CN114386879A - Grading and ranking method and system based on multi-product multi-dimensional performance indexes - Google Patents
Grading and ranking method and system based on multi-product multi-dimensional performance indexes Download PDFInfo
- Publication number
- CN114386879A CN114386879A CN202210282342.1A CN202210282342A CN114386879A CN 114386879 A CN114386879 A CN 114386879A CN 202210282342 A CN202210282342 A CN 202210282342A CN 114386879 A CN114386879 A CN 114386879A
- Authority
- CN
- China
- Prior art keywords
- index
- product
- quantization
- performance
- value
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
- G06Q10/06393—Score-carding, benchmarking or key performance indicator [KPI] analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/24—Classification techniques
- G06F18/243—Classification techniques relating to the number of classes
- G06F18/24323—Tree-organised classifiers
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Human Resources & Organizations (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Strategic Management (AREA)
- Entrepreneurship & Innovation (AREA)
- Economics (AREA)
- Educational Administration (AREA)
- Development Economics (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Biology (AREA)
- Game Theory and Decision Science (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Life Sciences & Earth Sciences (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Image Analysis (AREA)
- Complex Calculations (AREA)
Abstract
The invention discloses a grading and ranking method and system based on multi-product multi-dimensional performance indexes. The method comprises the following steps: s1, establishing a product classification system; designing an abstract product and abstract indexes, establishing an index quantization algorithm library, and storing an index quantization algorithm in the index quantization algorithm library; s2, instantiating a product with a plurality of dimension attributes from the abstract product; s3, instantiating a performance index with a plurality of dimension attributes from the abstract index; s4, establishing an index system of the product; s5, realizing the normalization of an index system; s6, reporting the brand product, and recording the product attribute value and the index value of the brand product; s7, quantifying the score; s8, ranking: repeatedly executing S6-S7, and realizing ranking based on the quantitative scores of different brand products; and S9, repeatedly executing S2-S8 to realize the quantitative score and ranking of the multi-product multi-dimensional performance indexes.
Description
Technical Field
The invention relates to the field of ranking and comprehensive ranking of products according to performance indexes, in particular to a ranking method and system suitable for multi-product multi-dimensional performance indexes.
Background
When people select products, the products can be selected according to indexes such as brands, prices, evaluations and the like of the products, more time, the products can be selected according to various performance indexes, and the quantitative ranking of the various performance indexes is very important. The existing ranking list can only be realized by adopting a uniform algorithm according to common indexes such as sales volume, price, comment quantity, favorable comment rate, access number, attention and the like, which are not individual indexes of commodities. For the performance indexes, different products have different performance indexes, the number of the performance indexes (index systems of the products) is different, the evaluation standards of various performance indexes of the same product are different, and how to realize quantitative scoring and ranking of various performance indexes of different products becomes the key for solving the problems. At present, the common practice is to design a specific program for a single type of product, and like ranking lists of automobiles according to various performance indexes, similar ranking lists with computers and mobile phones, the specific program is designed for the same type of product. If the types are as many as thousands, the workload of program development is enormous.
Therefore, a solution is needed to be provided, which solves the problem of large workload of development of the ranking list program based on the performance indexes of large comprehensive multi-product, and makes the scoring and ranking of the multi-dimensional personalized performance indexes of the multi-product possible. The key problems to be solved by the solution are how different products and different indexes are realized by the same solution, how to realize a quantization algorithm aiming at different index quantization requirements, and how to realize normalization aiming at different index systems, thereby calculating comprehensive scores and ranks.
Disclosure of Invention
In order to solve the technical problems, the invention provides a grading and ranking method and system based on multi-product multi-dimensional performance indexes.
In order to achieve the purpose, the invention adopts the technical scheme that: a grading and ranking method based on multi-product multi-dimensional performance indexes comprises the steps of collecting product information needing to be input into a system by defining an interactive program of a product, and forming an input interactive program of the product; an index system of the product is obtained by defining indexes for multiple times; and forming an input interactive program of the index and an index quantization algorithm based on the index value input requirement and the index key elements and associating the index algorithm. The method comprises the steps of inputting an interactive program and an index system quantization algorithm of a plurality of performance indexes of different types of products, collecting product information and index values through the formed product information and index system input interactive program, calculating all index quantization values of different manufacturers of the products through the index quantization algorithm, forming a ranking list of the products based on the performance indexes, and forming a comprehensive ranking list through the weight and normalization processing of each index. The method specifically comprises the following steps:
s1, establishing a product classification system: establishing a product classification tree and defining product classification;
designing abstract products and abstract indexes; the abstract product includes attributes describing all dimensions of the product for defining the product; the abstract index comprises attributes describing all dimensions of the index and is used for defining the performance index;
establishing an index quantization algorithm library, and storing an index quantization algorithm in the index quantization algorithm library;
s2, defining a product: selecting a product classification in a product classification tree, and instantiating a product with a plurality of dimension attributes from an abstract product;
s3, defining a performance index: instantiating a performance index with a plurality of dimension attributes from the abstract index for the instantiated product; defining a weight for the instantiated performance indicators; matching an index quantization algorithm for the performance index of the product, wherein the index quantization algorithm is used for quantizing the index value of the index into a specific score;
s4, completing the definition of all performance indexes of the product, thereby establishing an index system of the product;
s5, realizing the normalization of the index system, including judging whether the sum of the weights of each performance index in the index system is equal to 1:
P= 
=1
in the formula, P is the sum of the weights of an index system;
is the weight of the ith index;
s6, recording a product attribute value and a performance index attribute value of the brand product based on the brand product declaration; performance index attribute values, i.e., index values;
s7, quantification score: quantizing the input index value into scores of various performance indexes of the product according to an index quantization algorithm; calculating the comprehensive score of the product according to an index system and a normalization algorithm;
s8, ranking: when different brand products exist, repeatedly executing S6-S7, and realizing ranking based on the quantitative scores of the different brand products;
s9, aiming at different product classifications in the product classification tree, repeatedly executing S2-S8 to realize the quantitative score and ranking of the multi-product multi-dimensional performance indexes.
