An old custom cabinet storage partition requirement acquisition method and system

Abstract

The application discloses a kind of old custom cabinet storage partition demand acquisition method and system.The method includes: preference information pre-acquisition step, obtains the actual storage area information of old user and declarative preference selection;Question modeling and coding step, cabinet storage area is divided into matrix type grid layout and establishes chromosome coding;Initial population generation step, initial population is generated using orthogonal design method;Interactive genetic evolution step, obtain user score as fitness value by visual scheme presentation, execute hierarchical selection and cross variation operation Output satisfactory scheme set;Consistency check and decision step, actual storage, declarative preference and display preference are compared by three parties, identify certainty preference and uncertainty preference, generate optimal scheme.The application solves the problem of insufficient intuitiveness and poor accuracy of existing methods, and provides reliable demand decision basis for enterprises.

Description

Technical Field

[0001] This invention belongs to the interdisciplinary field of home intelligent design and computer application technology, specifically a method and system for obtaining storage zoning requirements of customized cabinets for the elderly based on an interactive genetic algorithm. Background Technology

[0002] As the population ages and the elderly population continues to expand, age-friendly product design has become a key focus for the home furnishing industry. The kitchen, as the core functional space of family life, directly impacts the daily quality and safety of elderly users through the rational design of its cabinetry and storage system. Studies show that older adults spend a significant amount of time in the kitchen daily, and their long-term cooking experience has fostered relatively stable storage habits and space usage preferences. Therefore, adapting to a new kitchen environment and the storage layout of new cabinets often presents a challenge.

[0003] In custom cabinet services, accurately understanding the storage and organization preferences of elderly users is a prerequisite for achieving personalized design. However, existing methods for obtaining preferences mainly rely on face-to-face conversations between sales staff and elderly users. This traditional approach has the following prominent problems: First, it lacks intuitiveness, as elderly users find it difficult to accurately express their preferences for abstract storage solutions through verbal descriptions; second, it is not timely, as a complete needs assessment often requires multiple communications, which is time-consuming and geographically limited; third, it lacks accuracy, as elderly users' cognitive and expressive abilities are limited, and there is often a discrepancy between their stated preferences and actual usage habits; fourth, communication barriers exist during the process, making it difficult for sales staff to effectively guide elderly users to express their deeper needs.

[0004] From the perspective of preference theory, user preferences can be divided into two types: declarative preferences and revealed preferences. Declarative preferences refer to the subjective intentions that users express directly through language, while revealed preferences refer to the true tendencies reflected through users' actual choices. Due to factors such as cognitive limitations, expression barriers, and situational differences, there is often an inconsistency between the declarative and revealed preferences of elderly users. This preference uncertainty makes it difficult for companies to accurately grasp the true needs of elderly users, thereby affecting user satisfaction with customized cabinet products.

[0005] In existing technologies, conjoint analysis is widely used in consumer preference research. This method designs orthogonal experimental schemes, requiring users to evaluate multiple attribute combinations to infer the utility value of each attribute level. However, conjoint analysis requires users to evaluate a large number of scheme combinations, which places a heavy cognitive burden on elderly users. Furthermore, this method essentially still falls under the category of declarative preference acquisition and cannot effectively solve the problem of preference uncertainty. Interactive genetic algorithms, as a human-machine collaborative optimization method, can incorporate user subjective evaluations into the evolutionary process, gradually approximating user preferences through iterative optimization. However, current applications are mainly concentrated in fields such as industrial design and graphic generation, and have not yet been adapted to the cognitive characteristics of elderly users and the special needs of cabinet storage scenarios.

[0006] Furthermore, existing preference acquisition methods generally lack verification mechanisms for the consistency between declarative and revealed preferences, failing to identify which preferences are certain preferences that users clearly recognize, and which are uncertain preferences that users are not yet clearly aware of. This deficiency leads to a lack of reliable decision-making basis for enterprises when formulating storage and zoning solutions, making it difficult to strike a balance between meeting users' personalized needs and controlling production costs.

[0007] Therefore, there is an urgent need for a method to obtain the storage and zoning needs of elderly customized kitchen cabinets that can overcome the above-mentioned technical defects. This method should have the following characteristics: it should be able to present abstract storage preferences in an intuitive and visual way, reducing the cognitive burden on elderly users; it should be able to efficiently obtain users' explicit preferences through interactive optimization, shortening the demand survey cycle; and it should be able to establish a consistency verification mechanism between declarative preferences and explicit preferences, identify deterministic preferences and uncertain preferences, thereby providing enterprises with a more accurate and reliable basis for demand decision-making. Summary of the Invention

[0008] To address the shortcomings of existing methods for obtaining storage preferences in customized kitchen cabinets for the elderly, such as insufficient intuitiveness, weak timeliness, poor accuracy, and lack of preference consistency verification mechanisms, this invention aims to provide a method and system for obtaining storage zoning needs in customized kitchen cabinets for the elderly. This method transforms the storage zoning optimization problem into a grid filling problem, combines orthogonal design and interactive genetic algorithms for iterative optimization, and establishes a three-way comparison mechanism between declarative and explicit preferences. This enables accurate and efficient acquisition of storage zoning needs of elderly users, effectively identifies preference uncertainties, and provides enterprises with reliable demand decision-making basis.

