. This is the weighting coefficient for the surface friction coefficient. This is a weighting factor for temperature comfort. This is the weighting coefficient for the perceived hardness value, and ; Apart from tactile texture, other evaluation indicators are calculated according to a unified framework of "sub-indicator collection—directional consistency—normalization—intra-dimensional weighting," that is: Let the... The first packaging scheme The first evaluation indicator Sub-indicators are ,like As a benefit-oriented indicator, according to Normalize, For the normalized first The first packaging scheme The first evaluation indicator Sub-indicators, For the first The first evaluation indicator The minimum value of each sub-indicator For the first The first evaluation indicator The maximum value of each sub-indicator; if As a cost-type indicator, according to Perform normalization; if If the evaluation score is already on a 1-10 scale, then take it directly. ; No. Packaging scheme #1 Output value of each evaluation indicator for: ； in, For the first The number of sub-indicators under each evaluation indicator For the first Within the evaluation indicators, the first Weights within the dimensions of each sub-indicator.

4. The packaging dynamic evaluation system based on the entropy weight TOPSIS method according to claim 3, characterized in that, In the dynamic entropy weighting module, the formula for calculating the information entropy value is: ； in, For the first The information entropy value of each evaluation indicator For the total number of packaging options, It is the natural logarithm function. , for Zhan Di The proportion of the standardized values ​​of all packaging schemes to the total value of each evaluation indicator. For the first The packaging scheme is in the first The standardized values ​​of each evaluation indicator For the index of packaging schemes, For the index of evaluation indicators; when hour, Process as 0.

5. The packaging dynamic evaluation system based on the entropy weight TOPSIS method according to claim 4, characterized in that, The TOPSIS comprehensive evaluation module includes a standardization processing unit, a weighted decision matrix construction unit, an ideal solution space construction unit, and a distance calculation unit. The standardization unit uses the range transformation method to standardize the original data. The standardization formula is as follows: ； in, For the first The packaging scheme is in the first The standardized values ​​of each evaluation indicator For the index of packaging schemes, For the index of evaluation indicators, For the first The packaging scheme is in the first The original observed values ​​of each evaluation indicator and The first The maximum and minimum values ​​of each evaluation indicator; The weighted decision matrix construction unit multiplies the standardized value by the entropy weight result to construct the weighted standardized decision matrix. Weighted standardized decision matrix The Middle Line number The elements of the column are , For the index of packaging schemes, , For the total number of packaging options, For the index of evaluation indicators, , The total number of evaluation indicators, , For the first The entropy weight coefficient of the first evaluation indicator; Coefficient of difference of each evaluation indicator , No. Entropy weight coefficient of each evaluation indicator The calculation formula is: ； in, The total number of evaluation indicators, ; Ideal solution space building blocks are used to determine positive ideal solutions. and negative ideal solution , , The first in the positive ideal solution One element, , This represents taking the weighted standardized decision matrix. The Middle The maximum value of the column elements; The first in the positive ideal solution One element, , The first of the negative ideal solutions One element, , This represents taking the weighted standardized decision matrix. The Middle The minimum value of a column element. The first in the set of negative ideal solutions One element; The distance calculation unit calculates the Euclidean distance from each packaging scheme to the positive ideal solution and the negative ideal solution. The calculation formula is as follows: ； ； in, For the first The distance between each packaging solution and the ideal solution. For the first The distance from each packaging scheme to the negative ideal solution; The formula for calculating overall similarity is: ； in, Indicates the first The overall fit of the packaging solutions.