Further, defining a quantization rule in the index quantization algorithm library, wherein the quantization rule comprises quantization by calculating an expression and quantization by corresponding relation between option content and option value; the corresponding relation quantification of the option content and the option value comprises single selection, multi-selection accumulation and multi-selection and coefficient accumulation; the quantization rule corresponds to an index quantization algorithm; and matching an index quantization algorithm according to the quantization rule of the performance index when the performance index is instantiated through the abstract index.
Furthermore, sometimes the index system is not normalized together, for example, some products need to consider the performance indexes of the product components besides the performance indexes of the product itself, and some products need to consider the indexes of manufacturers, so the invention provides the index system normalization algorithm based on the tree structure.
The performance indicators include tier attributes;
the combination of more than two nth-level performance indexes is an n-1 level index group, wherein n is more than 1;
the performance index of the high level is prior to the performance index of the low level to carry out normalization processing, and the performance index of the same level and the index group carry out normalization processing together; the sum of the performance index and the weight of the index group at the same level is equal to 1.
Further, the method for matching the performance index of the product with the index quantization algorithm comprises the following steps: the quantization rule comprises a quantization key element; the index quantization algorithm library stores the corresponding relation between the index quantization algorithm and the quantization key elements;
guiding and inputting the attribute of the product performance index through a rule, collecting quantitative key elements in the attribute, and matching an index quantization algorithm;
and if the matching is not achieved, adding or modifying the corresponding relation between the index quantization algorithm and the quantization key elements in the index quantization algorithm library.
Further, recording the optimal value of the index value in the performance index needing to dynamically extract the optimal value; when brand product declaration is newly added, inquiring performance indexes of all dynamic optimal values before re-issuing a rating ranking list based on different brand products or a rating ranking list based on multi-product multi-dimensional performance indexes; and if the index value better than the recorded optimal value exists, updating the optimal value, and quantitatively scoring all the products by using the updated optimal value. The method improves the calculation efficiency by one order of magnitude, and can greatly reduce the load of the server in a large number of calculation stages before release.
And for the performance index which does not need to dynamically extract the optimal value, carrying out quantitative scoring when the index value of the brand product is recorded. Because the values are fixed, the scores are not influenced by the change of the filled data of other suppliers, and the score is calculated and processed during filling, so that the values are directly used for ranking according to rules during publishing, calculation is not needed, and the server load before the list is published can be effectively reduced. And in the product declaration phase, the calculation of force load cost is almost not needed.
Further, the method for quantizing the score according to the recorded index value comprises the following steps:
the quantization rules quantized by the computational expression include:
linear function quantization, for the index of the type, defining a qualified index value x1 and an optimal index value x2 through an interactive program, respectively corresponding to quantized values y1=60 and y2=100, and establishing a linear function y = ax + b through (x 1, y 1) (x2, y 2); calculating quantization parameters a and b of a linear function through two points (x 1, y 1) (x2, y2) in a rectangular coordinate system, recording the quantization parameters a and b into a quantization table of the index, taking out the quantization parameters a and b from the quantization table according to a performance index attribute value x recorded in product declaration when the index is quantized, and calculating a quantization value y = ax + b of the declaration index value x;
the quantization rule quantized through the corresponding relation between the option content and the option value comprises the following steps:
and (3) an algorithm based on the contents and corresponding values of the single selection options: the quantization algorithm of the corresponding index is that the quantization value y = viV is the corresponding value of the selected content, and i is the selected serial number;
and (3) an algorithm based on multi-selection accumulation option content and corresponding values: selecting a content corresponding value v, selecting the maximum value max of index quantization score, and using a corresponding index quantization algorithm to obtain a quantization value y = min (sigma viMax), v is the corresponding value of the selected content, i is the selected sequence number, and min () takes the small value;
and (3) an algorithm for accumulating option contents and corresponding values based on the multiple-option coefficients: selecting the content quantity c, selecting the corresponding value v of the content, and the maximum value max of the index quantization score, wherein the corresponding index quantization algorithm is the quantization value y = min (Sigma c)iviMax), where v is the quantization value corresponding to the selected content, c is the number of the selected content, i is the selected sequence number, and min () takes a small value thereof.
Furthermore, the performance indexes of the index system are analyzed, the intervals of the performance index attributes are judged, the performance indexes are matched with an index quantization algorithm in different areas, and the input index values are quantized. Namely: a plurality of intervals may be defined, each interval corresponding to a different metric quantization requirement. The method can adapt to the situation that quantization is carried out in different regions through function expressions in actual demands, different regions have different expressions, and when the index algorithm is defined, the regions are defined and respectively correspond to different algorithms in a quantization algorithm library.