[0009] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides a method for obtaining the storage and partitioning needs of a customized kitchen cabinet for the elderly, comprising the following steps: S1 Preference Information Pre-collection Steps: Obtain information on the actual storage areas of various kitchen items in the current cabinets of elderly users, as well as the declarative preference choices of elderly users for the storage areas of various kitchen items.

[0010] The actual storage area information is obtained through in-home observation or image capture, reflecting the actual storage habits formed by elderly users over a long period of use. The declarative preference selection is obtained through direct questioning, recording the elderly users' subjective expressions of their desired storage areas for various items. This step may also include showing elderly users a storage style board, which presents typical storage methods for various kitchen items, to expand their understanding of storage methods and optimize their initial expression of needs.

[0011] S2 Problem Modeling and Coding Steps: Divide the cabinet storage area into a matrix grid layout according to a multi-dimensional spatial partitioning method and number it. Determine the set of fillable grid areas based on the location of fixed facilities. Establish a set of optional filling positions for various items to be stored to form item grid filling constraint rules. Use binary encoding to encode the filling positions of various items to form a chromosome representation.

[0012] The multi-dimensional spatial zoning method includes horizontal functional zoning and vertical spatial zoning. The horizontal functional zoning includes a cooking area, a food preparation area, and a washing area, corresponding to the cooking, food preparation, and washing operations in the kitchen, respectively. The vertical spatial zoning includes base cabinets, countertops, and walls, corresponding to storage spaces at different height levels within the cabinet system.

[0013] The fixed facilities include a gas stove and a sink. Because a range hood needs to be installed above the gas stove and there is a risk of open flame during operation, the countertop and wall-mounted shelves corresponding to the gas stove are removed. Similarly, because drainage pipes need to be installed under the sink and the countertop area needs to maintain a clean working space, the cabinet and countertop shelves corresponding to the sink are removed. After removing the aforementioned non-fillable shelves, the remaining shelves constitute the set of fillable shelf areas.

[0014] The item grid filling constraint rules are determined based on the usage characteristics of various kitchen items and the human-machine ergonomic correspondence of each functional area. The item grid filling constraint rules also include the following allocation constraints: (1) Integrity constraint: Each type of item to be stored must be assigned to a unique filling position in its set of optional filling positions, wherein the filling position is a single grid or a predefined grid combination; (2) Partition sharing constraint: Multiple different types of items can be accommodated in the same functional area; (3) Grid mutual exclusion constraint: Only one type of item can be accommodated in the same grid to avoid space conflicts. For example, seasonings and spatulas, which are frequently used during cooking, can be placed in wall shelves, countertop shelves, or cabinet shelves in the cooking area; knives, used for food preparation, can be placed in wall shelves, countertop shelves, or cabinet shelves in the food preparation area; cutting boards, used for both food preparation and drying after washing, can be placed in wall shelves, countertop shelves, or cabinet shelves in the food preparation or washing area; dishcloths, used for cleaning, can be placed in wall shelves, countertop shelves, or cabinet shelves in the washing area; bowls, plates, chopsticks, forks, and spoons, used for storing food after cooking and after washing, can be placed in cabinet shelves in the cooking area or wall shelves and countertop shelves in the washing area. In this invention, the storage area corresponds to the area unit formed by the combination of the horizontal functional partitions and the vertical spatial partitions. Each storage area contains several numbered shelves; the utility value is calculated using the storage area as the statistical unit, and the placement of the storage area can be mapped to the corresponding storage area for utility value calculation.

[0015] The binary encoding method determines the corresponding number of bits based on the number of optional filling positions for each type of item and the system implementation requirements. The position codes of each type of item are concatenated sequentially to form a chromosome representing a complete storage partitioning scheme.

[0016] S3 Initial population generation steps: An orthogonal array is generated using an orthogonal design method. Individuals are selected from the orthogonal array and combined with randomly generated individuals to form the initial population.

[0017] The orthogonal design method constructs an orthogonal array to cover uniform combinations of factor levels with fewer trials, thereby significantly reducing the number of initial schemes while ensuring population diversity. Preferably, the ratio of the number of individuals selected from the orthogonal array to the number of randomly generated individuals is 4-6:1, more preferably 5:1. For example, 25 individuals are selected from the orthogonal array as the initial population body, and another 5 individuals are randomly generated to supplement the initial population, so as to ensure random diversity of the population based on the systematic coverage of the orthogonal design.

[0018] S4 Interactive Genetic Evolution Steps: Decode each individual in the population into a partitioning scheme and present it to elderly users in a visual manner. Obtain the subjective rating of the elderly users as the fitness value. Based on the fitness value, perform hierarchical selection and perform crossover and mutation operations on individuals to be optimized to generate a new generation of population. Iterate the above process until the preset termination condition is met and output a set of satisfactory schemes.

[0019] The preferred visualization method is a rectangular planar graphic, in which the cooking area, food preparation area, and washing area are arranged horizontally from left to right, and the base cabinet area, countertop area, and wall area are arranged vertically from bottom to top. Various items to be stored are labeled with planar icons in their corresponding grid positions. This visualization method transforms abstract storage solutions into intuitive graphic representations, making them easier for elderly users to understand and evaluate.