6. The packaging dynamic evaluation system based on the entropy weight TOPSIS method according to claim 5, characterized in that, The cross-sensory mapping analysis module includes a correlation analysis unit, a coupling strength verification unit, a cross-modal mapping model construction unit, and a perception threshold determination unit; Calculate the Pearson correlation coefficient matrix among the various sensory dimensions of each evaluation index. , ,, For the first The evaluation index and the first Pearson correlation coefficients among the evaluation indicators; correlation analysis units are... As the result of full-scale cross-sensory relationship recognition, it is used to output the correlation strength between various evaluation indicators and to determine the set of strongly coupled indicator pairs. ,gather The indicators in the middle are for meeting and ,in This is a strong coupling screening threshold; The coupling strength verification unit extracts the correlation coefficients corresponding to preset core coupling index pairs from the Pearson correlation coefficient matrix and uses them as constraints in addition to the comprehensive evaluation for verification. The preset core coupling index pairs include tactile texture and emotional safety, and their corresponding correlation coefficients are , which represent the Pearson correlation coefficient between tactile texture and emotional safety. The coupling strength verification unit verifies whether the preset coupling threshold is reached, and other correlation coefficients are used as auxiliary diagnostic relationships to identify the linkage characteristics between visual, tactile, functional and emotional indicators. If the value is less than the preset coupling threshold, an alarm will be triggered and the corresponding packaging scheme will be sent to the design parameter refactoring module. The cross-modal mapping model building unit establishes a nonlinear mapping function between the physical parameters of touch and the sense of security. The nonlinear mapping function is as follows: ； in, Rate your sense of security. For material thickness, It is the natural logarithm function. Surface hardness, This is a weighting factor for material thickness. This is a weighting coefficient for surface hardness. For the fitting parameters, , and The data was obtained by fitting the sample packaging material thickness, surface hardness, and safety perception score data; the target safety perception score was... Surface hardness When given, the material thickness adjustment unit adjusts according to... Calculate the target material thickness and limit it to the range of 0.15mm-0.25mm; if the target safety score is still not met after the thickness reaches the boundary of the range, adjust the surface hardness to Shore A hardness of 50-70. The perception threshold determination unit determines the critical perception threshold of tactile texture. When the tactile texture score is lower than the critical perception threshold, the sense of emotional security decreases.

7. The packaging dynamic evaluation system based on the entropy weight TOPSIS method according to claim 6, characterized in that, The design parameter reconstruction module includes a parameter adjustment target determination unit, a candidate parameter generation unit, a parameter constraint unit, and an effect verification unit; The parameter adjustment target determination unit determines the type of packaging design parameters to be adjusted based on the evaluation index that the overall closeness does not reach the preset threshold and the core coupling index pair that does not reach the preset coupling threshold. The candidate parameter generation unit generates at least one candidate adjustment parameter from among the following: packaging surface tactile parameters, packaging structural morphology parameters, material thickness parameters, and material hardness parameters, based on the type of packaging design parameter to be adjusted. The parameter constraint unit filters and constrains candidate adjustment parameters based on packaging materials, packaging structure, processing technology, and safety requirements to generate an optimized packaging solution. The effect verification unit retests the evaluation indicators of the optimized packaging scheme, and re-executes the entropy weight TOPSIS comprehensive evaluation and core coupling constraint verification to determine whether the comprehensive closeness of the optimized packaging scheme reaches the preset threshold and whether the correlation coefficient of the preset core coupling indicators reaches the preset coupling threshold.

8. A packaging dynamic evaluation method based on the entropy weight TOPSIS method, implemented based on the packaging dynamic evaluation system based on the entropy weight TOPSIS method as described in any one of claims 1-7, characterized in that, Includes the following steps: Step 1, Multimodal Perception Data Acquisition: In a standardized environment, collect multimodal perception data from multiple target users on several packaging schemes; Step 2, Dynamic Entropy Weighting: Based on multimodal sensing data, calculate the information entropy value of each sensing dimension and dynamically determine the weight coefficients of the evaluation indicators; Step 3, Comprehensive Evaluation: Construct a space of positive and negative ideal solutions, calculate the comprehensive closeness of each packaging scheme, and rank them; Step 4, cross-sensory mapping analysis: calculate the correlation matrix between all evaluation indicators and verify the coupling strength of the preset core coupling indicator pairs; Step 5, Design Packaging Scheme Verification: Verify the new packaging design from multiple dimensions to determine whether it meets user experience requirements; Step 6, Design Parameter Reconstruction: If the new packaging design does not meet the requirements, the parameter reconstruction module will be activated to generate an optimized packaging design.

9. The packaging dynamic evaluation method based on entropy weight TOPSIS according to claim 8, characterized in that, Step five involves the following specific process for validating the packaging design: First, the new packaging design scheme is input into the data perception and acquisition module to obtain perception data of the new packaging design scheme on 7 evaluation indicators; Then, the overall fit of the new packaging design is calculated using the entropy weight dynamic weighting module and the TOPSIS comprehensive evaluation module. Next, the overall relevance of the new packaging design will be compared with the preset threshold. For comparison, the overall similarity is used to characterize the packaging scheme's performance. to The overall performance across all evaluation metrics; simultaneously, using pre-set core coupling metrics as baseline constraints, to verify tactile texture. With emotional security Pearson correlation coefficient Does it meet the requirement of being greater than or equal to the preset coupling threshold? ; If the overall relevance of the new packaging design is less than the preset threshold or tactile texture With emotional security Pearson correlation coefficient Less than the preset coupling threshold If the new packaging design does not meet user experience requirements, the parameters will be refactored. Finally, if the requirements are still not met after three consecutive parameter reconstructions, an exception report will be generated and a redesign prompt report will be output.

Images