Further, an index template is developed, the established performance index is stored as the index template in the process of establishing the performance index, and when the performance index is established, the index template is called to accelerate the generation of a new performance index;
and developing an index system template, storing the established index system as the index system template in the process of establishing the index system, and accelerating the generation of a new index system by calling the index system template when establishing the index system.
When a similar product performance index or an index system is established, the template is called, a new index system can be generated only by little change, and the implementation workload can be saved by more than 50%.
On the other hand, the invention also discloses a grading and ranking system based on the multi-product multi-dimensional performance indexes, which comprises the following steps:
the abstract product is designed, describes the attributes of all dimensions of the product and is used for defining the product class which really exists; establishing a product classification tree, and defining product classification by using a tree structure;
abstract indexes, which are abstract indexes designed, describe the attributes of each dimension of the indexes and are used for defining the performance indexes of actual products;
the index quantization algorithm library stores index quantization algorithms and quantization rules, and the quantization rules correspond to the index quantization algorithms; matching an index quantization algorithm according to a quantization rule of the performance index when the performance index is instantiated through the abstract index;
defining a product module: selecting a product classification in a product classification tree, and instantiating a product with a plurality of dimension attributes from an abstract product;
a define performance indicators module: instantiating a performance index with a plurality of dimension attributes from the abstract index for the instantiated product; matching an index quantization algorithm for the performance index of the product, wherein the index quantization algorithm is used for quantizing the index value of the index into a specific score;
an index algorithm matching module: when the performance indexes are instantiated, matching the performance indexes of the product with the index quantization algorithm through the corresponding relation between the quantization rules and the index quantization algorithm;
a product declaration module: recording a product attribute value and a performance index attribute value of a brand product;
a quantization scoring module: quantizing the input index value into scores of various performance indexes of the product according to an index quantization algorithm; calculating the comprehensive score of the product according to an index system and a normalization algorithm;
the ranking list publishing module: and forming and publishing a ranking list based on the quantitative scores of different brand products and/or the quantitative scores of the multi-product multi-dimensional performance indexes.
Further, the quantization rule includes a quantization key element; the index quantization algorithm library stores the corresponding relation between the index quantization algorithm and the quantization key elements;
the index algorithm matching module collects quantitative key elements in the attribute of the input product performance index and matches an index quantitative algorithm;
and if the matching is not achieved, the index quantization algorithm library adds or modifies the corresponding relation between the index quantization algorithm and the quantization key elements.
The system further comprises an index template module, wherein the established performance indexes are stored as index templates in the process of establishing the performance indexes, and when the performance indexes are established, the index templates are called to accelerate the generation of new performance indexes;
index system template module: in the process of establishing the index system, the established index system is stored as a template, and when the index system is newly established, the generation of a new index system is accelerated by calling the template of the index system.
The invention has the following beneficial effects:
the algorithm and the application of the invention can be suitable for the ranking of any product based on the performance index without the realization of personalized programming. The system can adapt to the scoring and ranking of any type of products according to different performance indexes and comprehensive scoring and ranking, can greatly improve the development and maintenance efficiency of ranking lists according to the performance indexes of products, and is particularly suitable for large comprehensive product ranking list systems based on the performance indexes, such as large building material ranking list systems based on the performance indexes, daily department goods ranking list systems based on the performance indexes, commodity ranking selection systems of large e-commerce platforms based on the performance indexes, and the like;
inputting a class of products, generating a data structure expression required by a class of management program, and generating a management function related to the class of products and a performance index system through the data structure expression, so that the development of the management program aiming at different products and index systems is converted into the input of product attributes and the index systems thereof, the efficiency of establishing the management function of the index system is greatly improved, and a huge management system of a plurality of products and index systems can be established in a short time;
with the increase of the requirement on the level of designers, the requirement on personnel for establishing the index system is relatively reduced, and ordinary personnel can participate in the establishment of the index system through training without professional IT personnel. The cost is greatly reduced;
the index system is maintained, maintenance programs are not needed, the maintenance efficiency is improved, and the maintenance cost is reduced.
Drawings
Fig. 1 is a flowchart of a scoring and ranking method based on multi-product multi-dimensional performance indexes according to an embodiment of the present invention.
Fig. 2 is a schematic diagram of an inter-partition matching index quantization algorithm of a scoring and ranking method based on multi-product multi-dimensional performance indexes according to an embodiment of the present invention.
Fig. 3 is a schematic diagram of performance index level attributes of a multi-product multi-dimensional performance index-based scoring and ranking method according to an embodiment of the present invention.
Detailed Description
In order to facilitate understanding of those skilled in the art, the present invention will be further described with reference to the following embodiments and accompanying drawings.
The scoring and ranking method based on multi-product and multi-dimensional performance indexes of the embodiment is shown in fig. 1, and includes the following contents:
(1) defining product classifications
The tree structure is used for defining product classification, a classification system of the product is established, and a product classification tree can be established and maintained according to needs through an interactive program, so that the system can dynamically manage the needed product classes.
(2) Design of abstract product
Abstract products are designed to define the product classes that exist in reality. The abstract product is designed to describe the attributes of each dimension of the product, and the possible product attributes are fully considered and are as comprehensive as possible. For example, the dimensions of the following product attributes can be designed: product name, place of production, manufacturer, brand, specification and model, picture, three-dimensional model, index system and the like.