[0020] The graded selection employs a dual-threshold mechanism. Preferably, the first threshold is set to 80 points and the second threshold to 20 points, using a percentage-based scoring system. Individuals with fitness values ​​higher than the first threshold are considered satisfactory and are directly replicated to the next generation population and stored in the satisfactory scheme set. Individuals with fitness values ​​lower than the second threshold are considered unsatisfactory and are eliminated, with new individuals randomly selected from the orthogonal array to replace them. Individuals with fitness values ​​between the first and second thresholds are considered individuals to be optimized and are sent to subsequent crossover and mutation operations.

[0021] The crossover operation includes: selecting two parent individuals according to a preset crossover probability, randomly determining the crossover gene position, and exchanging gene segments of the two parent individuals at that gene position to generate two new individuals. The mutation operation includes: selecting a parent individual according to a preset mutation probability, randomly selecting a mutation position, and randomly selecting a position from the set of available filling positions according to the item type corresponding to the mutation position to replace the original position, forming a new mutated individual.

[0022] Furthermore, when the number of iterations exceeds a preset iteration threshold and the termination condition is still not met, the selection strategy switches to a roulette wheel selection method. This involves calculating the proportion of each individual's fitness value to the total fitness value of the population in each generation, and allocating the probability of each individual being selected for crossover and mutation operations according to this proportion. The preset iteration threshold is preferably 10 generations. This threshold is set considering the testing fatigue of older users and the efficiency advantage of hierarchical selection in rapid screening in the early stages. In practical applications, it can be adjusted based on user feedback and convergence results.

[0023] The preset termination conditions include: the number of solutions in the set of satisfactory solutions reaches a preset target number, or the elderly user issues a termination iteration command through the interactive interface to indicate that an ideal solution has been found. The preset target number is preferably 6.

[0024] S5 Consistency Check and Decision-Making Steps: Based on the fitness values ​​of each solution in the set of satisfactory solutions, the utility values ​​of various items in each storage area are statistically calculated. The actual storage area information, the declarative preference selection, and the explicit preference reflected by the utility values ​​are compared to identify deterministic preferences and uncertain preferences. Based on this, the optimal storage partition outline solution is generated.

[0025] The utility value is calculated as follows: The candidate storage area is a set of storage areas mapped from the set of possible filling positions for that type of item. For each candidate storage area of ​​each type of item, the number of solutions in the set of satisfactory solutions that assign the item to that area and their corresponding fitness values ​​are counted. These fitness values ​​are summed and divided by the total number of satisfactory solutions to obtain the utility value of the item in that area. The calculation formula is: ; For the first Items of this type in the first The utility value of the storage area; For a set of satisfactory solutions, The total number of its options; For a satisfactory solution fitness value; As an indicator function, when the scheme The Middle Items of this type are assigned to the first The value is 1 if it is in the specified region, and 0 otherwise.

[0026] The utility value reflects the degree of preference shown by elderly users through their actual selection behavior in the interactive genetic evolution process. A higher utility value indicates a stronger preference among elderly users for storing the item in that area. The revealed preference refers to the true preference tendency reflected by the actual selection behavior of elderly users in the interactive genetic evolution process.

[0027] The rule for identifying deterministic and uncertain preferences is as follows: Regions with a utility value greater than zero in revealed preferences are considered candidate regions. When the actual storage area of ​​the same item matches the region selected in declarative preferences, and this matching region belongs to the candidate region, the storage area preference for that item is determined to be a deterministic preference; otherwise, the storage area preference for that item is determined to be an uncertain preference. A utility value greater than zero indicates that the region has been accepted at least once by older users through rating behavior during the interactive genetic evolution process.

[0028] The method for generating the optimal storage partition outline includes: for items corresponding to deterministic preferences, keeping their original storage area settings unchanged; for items corresponding to uncertain preferences, first selecting the area with the highest utility value from the candidate areas; if there are multiple candidate areas with similar utility values, then prioritizing the area consistent with the declarative preference as the optimal storage area; and generating the optimal storage partition outline by combining the storage area determination results of various items.

[0029] Secondly, the present invention provides a system for obtaining storage and partitioning needs of customized cabinets for the elderly, used to execute the above method, including a user decision-making layer and an enterprise decision-making layer.

[0030] The user decision-making level for elderly users' preference acquisition and solution optimization includes the following functional modules: a preference pre-collection module, used to acquire information on the actual storage areas of various kitchen items in the elderly user's current cabinet and the elderly user's declarative preference selection for storage areas of various kitchen items; this module also includes a storage style dashboard module, used to display various kitchen item storage styles to the elderly user to expand their understanding of storage forms.

[0031] The problem modeling module is used to divide the cabinet storage area into a matrix grid layout, define the grid filling constraints for storage items, and perform chromosome coding on the filling positions of various items.

[0032] The population initialization module is used to generate orthogonal arrays and construct the initial population based on orthogonal design methods.

[0033] The interactive genetic evolution module is used to decode the population individuals into storage partition outline schemes for visualization, obtain subjective ratings from elderly users, perform hierarchical selection and crossover mutation operations, and control the iteration process to output a set of satisfactory schemes. This module includes a rating acquisition unit, a selection operation unit, a crossover operation unit, a mutation operation unit, and an iteration control unit. The selection operation unit adopts a hierarchical selection mechanism and switches to a roulette wheel selection method after the number of iterations exceeds a preset iteration threshold. The interactive genetic evolution module visualizes the storage partition outline scheme as a rectangular planar graphic, in which the cooking area, food preparation area, and washing area are arranged horizontally from left to right, and the cabinet area, countertop area, and wall area are arranged vertically from bottom to top.