(3) Design of abstract index
And designing abstract indexes to define the performance indexes of the actual product. The abstract index is the attribute of each dimension of the design description index, and the possible index attributes are fully considered and are as comprehensive as possible. For example, the following dimensions for the index attribute may be set: index name, index quantization algorithm requirement, index quantization interval (a plurality of intervals can be defined, each interval corresponds to different index quantization requirements), index sum value, index optimal value, index unit, index national standard, industry standard, index calculation type, weight, detection standard and method, evidence-based material requirement, declaration and the like.
(4) Defining a product
Selecting a certain product from a product classification tree, changing an abstract product into a certain actual product through program interaction, for example, selecting a ceramic tile, namely a glazed tile, namely a full-polished tile, in the product classification tree, namely 600X600, obtaining the glazed full-polished tile of the specific product 600X600, then selecting related attributes required by the product according to the attributes of the abstract product, for example, defining the attributes of the 600X600 glazed full-polished ceramic tile, such as production place, production manufacturer, brand, specification model, picture and the like, and establishing an index system through the following algorithm of 5-11, thereby defining a specific 600X600 glazed full-polished ceramic tile product. Thereby defining the relevant data to be entered for the manufacturer to declare the product.
And (3) adding product classification in the product classification tree, defining abstract products into concrete products in the product classification tree through the step (2), and repeatedly operating in such a way, thus infinitely building products needing to be managed.
(5) Defining specific performance indicators
And (3) defining a concrete performance index through the abstract index defined in the step (3), for example, the surface flatness deviation of the 600X600 glazed full-polishing ceramic tile, and defining relevant attributes for the performance index.
Thus, a specific performance index is defined: deviation of surface flatness of the tile.
(6) Defining an index system
Repeating the step (5), defining all performance indexes of the product, and establishing an index system of the product;
for example, the following table establishes an index system for a 600X600 glazed fully polished ceramic tile by repeating (5) multiple times.
(7) Index quantification algorithm library capable of being dynamically managed
An index quantization algorithm library is designed, namely, the algorithm library for quantizing the index value into specific scores is dynamically manageable, and the index quantization algorithm can be added and maintained through an interactive program.
Defining an algorithm rule name in an algorithm library, defining a quantization rule, comprising the steps of quantizing through a calculation expression, quantizing through the corresponding relation between option content and option value and the like, defining algorithms such as single selection, multiple selection accumulation, multiple selection and coefficient accumulation and the like in the quantizing of the corresponding relation between the option content and the option value, leading a user to select different quantization rules through rule guidance when a performance index is input, associating a dynamic quantization algorithm library, extracting quantization parameters needing to be input, for example, if a certain quantization interval is selected, quantizing through the calculation expression, associating the dynamic quantization algorithm library by a program, taking out the existing expression type for the user to select, for example, further selecting a linear expression (y = ax + b), interacting the program and requiring to input the performance index values (key elements) of two points corresponding to a grid value and an optimal value, calculating the related parameters (a, b) and recording the performance index in a quantization table of the interval, and extracting quantization parameters (a, b) according to the performance index declaration value x when the actual product is declared, so that the quantization value y = ax + b can be easily obtained. If the optimal value of the actual declared value changes, (a, b) is recalculated and the quantization table is updated. And if the dynamic quantization algorithm library has no related expression, adding the related expression and the related quantization key element through an interactive program. Other quantization rules are similar to the quantization algorithm.
Dynamic quantization algorithm library content:
(8) analysis of performance index, correlation quantization algorithm
When a concrete performance index is defined through an abstract index, the system automatically matches a related quantization algorithm according to a selected quantization rule and a quantization name, if the matching is not achieved, a dynamic quantization algorithm library is maintained through an interactive program, and the algorithm is increased or quantization key elements related to the quantization algorithm are modified.
(9) Index quantization algorithm
Corresponding the index value to 0-100 through an index quantization algorithm in an index quantization algorithm library;
the quantization is carried out through a function expression, most of the function expression is divided into regions, different regions have different expressions, and when an index quantization algorithm is defined, the regions are defined and respectively correspond to different algorithms in a quantization algorithm library.
For example, FIG. 2:
the absolute value of the index value is greater than 10, the quantization is 0, 10< = index value <0, and the quantization is 60;
0< = index value < =10, linear increase, and when the index value reaches 10, quantization is 100.
(10) Normalization algorithm of index system
And realizing the normalization of the index system through the quantization scores of the performance indexes and the weights of the performance indexes. And judging whether the weight sum is equal to 1 in an interactive program of a defined index system, and preventing weight errors;
P=
=1；
in the formula, P is the sum of the weights of an index system;
is the weight of a certain performance index.
(11) Tree structure based index system normalization algorithm
Sometimes, not all indexes of the index system are normalized together, for example, some products need to consider the performance indexes of product components besides the performance indexes of the products themselves, and some products need to consider the indexes of manufacturers, and a tree structure-based normalization algorithm is adopted to solve the requirement, as shown in fig. 3.
The defined performance indexes comprise first-level indexes and second-level indexes, more than two second-level indexes are combined into an index group according to a quantization algorithm, and normalization processing is performed firstly; then, the index group and the first-level index are normalized again to obtain the score of the index system. The sum of the weights of the secondary indexes of the same index group is equal to 1; the sum of the weights of the index group and the primary index is equal to 1. The index 1 and the index 2 are used as an index group 1 to be normalized to obtain the score of the index group 1, and the index 3, the index 4 and the index 5 are used as an index group 2 to be normalized to obtain the score of the index group 2. And normalizing the index group 1, the index group 2, the index 6 and the index 7 to obtain the value of the index system of the product.
(12) Inputting product information and index information
Through the operation, a specific product and a product index system are defined, and the related information of a brand of the product can be input through an interactive program according to the defined attribute.
The following is a glazed fully polished tile according to definition 600X600, a brand of product entered and its index system.
The product name is as follows: ceramic tile glazed tile YG66188
Product category: 600*600
Brand name: a certain
The manufacturer: hangzhou ceramics Ltd
Photo:
three-dimensional model:
index system:
(13) index quantization and weighted normalization
Calculating the score of each index of the brand product according to the defined performance index, the related quantitative algorithm and the declared index value; and calculating the comprehensive score of the brand product according to a defined index system and a normalization algorithm.
The following are the quantized values of a system of certain brand indexes of the glazed fully-polished tiles of 600X600 and the overall scores after the normalization treatment after weighting.
(14) Product of one brand forming a certain type of product and quantified performance index
And forming all information of one brand of a certain product by the input basic information of the certain product, the quantitative information of the brand index system formed in the step and the comprehensive score after the weighting normalization processing.
(15) Ranking list and comprehensive ranking list based on product performance indexes and forming multiple brands of certain products
And (5) repeating the steps (12) to (14) to form index scores, comprehensive scores and ranking lists of different brands of different manufacturers of the products.
(16) System for forming ranking list of multiple products
And (4) adding new products in the product classification tree, repeating the steps (4) to (15), and forming the ranking list and the comprehensive ranking list of the various products based on the performance indexes.
The embodiment also discloses a scoring and ranking system based on multi-product multi-dimensional performance indexes, which comprises:
the abstract product is designed, describes the attributes of all dimensions of the product and is used for defining the product class which really exists; establishing a product classification tree, and defining product classification by using a tree structure;
abstract indexes, which are abstract indexes designed, describe the attributes of each dimension of the indexes and are used for defining the actual performance indexes of actual products;
the index quantization algorithm library stores index quantization algorithms and quantization rules, and the quantization rules correspond to the index quantization algorithms; matching an index quantization algorithm according to a quantization rule of the performance index when the performance index is instantiated through the abstract index; the quantization rule comprises a quantization key element; the index quantization algorithm library stores the corresponding relation between the index quantization algorithm and the quantization key elements;
defining a product module: selecting a product classification in a product classification tree, and instantiating a product with a plurality of dimension attributes from an abstract product;
a define performance indicators module: instantiating a performance index with a plurality of dimension attributes from the abstract index for the instantiated product; matching an index quantization algorithm for the performance index of the product, wherein the index quantization algorithm is used for quantizing the index value of the index into a specific score;
an index algorithm matching module: when the performance indexes are instantiated, the index algorithm matching module collects quantitative key elements in the recorded attributes of the performance indexes of the products and matches an index quantitative algorithm; if the matching is not achieved, the index quantization algorithm library adds or modifies the corresponding relation between the index quantization algorithm and the quantization key elements;
a product declaration module: recording a product attribute value and a performance index attribute value of a brand product;
a quantization scoring module: quantizing the input index value into scores of various performance indexes of the product according to an index quantization algorithm; calculating the comprehensive score of the product according to an index system and a normalization algorithm;
the ranking list publishing module: and forming and publishing a ranking list based on the quantitative scores of different brand products and/or the quantitative scores of the multi-product multi-dimensional performance indexes.
The classified and graded ranking list management of the building material equipment tested by the method provided by the embodiment comprises over 3000 material equipment, and by means of bringing 3000 material equipment into management, the template function of the test index system can save about 50% of implementation workload, and greatly shorten project implementation time.
The above embodiments are only for illustrating the technical idea of the present invention, and the protection scope of the present invention is not limited thereby, and any modification made on the basis of the technical solution according to the technical idea of the present invention falls within the protection scope of the present invention.
Claims (10)
1. A scoring and ranking method based on multi-product and multi-dimensional performance indexes is characterized by comprising the following steps:
s1, establishing a product classification system: establishing a product classification tree and defining product classification;
designing abstract products and abstract indexes; the abstract product includes attributes describing all dimensions of the product for defining the product; the abstract index comprises attributes describing all dimensions of the index and is used for defining the performance index;
establishing an index quantization algorithm library, and storing an index quantization algorithm in the index quantization algorithm library;
s2, defining a product: selecting a product classification in a product classification tree, and instantiating a product with a plurality of dimension attributes from an abstract product;
s3, defining a performance index: instantiating a performance index with a plurality of dimension attributes from the abstract index for the instantiated product; defining a weight for the instantiated performance indicators; matching an index quantization algorithm for the performance index of the product, wherein the index quantization algorithm is used for quantizing the index value of the index into a specific score;
s4, completing the definition of all performance indexes of the product, thereby establishing an index system of the product;
s5, realizing the normalization of the index system, including judging whether the sum of the weights of each performance index in the index system is equal to 1;
s6, recording a product attribute value and a performance index attribute value of the brand product based on the brand product declaration; performance index attribute values, i.e., index values;
s7, quantification score: quantizing the input index value into scores of various performance indexes of the product according to an index quantization algorithm; calculating the comprehensive score of the product according to an index system and a normalization algorithm;
s8, ranking: when different brand products exist, repeatedly executing S6-S7, and realizing ranking based on the quantitative scores of the different brand products;
s9, aiming at different product classifications in the product classification tree, repeatedly executing S2-S8 to realize the quantitative score and ranking of the multi-product multi-dimensional performance indexes.