[0034] The enterprise decision-making level provides demand analysis and decision support for cabinet customization enterprises, including the following functional modules: a consistency verification module, which calculates utility values ​​based on the set of satisfactory solutions and compares the actual storage area information, declarative preference selection, and the revealed preferences reflected by the utility values ​​to identify deterministic and uncertain preferences; this module adopts a three-way comparison rule, using areas with utility values ​​greater than zero in the revealed preferences as candidate areas for judgment.

[0035] The decision optimization module is used to generate storage partitioning requirement decision suggestions and determine the optimal storage partitioning outline scheme based on the consistency test results; this module combines the consistency test results with the enterprise's production feasibility and cost factors to generate the final decision suggestions.

[0036] Compared with the prior art, the present invention has the following beneficial effects: (1) This invention presents the abstract storage partition scheme to elderly users in a visual way. Elderly users only need to rate the intuitive graphic scheme to express their preferences. There is no need to describe the abstract concepts in words, which effectively reduces the cognitive burden of elderly users, improves the accuracy of obtaining preference information, solves the problem of insufficient intuitiveness in the prior art leading to the distortion of demand information, and avoids the communication barrier between sales personnel and elderly users in the traditional way.

[0037] (2) This invention uses an orthogonal design method to generate an initial population, achieving uniform coverage of the solution space with a smaller number of schemes; it employs a hierarchical selection mechanism to quickly screen satisfactory schemes and eliminate inferior schemes; and it uses an interactive genetic evolution method to gradually approach user preferences. Compared with traditional joint analysis methods that require users to evaluate a large number of scheme combinations, this invention can reduce the number of evaluations by elderly users, shorten the demand survey cycle, and is not limited by geographical location, thus solving the problem of weak timeliness in the prior art.

[0038] (3) This invention creatively establishes a three-way comparison mechanism between declarative preferences, revealed preferences, and actual storage behavior, and identifies deterministic and uncertain preferences through consistency checks. For items with deterministic preferences, their storage area can be directly determined; for items with uncertain preferences, optimization suggestions are generated by combining utility value distribution and consistency degree. This mechanism can improve the problem of difficulty in identifying preference uncertainty in the prior art and improve the reliability of demand decisions.

[0039] (4) The present invention constructs a two-layer system architecture that includes a user decision-making layer and an enterprise decision-making layer. The user decision-making layer ensures that the personalized preferences of elderly users are fully expressed and identified. The enterprise decision-making layer generates final decision suggestions based on the consistency test results and the factors of production feasibility and cost, thereby achieving a balance between user satisfaction and enterprise benefits.

[0040] (5) The present invention has made adaptive designs for the cognitive characteristics of elderly users, including adopting an intuitive planar graphic visualization method, setting up a storage style board to expand cognition, adopting a simple scoring interaction method, and reducing the number of evaluations through orthogonal design, which effectively alleviates the test fatigue problem of elderly users and improves the usability of the system. Attached Figure Description

[0041] Figure 1This is a flowchart illustrating the method for obtaining storage partitioning requirements of a customized cabinet for the elderly according to the present invention. Figure 2 This is a schematic diagram of the structure of a system for obtaining storage and partitioning needs of a customized cabinet for the elderly, according to the present invention. Detailed Implementation

[0042] The present invention will be further described below with reference to specific embodiments, but the invention is not limited to these embodiments. Those skilled in the art should recognize that the present invention covers all possible alternatives, improvements, and equivalents included within the scope of the claims.

[0043] In this invention, unless otherwise stated, the following terms have the following meanings: Storage zoning outline scheme: refers to the spatial distribution and configuration scheme of various items in the cabinet storage area, which is represented in the form of a matrix grid layout.

[0044] Grid: refers to the basic spatial unit after the cabinet storage area is divided into a matrix.

[0045] Fill position: refers to the cell number or cell combination that a certain type of item can be assigned in a matrix grid layout.

[0046] Declarative preferences: These refer to users' subjective desire to store items, expressed directly through language.

[0047] Revealed preferences: refer to the true preference tendencies reflected through users' actual selection behavior (such as rating).

[0048] This invention provides a method for obtaining the storage zoning needs of customized kitchen cabinets for the elderly. The method targets elderly people aged 55-65 who are often cooks, and identifies seven categories of commonly used kitchen items to be stored: knives, bowls and plates, cutting boards, dishcloths, seasonings, chopsticks, forks and spoons, and cookware. To improve the system's coding accuracy, this invention further subdivides these seven categories into 10 subcategories based on characteristics such as volume and frequency of use, using chromosome coding. Seasonings are further subdivided into small-volume and large-volume seasonings, and cookware is subdivided into ladles, spatulas, pot bodies, and lids. Figure 1 As shown, the method includes the following steps: S1 Preference Information Pre-collection Steps The preference information pre-collection step is used to obtain the initial preference information of elderly users, including actual storage area information and declarative preference selections.

[0049] The actual storage area information was obtained through in-home observation, recording the actual storage location of 7 types of items in the elderly user's kitchen, including the vertical spatial location (upper cabinet area, wall area, countertop area or base cabinet area) and the horizontal functional location (cooking area, food preparation area or washing area).

[0050] Declarative preference selection was obtained through direct questioning, asking elderly users about their subjective preferences for ideal storage areas for seven types of items. The questioning was conducted using single-choice or multiple-choice methods.

[0051] Before the formal inquiry, show the elderly users a storage style display board. The display board presents various typical storage methods for seven types of kitchen items in a graphic and textual way, including wall-mounted, countertop shelf, pull-out basket, and drawer types in base cabinets, to help the elderly users expand their knowledge of storage methods.

[0052] S2 Problem Modeling and Coding Steps The problem modeling and coding steps are used to transform the storage and partitioning problem into a mathematical model that can be handled by a genetic algorithm.

[0053] Matrix-style grid layout: The cabinet storage area is divided into a matrix-style grid layout according to a multi-dimensional spatial partitioning method. Horizontal functional partitions, from left to right, are the cooking area, food preparation area, and washing area; vertical spatial partitions, from bottom to top, are the base cabinet area, countertop area, and wall area. Based on the statistical results of pre-collected preference information, elderly users rarely choose the top cabinet area as a storage location for frequently used items; therefore, the matrix-style grid layout of this invention does not include a top cabinet partition. Through the combination of three horizontal and three vertical partitions, the cabinet storage area is divided into a nine-square matrix layout, with each partition further subdivided into several numbered grids, totaling 36 grids.

[0054] The set of fillable grid areas is determined by eliminating unfillable areas based on the location of the fixed facilities. For example, if the gas stove is located in the cooking area, its corresponding countertop and wall grids are eliminated; if the sink is located in the washing area, its corresponding base cabinet and countertop grids are eliminated. The remaining grids constitute the set of fillable grid areas.

[0055] Establishment of item grid filling constraint rules: Based on the usage characteristics of 7 types of kitchen items and the human-machine ergonomic correspondence of each functional area, establish item grid filling constraint rules. Seasonings can be filled in the wall shelves, countertop shelves, or cabinet shelves in the cooking area, with a total of 6 possible locations; ladles and spatulas can be filled in the wall shelves, countertop shelves, or cabinet shelves in the cooking area, with a total of 5 possible locations; knives can be filled in the wall shelves, countertop shelves, or cabinet shelves in the food preparation area, with a total of 5 possible locations; cutting boards can be filled in the wall shelves, countertop shelves, or cabinet shelves in the food preparation area or washing area, with a total of 4 possible locations; dishcloths can be filled in the wall shelves, countertop shelves, or cabinet shelves in the washing area, with a total of 4 possible locations; bowls and plates can be filled in the cabinet shelves in the cooking area or the wall shelves and countertop shelves in the washing area, with a total of 5 possible locations; chopsticks, forks, and spoons can be filled in the cabinet shelves in the cooking area or the wall shelves and countertop shelves in the washing area, with a total of 3 to 4 possible locations.

[0056] Chromosome Coding: To improve coding accuracy, this invention employs a 10-category subdivision scheme for chromosome coding. Seasonings are subdivided by volume into small-volume seasonings (number 0) and large-volume seasonings (number 8). Cookware is subdivided by function into ladles and spatulas (number 1), pot bodies (number 7), and pot lids (number 9). The remaining five categories are numbered sequentially as follows: knives (number 2), cutting boards (number 3), dishcloths (number 4), bowls and plates (number 5), and chopsticks, forks, and spoons (number 6). A binary coding method is used to encode the filling positions of each type of item. The number of bits in the code is determined based on the number of selectable filling positions for each type of item and the system implementation requirements. The position codes of the 10 categories of items are concatenated sequentially to form a 46-bit chromosome, representing a complete storage and partitioning scheme. The correspondence between item numbers and chromosome codes is as follows: .

[0057] S3 Initial Population Generation Steps The initial population generation step is used to generate the initial solution set for the genetic algorithm.

[0058] An orthogonal design method was used. The number of factors and levels of the orthogonal array were determined based on the number of item types and the number of possible locations for each item. The orthogonal array was then generated using the orthogonal design module of statistical analysis software. Twenty-five individuals were selected from the orthogonal array as the main body of the initial population, and five additional individuals were randomly generated to supplement the initial population, resulting in an initial population of 30 individuals. The ratio of the number of individuals selected from the orthogonal array to the number of randomly generated individuals was 5:1.

[0059] The advantage of using orthogonal design to generate the initial population is that orthogonal arrays have the characteristics of balanced distribution and neat comparability, which can cover a large solution space with fewer trials, making the initial population evenly distributed in the solution space and avoiding the genetic algorithm from getting trapped in local optima.

[0060] S4 Interactive Genetic Evolutionary Steps The interactive genetic evolution step is used to drive the evolution of the genetic algorithm through the subjective evaluation of elderly users, and gradually optimize the storage partitioning scheme.

[0061] Individual Decoding and Visualization: The system decodes each chromosome in the population into a storage partition outline, presenting it to elderly users as a rectangular planar graphic. In this graphic, the cooking area, preparation area, and washing area are arranged horizontally from left to right, and the cabinet area, countertop area, and wall area are arranged vertically from bottom to top. Seven categories of items to be stored are labeled with intuitive planar icons in their corresponding grid positions. During visualization, the system automatically merges the 10 categories into 7 for display; for example, small-volume seasonings are combined with large-volume seasonings and displayed as "seasonings," while ladles, spatulas, pots, and lids are combined and displayed as "cookware," reducing the cognitive burden on elderly users. Each round, six storage partition outlines are presented to elderly users for their evaluation.