2. The multi-product multi-dimensional performance index-based scoring and ranking method of claim 1, wherein:
defining quantization rules in the index quantization algorithm library, wherein the quantization rules comprise quantization by calculating expressions and quantization by corresponding relations between option contents and option values; the corresponding relation quantification of the option content and the option value comprises single selection, multi-selection accumulation and multi-selection and coefficient accumulation;
the quantization rule corresponds to an index quantization algorithm; and matching an index quantization algorithm according to the quantization rule of the performance index when the performance index is instantiated through the abstract index.
3. The multi-product multi-dimensional performance index-based scoring and ranking method of claim 1 or 2, wherein:
the performance indicators include tier attributes;
the combination of more than two nth-level performance indexes is an n-1 level index group, wherein n is more than 1;
the performance index of the high level is prior to the performance index of the low level to carry out normalization processing, and the performance index of the same level and the index group carry out normalization processing together; the sum of the performance index and the weight of the index group at the same level is equal to 1.
4. The method for scoring and ranking based on multi-product multi-dimensional performance indicators according to claim 1 or 2, wherein the method of matching the indicator quantification algorithm for the performance indicators of the product is:
the quantization rule comprises a quantization key element; the index quantization algorithm library stores the corresponding relation between the index quantization algorithm and the quantization key elements;
guiding and inputting the attribute of the product performance index through a rule, collecting quantitative key elements in the attribute, and matching an index quantization algorithm;
and if the matching is not achieved, adding or modifying the corresponding relation between the index quantization algorithm and the quantization key elements in the index quantization algorithm library.
5. The multi-product multi-dimensional performance index-based scoring and ranking method of claim 4, wherein:
recording the optimal value of the index value in the performance index needing to dynamically extract the optimal value;
when brand product declaration is newly added, inquiring performance indexes of all dynamic optimal values before re-issuing a rating ranking list based on different brand products or a rating ranking list based on multi-product multi-dimensional performance indexes; if the index value better than the recorded optimal value exists, updating the optimal value, and quantitatively scoring all the products by using the updated optimal value;
and for the performance index which does not need to dynamically extract the optimal value, carrying out quantitative scoring when the index value of the brand product is recorded.
6. The multi-product multi-dimensional performance index-based scoring and ranking method of claim 5, wherein the method for quantifying the score according to the entered index value is:
a quantization rule quantized by a computational expression, comprising:
linear function quantization, for the index of the type, defining a qualified index value x1 and an optimal index value x2 through an interactive program, respectively corresponding to quantized values y1=60 and y2=100, and establishing a linear function y = ax + b through (x 1, y 1) (x2, y 2); calculating quantization parameters a and b of a linear function through two points (x 1, y 1) (x2, y2) in a rectangular coordinate system, recording the quantization parameters a and b into a quantization table of the index, taking out the quantization parameters a and b from the quantization table according to a performance index attribute value x recorded in product declaration when the index is quantized, and calculating a quantization value y = ax + b of the declaration index value x;
the quantization rule quantized through the corresponding relation between the option content and the option value comprises the following steps:
and (3) an algorithm based on the contents and corresponding values of the single selection options: the quantization algorithm of the corresponding index is that the quantization value y = viV is the corresponding value of the selected content, and i is the selected serial number;
and (3) an algorithm based on multi-selection accumulation option content and corresponding values: selecting a content corresponding value v, selecting the maximum value max of index quantization score, and using a corresponding index quantization algorithm to obtain a quantization value y = min (sigma viMax), v is the corresponding value of the selected content, i is the selected sequence number, and min () takes the small value;
and (3) an algorithm for accumulating option contents and corresponding values based on the multiple-option coefficients: selecting the content quantity c, selecting the corresponding value v of the content, and the maximum value max of the index quantization score, wherein the corresponding index quantization algorithm is the quantization value y = min (Sigma c)iviMax), where v is the quantization value corresponding to the selected content, c is the number of the selected content, i is the selected sequence number, and min () takes a small value thereof.
7. The multi-product multi-dimensional performance index-based scoring and ranking method of claim 4, wherein: and analyzing the performance indexes of the index system, judging the intervals of the performance index attributes, matching the performance indexes with an index quantization algorithm in different areas, and quantizing the input index values.
8. The multi-product multi-dimensional performance index-based scoring and ranking method of claim 1, wherein:
developing an index template, storing the established performance index as the index template in the process of establishing the performance index, and accelerating the generation of a new performance index by calling the index template when the performance index is newly established;
and developing an index system template, storing the established index system as the index system template in the process of establishing the index system, and accelerating the generation of a new index system by calling the index system template when establishing the index system.