[0062] Subjective rating acquisition: Elderly users subjectively rate the six presented storage partition outlines based on their personal preferences. The rating is based on a 100-point scale, and the subjective rating is used as the fitness value for the corresponding individual.

[0063] Hierarchical selection operation: Individuals in the population are selected hierarchically based on their fitness values, with a first threshold of 80 points and a second threshold of 20 points. Individuals with fitness values ​​higher than 80 points are considered satisfactory and are directly replicated to the next generation of the population and stored in the satisfactory scheme set. Individuals with fitness values ​​lower than 20 points are considered unsatisfactory and are eliminated, with new individuals randomly selected from the orthogonal array to replace them. Individuals with fitness values ​​between 20 and 80 points are considered individuals to be optimized and are sent to subsequent crossover and mutation operations.

[0064] Crossover operation: Set the crossover probability to 0.8. Select two parent individuals Pa and Pb according to the probability, randomly determine the crossover gene position X, obtain the start and end positions of the item corresponding to this gene position in the chromosome according to the chromosome coding data structure, exchange the gene segments of the two parent individuals in this interval, and generate two new individuals Pc and Pd.

[0065] Mutation operation: Set the mutation probability to 0.1. Select the parent individual Pb according to the probability, randomly select the item number corresponding to the mutation bit X, and randomly select a position from the set of possible filling positions of the item to replace the original position, forming a new individual after mutation.

[0066] Constraint Satisfaction Handling: After the initial population generation, crossover operation, and mutation operation are completed, the system automatically performs grid conflict detection. If the same grid is detected to be assigned to multiple types of items, the conflicting items are randomly reassigned to empty grids according to their set of possible fill positions until all individuals satisfy the grid mutual exclusion constraint; if the conflict cannot be eliminated after a preset number of reassignments, the individual is determined to be an invalid individual and a new individual is randomly selected from the orthogonal array to replace it.

[0067] Selection strategy switching: When the number of iterations exceeds 10 generations and the termination condition is still not met, the selection strategy is switched from hierarchical selection to roulette wheel selection method. The probability of being selected is allocated according to the proportion of each individual's fitness value to the total fitness value of the population. That is, the selection probability of each individual is equal to its fitness value divided by the sum of the fitness values ​​of all individuals in the population.

[0068] Termination Condition Determination: The preset termination conditions include any of the following: the number of solutions in the satisfactory solution set reaches 6, or the elderly user issues a termination iteration command through the interactive interface. Once the termination condition is met, the satisfactory solution set is output.

[0069] S5 Consistency Check and Decision-Making Steps Consistency checks and decision steps are used to identify the certainty of elderly users' preferences and generate the optimal storage partition outline scheme.

[0070] Utility value calculation: For each candidate storage area of ​​each type of item, all options for assigning the item to that area are selected from the set of satisfactory options. The fitness values ​​of these options are summed and then divided by the total number of options in the set of satisfactory options to obtain the utility value of the item in that area. A utility value greater than zero indicates that elderly users have a positive preference for placing the item in that area, and a higher utility value indicates a stronger preference.

[0071] Three-way comparison and preference identification: A three-way comparison is performed on the actual storage area information, the declarative preference selection, and the explicit preference reflected by the utility value. The area with a utility value greater than zero in the explicit preference is used as the candidate area. When the actual storage area of ​​a certain type of item is consistent with the declarative preference selection area, and the consistent area belongs to the candidate area, the storage area preference of the item is determined to be a deterministic preference; otherwise, it is determined to be an uncertain preference.

[0072] Optimal storage partition outline generation: For items with deterministic preferences, their original storage area settings remain unchanged; for items with uncertain preferences, the storage area with the highest utility value is selected as the recommended area. The optimal storage partition outline scheme is generated by combining the storage area determination results for various items.

[0073] This invention also provides a system for obtaining the storage and zoning needs of customized kitchen cabinets for the elderly, used to execute the above-described method. For example... Figure 2 As shown, the system includes a user decision-making layer and an enterprise decision-making layer.

[0074] The user decision-making process focuses on acquiring and optimizing the preferences of elderly users, with the optimization goal of highly satisfying individual preferences. This includes the following functional modules: The preference pre-collection module provides a user login interface, a personal basic information entry interface, an actual storage zoning survey interface, and a storage zoning inquiry test interface. The actual storage zoning survey interface allows elderly users to submit the actual storage locations of various items in their kitchen via image upload or option selection. The storage zoning inquiry test interface uses single-choice or multiple-choice formats to obtain elderly users' subjective preferences for ideal storage areas for various items. This module also includes a storage style dashboard module, which displays various kitchen item storage styles to elderly users in a carousel format with pictures and text.

[0075] Problem modeling module: This module includes built-in matrix grid layout rules for cabinet storage areas, fixed facility location information, a set of optional filling positions for various items, and binary chromosome coding rules. Problem modeling is automatically completed during system initialization. This module employs a 10-category detailed coding scheme and includes built-in mapping rules between categories 7 and 10.

[0076] Population initialization module: It has a built-in pre-generated orthogonal array. When the system starts, it automatically selects 25 individuals from the orthogonal array and randomly generates 5 individuals to form the initial population.

[0077] The interactive genetic evolution module provides a storage partition outline testing interface, displaying six storage partition outline diagrams to elderly users each round, with zoom functionality. This module includes a scoring unit, a selection operation unit, a crossover operation unit, a mutation operation unit, and an iteration control unit. When presenting the scheme to the user, this module automatically merges the 10 code categories into 7 categories. The interactive interface is developed using a high-level programming language and implemented based on a graphical user interface framework for visual interaction. In one specific embodiment, it is implemented using Python and the Tkinter graphics library.

[0078] Enterprise decision-making level; the enterprise decision-making level provides demand analysis and decision support for cabinet customization companies, with the optimization goal of facilitating the fulfillment of customization needs and balancing enterprise costs and profits, including the following functional modules: Consistency Verification Module: This module provides a customer decision-making information management interface, allowing enterprise users to view and manage basic information of elderly users, actual storage survey results, storage zoning inquiry results, and zoning profile test results. It automatically calculates the utility value of various items in each storage area and executes three-way comparison rules to identify deterministic and uncertain preferences.

[0079] The decision optimization module provides a decision interface for storage zoning needs. Based on consistency check results, it automatically generates storage zoning needs decision suggestions, including the degree of certainty of preferences for various items, recommended storage areas, and optimization suggestions. This module also considers the company's storage accessory reserves, production feasibility, and cost factors to generate final decision suggestions and the optimal storage zoning outline scheme.

[0080] The system adopts a client-server architecture. Elderly users access the user decision-making interface through mobile terminals or PC terminals, while enterprise users access the enterprise decision-making interface through the back-end management system. The system data is stored on a cloud server, supporting remote access to services across regions. Example 1

[0081] The following uses an elderly user (ID 10, female, 60 years old) as an example to illustrate the specific implementation process of the method of the present invention. The user interaction process adopts a 7-category merging scheme to reduce the cognitive burden of elderly users, and the internal coding calculation of the system adopts a 10-category subdivision scheme.

[0082] S1 Preference Information Pre-collection: Through in-home observation, the user's actual storage situation was obtained as follows: seasonings were stored in the cooking area on the countertop; pots and pans were stored in the cooking area on the wall; knives were stored in the food preparation area on the countertop; cutting boards were stored in the food preparation area on the countertop; dishcloths were hung in the washing area on the wall; dishes were stored in the washing area in the base cabinet; and chopsticks, forks, and spoons were stored in the washing area on the countertop. Through direct questioning, the user's stated preferences were obtained as follows: seasonings preferred to be stored on the countertop; pots and pans preferred to be stored on the wall; knives preferred to be stored on the wall; cutting boards preferred to be stored on the countertop; dishcloths preferred to be stored on the wall; dishes preferred to be stored in the base cabinet; and chopsticks, forks, and spoons preferred to be stored on the countertop.

[0083] S2 Problem Modeling and Coding: The cabinet storage areas are divided according to a 3x3 matrix layout. After removing the unfillable cells corresponding to the gas stove and sink, the set of fillable cell areas is determined. Based on the cell filling constraints and binary encoding rules, the user's storage scheme is encoded as a 46-bit chromosome. The system automatically maps the user's input of 7 categories of items to 10 categories of codes. Seasonings are further subdivided into small-volume and large-volume seasonings based on the user's actual storage situation. Pots and pans are subdivided into pots and pans, pot bodies, and pot lids.

[0084] S3 Initial population generation: Select 25 individuals from the pre-generated orthogonal array, and randomly generate 5 individuals to form an initial population of 30 individuals.

[0085] S4 Interactive Genetic Evolution: In the first round of evaluation, the user scored the 6 schemes 45, 60, 35, 70, 55 and 40 points respectively, with no satisfactory individuals, and all were sent to crossover and mutation operations; in the fifth round of evaluation, the first satisfactory individual (85 points) appeared; after the tenth round of evaluation, the selection was switched to roulette wheel selection; after the fourteenth round of evaluation, the set of satisfactory schemes accumulated to 6 schemes (scores of 85, 82, 88, 80, 86 and 83 points respectively), reaching the termination condition.

[0086] S5 Consistency Check and Decision: Calculate the utility value of each item in the satisfactory solution set for each area, and compare it with the actual storage area and declarative preference. Results show that the three results for knives and dishes are consistent, indicating a deterministic preference; the three results for cutting boards, dishcloths, and seasonings are not entirely consistent, indicating an uncertain preference. The final decision recommendation is: keep the existing areas for knife and dish storage unchanged; optimize cutting board and seasoning storage around the countertop and base cabinet areas; optimize dishcloth storage around the wall and countertop areas.

[0087] This embodiment is only a typical application scenario of the present invention. Those skilled in the art can adjust parameters such as the number of item categories, scoring threshold and number of iterations according to actual needs.

[0088] The above description is only a preferred embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A method for obtaining the storage and partitioning needs of customized kitchen cabinets for the elderly, characterized in that, Includes the following steps: S1 Preference Information Pre-collection Steps: Obtain information on the actual storage areas of various kitchen items in the current cabinets of elderly users, as well as the declarative preference choices of elderly users for the storage areas of various kitchen items; S2 Problem Modeling and Coding Steps: Divide the cabinet storage area into a matrix grid layout according to a multi-dimensional spatial partitioning method and number it. Determine the set of fillable grid areas based on the location of fixed facilities. Establish a set of optional filling positions for various items to be stored to form item grid filling constraint rules. Use binary encoding to encode the filling positions of various items to form a chromosome representation. S3 Initial population generation steps: An orthogonal array is generated using an orthogonal design method. Individuals are selected from the orthogonal array and combined with randomly generated individuals to form the initial population. S4 Interactive Genetic Evolution Steps: Decode each individual in the population into a partitioning outline scheme and present it to elderly users in a visual way. Obtain the subjective rating of the elderly users as the fitness value. Based on the fitness value, perform hierarchical selection and perform crossover and mutation operations on individuals to be optimized to generate a new generation of population. Iterate the above process until the preset termination condition is met and output a set of satisfactory schemes. S5 Consistency Check and Decision-Making Steps: Based on the fitness values ​​of each solution in the set of satisfactory solutions, the utility values ​​of various items in each storage area are statistically calculated. The actual storage area information, the declarative preference selection, and the explicit preference reflected by the utility values ​​are compared to identify deterministic preferences and uncertain preferences. Based on this, the optimal storage partition outline solution is generated.

2. The method according to claim 1, characterized in that, In step S2, the multi-dimensional space zoning method includes horizontal functional zoning and vertical space zoning. The horizontal functional zoning includes a cooking area, a food preparation area, and a washing area. The vertical space zoning includes a base cabinet area, a countertop area, and a wall area. The fixed facilities include a gas stove and a sink. The countertop and wall grids corresponding to the gas stove and the base cabinet and countertop grids corresponding to the sink are removed.

3. The method according to claim 1, characterized in that, In step S2, the corresponding number of binary encoding bits is used according to the number of optional filling positions for each type of item; in step S3, the ratio of the number of individuals selected from the orthogonal array to the number of randomly generated individuals is 4-6:1; in step S4, the first threshold for graded selection is 80 points and the second threshold is 20 points; the preset termination condition is that the number of solutions in the set of satisfactory solutions reaches the preset target number, or the elderly user issues a termination iteration command through the interactive interface.

4. The method according to claim 3, characterized in that, The hierarchical selection in step S4 specifically includes: storing individuals with fitness values ​​higher than the first threshold into the set of satisfactory solutions; eliminating individuals with fitness values ​​lower than the second threshold and supplementing new individuals from the orthogonal array; sending individuals with fitness values ​​between the two thresholds into the crossover and mutation operation; when the number of iterations exceeds the preset iteration number threshold, switching to the roulette wheel selection method, allocating the probability of selection of each individual according to the proportion of each individual's fitness value to the total fitness value of the population.

5. The method according to claim 1, characterized in that, The rule for identifying deterministic and uncertain preferences in step S5 is as follows: the region with a utility value greater than zero in the explicit preference is the candidate region. When the actual storage area of ​​the same item is consistent with the region selected by the declarative preference, and the consistent region belongs to the candidate region, it is determined to be a deterministic preference; otherwise, it is determined to be an uncertain preference. For items with deterministic preferences, the storage area setting of the consistent region is maintained. For items with uncertain preferences, the storage area with the highest utility value is selected as the optimal storage area.

6. The method according to claim 1, characterized in that, Step S1 also includes showing elderly users a storage style board to expand their understanding of storage methods; the visualization method described in step S4 is a rectangular plane graphic, with functional areas arranged horizontally and spatial areas arranged vertically, and various items to be stored are marked in the form of flat icons in the corresponding grid positions.

7. A system for obtaining storage zoning needs of customized kitchen cabinets for the elderly, used to execute the method of claim 1, characterized in that, This includes both user-level decision-making and corporate-level decision-making. The user decision-making layer includes: The preference pre-collection module is used to obtain information on the actual storage areas of various kitchen items in the current cabinets of elderly users, as well as the elderly users' declarative preference choices for the storage areas of various kitchen items; The problem modeling module is used to divide the cabinet storage area into a matrix grid layout, define the grid filling constraints for storage items, and perform chromosome coding on the filling positions of various items. The population initialization module is used to generate an orthogonal array and construct an initial population based on orthogonal design methods. The interactive genetic evolution module is used to decode the population individuals into a storage partition outline scheme for visualization, obtain subjective ratings from elderly users, perform hierarchical selection and crossover mutation operations, and control the iteration process to output a set of satisfactory schemes. The corporate decision-making level includes: The consistency check module is used to compare the actual storage area information, declarative preference selections, and explicit preferences reflected in the set of satisfactory solutions to identify deterministic and uncertain preferences. The decision optimization module is used to generate storage partitioning requirement decision suggestions and determine the optimal storage partitioning outline scheme based on the consistency test results.

8. The system according to claim 7, characterized in that, The preference pre-collection module also includes a storage style display board module, which is used to show elderly users a variety of kitchen item storage styles to expand their knowledge of storage forms; the interactive genetic evolution module visualizes the storage partition outline scheme as a rectangular planar graphic, in which the cooking area, food preparation area and washing area are arranged horizontally from left to right, and the base cabinet area, countertop area and wall area are arranged vertically from bottom to top.

9. The system according to claim 7, characterized in that, The interactive genetic evolution module includes a scoring acquisition unit, a selection operation unit, a crossover operation unit, a mutation operation unit, and an iteration control unit; the selection operation unit adopts a hierarchical selection mechanism and switches to a roulette wheel selection method after the number of iterations exceeds a preset iteration threshold.

10. The system according to claim 7, characterized in that, The consistency verification module uses a three-way comparison rule to determine candidate regions based on the region with a utility value greater than zero in the revealed preference. The decision optimization module combines the consistency verification results with the enterprise's production feasibility and cost factors to generate the final decision recommendation.

Images