9. A scoring and ranking system based on multi-product multi-dimensional performance indexes is characterized by comprising:
the abstract product is designed, describes the attributes of all dimensions of the product and is used for defining the product class which really exists; establishing a product classification tree, and defining product classification by using a tree structure;
abstract indexes, which are abstract indexes designed, describe the attributes of each dimension of the indexes and are used for defining the performance indexes of actual products;
the index quantization algorithm library stores index quantization algorithms and quantization rules, and the quantization rules correspond to the index quantization algorithms; matching an index quantization algorithm according to a quantization rule of the performance index when the performance index is instantiated through the abstract index;
defining a product module: selecting a product classification in a product classification tree, and instantiating a product with a plurality of dimension attributes from an abstract product;
a define performance indicators module: instantiating a performance index with a plurality of dimension attributes from the abstract index for the instantiated product; matching an index quantization algorithm for the performance index of the product, wherein the index quantization algorithm is used for quantizing the index value of the index into a specific score;
an index algorithm matching module: when the performance indexes are instantiated, matching the performance indexes of the product with the index quantization algorithm through the corresponding relation between the quantization rules and the index quantization algorithm;
a product declaration module: recording a product attribute value and a performance index attribute value of a brand product;
a quantization scoring module: quantizing the input index value into scores of various performance indexes of the product according to an index quantization algorithm; calculating the comprehensive score of the product according to an index system and a normalization algorithm;
the ranking list publishing module: and forming and publishing a ranking list based on the quantitative scores of different brand products and/or the quantitative scores of the multi-product multi-dimensional performance indexes.
10. The multi-product multi-dimensional performance indicator-based scoring and ranking system of claim 9 wherein:
the quantization rule comprises a quantization key element; the index quantization algorithm library stores the corresponding relation between the index quantization algorithm and the quantization key elements;
the index algorithm matching module collects quantitative key elements in the attribute of the input product performance index and matches an index quantitative algorithm; and if the matching is not achieved, the index quantization algorithm library adds or modifies the corresponding relation between the index quantization algorithm and the quantization key elements.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210282342.1A CN114386879B (en) | 2022-03-22 | 2022-03-22 | Grading and ranking method and system based on multi-product multi-dimensional performance indexes |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210282342.1A CN114386879B (en) | 2022-03-22 | 2022-03-22 | Grading and ranking method and system based on multi-product multi-dimensional performance indexes |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN114386879A true CN114386879A (en) | 2022-04-22 |
| CN114386879B CN114386879B (en) | 2022-07-22 |
Family
ID=81205365
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210282342.1A Active CN114386879B (en) | 2022-03-22 | 2022-03-22 | Grading and ranking method and system based on multi-product multi-dimensional performance indexes |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114386879B (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115910381A (en) * | 2022-11-17 | 2023-04-04 | 天津大学温州安全(应急)研究院 | Normative determination method and system for cardio-pulmonary resuscitation operation flow |
| CN116523497A (en) * | 2023-04-07 | 2023-08-01 | 国网安徽省电力有限公司滁州供电公司 | PMS-based integrating worker Shan Yunwei cost quantification system |
Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103136227A (en) * | 2011-11-25 | 2013-06-05 | 阿里巴巴集团控股有限公司 | Method and device capable of generating list |
| US20140122494A1 (en) * | 2012-10-31 | 2014-05-01 | Mapmyfitness, Inc. | System and method for personal and peer performance ranking of outdoor activities |
| CN104077693A (en) * | 2013-03-27 | 2014-10-01 | 腾讯科技（深圳）有限公司 | Commodity comparison method, server, client side and e-commerce system |
| CN106485554A (en) * | 2016-10-18 | 2017-03-08 | 希音（深圳）投资有限公司 | A kind of musical instruments chooses system |
| CN107016570A (en) * | 2017-03-30 | 2017-08-04 | 北京汽车股份有限公司 | Product methods of marking and device based on voice of customers VOC |
| CN107578161A (en) * | 2017-08-30 | 2018-01-12 | 咪咕文化科技有限公司 | Product evaluation method and device and storage medium |
| CN108491377A (en) * | 2018-03-06 | 2018-09-04 | 中国计量大学 | A kind of electric business product comprehensive score method based on multi-dimension information fusion |
| CN108934026A (en) * | 2018-06-30 | 2018-12-04 | 江苏有线数据网络有限责任公司 | A kind of mobile terminal Wi-Fi quality determining method |
| CN109800989A (en) * | 2019-01-21 | 2019-05-24 | 石家庄职业技术学院（石家庄广播电视大学） | Using the supplier selection method of evaluation index classification overall merit |
| CN112667969A (en) * | 2020-12-24 | 2021-04-16 | 一汽奔腾轿车有限公司 | Vehicle commodity score calculation method |
| CN112732555A (en) * | 2020-12-29 | 2021-04-30 | 深圳市磐锋精密技术有限公司 | Product performance testing system and method based on internet big data |
| CN113836381A (en) * | 2021-09-30 | 2021-12-24 | 广东南方信息安全研究院 | System scoring coverage degree tuning method |
-
2022
- 2022-03-22 CN CN202210282342.1A patent/CN114386879B/en active Active
Patent Citations (12)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103136227A (en) * | 2011-11-25 | 2013-06-05 | 阿里巴巴集团控股有限公司 | Method and device capable of generating list |
| US20140122494A1 (en) * | 2012-10-31 | 2014-05-01 | Mapmyfitness, Inc. | System and method for personal and peer performance ranking of outdoor activities |
| CN104077693A (en) * | 2013-03-27 | 2014-10-01 | 腾讯科技（深圳）有限公司 | Commodity comparison method, server, client side and e-commerce system |
| CN106485554A (en) * | 2016-10-18 | 2017-03-08 | 希音（深圳）投资有限公司 | A kind of musical instruments chooses system |
| CN107016570A (en) * | 2017-03-30 | 2017-08-04 | 北京汽车股份有限公司 | Product methods of marking and device based on voice of customers VOC |
| CN107578161A (en) * | 2017-08-30 | 2018-01-12 | 咪咕文化科技有限公司 | Product evaluation method and device and storage medium |
| CN108491377A (en) * | 2018-03-06 | 2018-09-04 | 中国计量大学 | A kind of electric business product comprehensive score method based on multi-dimension information fusion |
| CN108934026A (en) * | 2018-06-30 | 2018-12-04 | 江苏有线数据网络有限责任公司 | A kind of mobile terminal Wi-Fi quality determining method |
| CN109800989A (en) * | 2019-01-21 | 2019-05-24 | 石家庄职业技术学院（石家庄广播电视大学） | Using the supplier selection method of evaluation index classification overall merit |
| CN112667969A (en) * | 2020-12-24 | 2021-04-16 | 一汽奔腾轿车有限公司 | Vehicle commodity score calculation method |
| CN112732555A (en) * | 2020-12-29 | 2021-04-30 | 深圳市磐锋精密技术有限公司 | Product performance testing system and method based on internet big data |
| CN113836381A (en) * | 2021-09-30 | 2021-12-24 | 广东南方信息安全研究院 | System scoring coverage degree tuning method |
Non-Patent Citations (3)
| Title |
|---|
| 唐沰: "手机阅读应用产品评价指标体系构建及应用", 《中国优秀博硕士学位论文全文数据库(硕士)信息科技辑》 * |
| 唐沰: "手机阅读应用产品评价指标体系构建及应用", 《中国优秀博硕士学位论文全文数据库(硕士)信息科技辑》, no. 12, 15 December 2013 (2013-12-15), pages 141 - 3 * |
| 覃燕红等: "《模块化产品智能配置理论与方法》", 31 January 2011, 成都：西南交通大学出版社, pages: 50 - 51 * |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN115910381A (en) * | 2022-11-17 | 2023-04-04 | 天津大学温州安全(应急)研究院 | Normative determination method and system for cardio-pulmonary resuscitation operation flow |
| CN116523497A (en) * | 2023-04-07 | 2023-08-01 | 国网安徽省电力有限公司滁州供电公司 | PMS-based integrating worker Shan Yunwei cost quantification system |
| CN116523497B (en) * | 2023-04-07 | 2023-11-17 | 国网安徽省电力有限公司滁州供电公司 | PMS-based integrating worker Shan Yunwei cost quantification system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN114386879B (en) | 2022-07-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US9892187B2 (en) | Data analysis method, data analysis device, and storage medium storing processing program for same | |
| CN114386879B (en) | Grading and ranking method and system based on multi-product multi-dimensional performance indexes | |
| Branke et al. | Hyper-heuristic evolution of dispatching rules: A comparison of rule representations | |
| CN107220217A (en) | Characteristic coefficient training method and device that logic-based is returned | |
| Achlioptas et al. | Fast sampling of perfectly uniform satisfying assignments | |
| US11748452B2 (en) | Method for data processing by performing different non-linear combination processing | |
| CN107424007A (en) | A kind of method and apparatus for building electronic ticket susceptibility identification model | |
| US11182841B2 (en) | Prospect recommendation | |
| CN112016967B (en) | Transaction data processing method and device | |
| CN111966886A (en) | Object recommendation method, object recommendation device, electronic equipment and storage medium | |
| CN113554307B (en) | RFM model-based user grouping method, device and readable medium | |
| JP6242540B1 (en) | Data conversion system and data conversion method | |
| US20070198252A1 (en) | Optimum design management apparatus, optimum design calculation system, optimum design management method, and optimum design management program | |
| CN114741392A (en) | Data query method and device, electronic equipment and storage medium | |
| CN113283795A (en) | Data processing method and device based on two-classification model, medium and equipment | |
| CN116562588A (en) | Enterprise supply chain analysis system, method and equipment based on ERP | |
| Wu et al. | A hierarchical parallel multi-station assembly sequence planning method based on GA-DFLA | |
| CN113763031A (en) | Commodity recommendation method and device, electronic equipment and storage medium | |
| JP2005032079A (en) | Project pre-evaluation method | |
| CN115049446A (en) | Merchant identification method and device, electronic equipment and computer readable medium | |
| Behdani et al. | An alternative approach to rank efficient DMUs in DEA via cross-efficiency evaluation, Gini coefficient, and Bonferroni mean | |
| CN111324594A (en) | Data fusion method, device, equipment and storage medium for grain processing industry | |
| CN116431498A (en) | Performance test method and device, electronic equipment and computer readable storage medium | |
| Goossens et al. | From OCEL to DOCEL–Datasets and Automated Transformation | |
| de Moraes et al. | An Efficiency Frontier based Model for Cloud Computing Provider Selection and Ranking. |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |