A method and system for reviewing technical parameters in bidding for medical device products.

Abstract

This application relates to the field of bidding review technology, and in particular to a method and system for reviewing technical parameters of medical device products in bidding. It includes: obtaining bidding documents and all tender documents; setting multiple technical sub-components based on the preprocessing results of the bidding documents, and generating abnormal parameter values ​​for each technical sub-component based on a preset anomaly assessment model; setting a review strategy based on a preset correlation review model and all abnormal parameter values, and generating review results based on the review strategy; and determining whether any technical sub-components are abnormal by screening the bidding documents, segmenting them into multiple technical sub-components, initially screening each technical sub-component based on the constructed anomaly assessment model, and constructing a comparison model based on the screening results to screen all tender documents, thereby improving the efficiency of reviewing the compliance of technical parameters of medical device products.

Description

Technical Field

[0001] This application relates to the field of tender review technology, and in particular to a method and system for reviewing the technical parameters of a medical device product tender. Background Technology

[0002] In the government procurement and hospital bidding processes for medical equipment, the review of technical parameters is a core step in ensuring the quality, performance, and suitability of the procured products. Currently, the review of technical parameters for medical devices in bidding documents mainly relies on manual review.

[0003] Existing manual review methods have several drawbacks. First, they heavily rely on the individual knowledge and experience of review experts, making them highly subjective and difficult to guarantee the consistency and objectivity of review standards. Different experts may have differing interpretations of the same technical clauses, easily leading to disputes. Second, faced with a large number of complex tender documents, manually comparing each item against the tender requirements is labor-intensive, inefficient, and prone to oversights and misjudgments due to fatigue. In particular, manual review struggles to comprehensively and accurately identify and assess the compliance of key performance indicators, the authenticity of responses, and the existence of hidden or negative deviations. Furthermore, this method lacks systematic quantitative analysis tools, making it impossible to conduct horizontal comparisons and priority rankings of numerous tender parameters, thus failing to generate data-driven decision support. Summary of the Invention

[0004] The purpose of this application is to provide a method and system for reviewing the technical parameters of medical device products in order to solve the above-mentioned technical problems, thereby improving the efficiency and accuracy of reviewing the compliance of the technical parameters of medical device products.

[0005] In some embodiments of this application, the tender documents are screened, divided into multiple technical sub-components, and each technical sub-component is initially screened according to the constructed anomaly assessment model. A comparison model is constructed based on the screening results to screen all tender documents, thereby determining whether each technical sub-component has any anomalies, thus improving the efficiency of reviewing the compliance of technical parameters of medical device products.

[0006] In some embodiments of this application, after anomaly screening of each technical sub-component, compliance risk values ​​of each technical sub-component are generated through the tender documents, thereby achieving dual judgment of each technical sub-component. By constructing a dual review model, the accuracy of reviewing the compliance of technical parameters of medical device products is improved.

[0007] In some embodiments of this application, a method for reviewing technical parameters for bidding on medical device products is provided, including:

[0008] Obtain the tender documents and all tender documents;

[0009] Set multiple technical sub - components according to the pre - processing results of the tender documents, and generate the parameter anomaly values of each technical sub - component according to the preset anomaly evaluation model;

[0010] Set the review strategy according to the preset associated review model and all parameter anomaly values, and generate the review result according to the review strategy;

[0011] Among them, generating multiple technical sub - components includes:

[0012] Establish a technical sub - component sequence A, A=(a1, a2…a …a n )), where a i is the i - th technical sub - component; n is the number of technical sub - components.

[0013] In some embodiments of the present application, generating the parameter anomaly values of each technical sub - component includes:

[0014] Set a i as the target technical sub - component in sequence according to the technical sub - component sequence A;

[0015] Generate the parameter anomaly value b of the target technical sub - component according to the anomaly evaluation model;

[0016] b= η i *s i ;

[0017] Among them, θ1 is the number of parameter evaluation indicators of the target technical sub - component; η i is the influence factor of the i - th evaluation indicator; s i is the reference value of the i - th parameter evaluation indicator in the target technical sub - component;

[0018] Generate the parameter anomaly values of each technical sub - component in sequence.

[0019] In some embodiments of the present application, setting the review strategy includes:

[0020] Establish a parameter anomaly value sequence B, B=(b1, b2…b i …b n ), where b i is the parameter anomaly value of the i - th technical sub - component; n is the number of technical sub - components;

[0021] Preset the parameter anomaly value threshold B1;

[0022] Select the maximum value b max in the parameter anomaly value sequence B;

[0023] If b max <B1, generate a first - level review instruction;

[0024] If b max > B1, generate a secondary review instruction;

[0025] Among them, the secondary review instruction includes:

[0026] If b i > B1, set the i-th technical sub-component as an abnormal sub-component;

[0027] Judge whether each technical sub-component is an abnormal sub-component in turn;

[0028] Obtain all abnormal sub-components according to the judgment result, and construct a primary comparison model based on all abnormal sub-components;

[0029] Judge whether to generate a primary review instruction according to the primary comparison model.

[0030] In some embodiments of the present application, the primary review instruction includes:

[0031] Obtain all tender documents;

[0032] Generate multiple response parameter packets according to the preprocessing results of each tender document;

[0033] Establish a response parameter packet sequence W, W = (w1, w2…w s i …w m ), where w i is the response parameter packet corresponding to the i-th tender document; m is the number of tender documents;

[0034] Set a i as the sub-component to be reviewed in turn according to the technical sub-component sequence A;

[0035] Generate a compliance risk value f for the sub-component to be reviewed according to all response parameter packets;

[0036] Preset a compliance risk value threshold F1;

[0037] If f > F1, set the sub-component to be reviewed as a risk sub-component;

[0038] If f < F1, set the sub-component to be reviewed as a compliant sub-component;

[0039] Generate the compliance risk values of each technical sub-component in turn, and generate a review result according to all compliance risk values.

[0040] In some embodiments of the present application, generating the compliance risk value f includes:

[0041] f = U1 * e * k i ;

[0042] Among them, U1 is a preset first conversion coefficient; e is a risk compensation coefficient set according to the parameter outlier of the sub-component to be reviewed; m is the number of tender documents; k i is the fit value between the sub-component to be reviewed and the i-th response parameter package.

[0043] In some embodiments of the present application, determining whether to generate a first-level review instruction includes:

[0044] Set w in sequence according to the response parameter package sequence W i as the target response package;

[0045] Generate the parameter similarity value g of the target response package according to the first-level comparison model;

[0046] Generate the parameter similarity values of each response parameter package in sequence;

[0047] Establish a parameter similarity value sequence G, G = (g1, g2... g i …g m ), where g i is the i-th parameter similarity value; m is the number of response parameter packages;

[0048] Preset a parameter similarity value threshold G1;

[0049] Select the maximum value g in the parameter similarity value sequence G max ;

[0050] If g max > G1, generate a first-level warning instruction;

[0051] If g[[ID=四十一]] max < G1, generate a first-level review instruction.

[0052] In some embodiments of the present application, generating the parameter similarity value g includes:

[0053] g = r * β i * k 1i ;

[0054] r = U2 * k 2i ;

[0055] Among them, r is a similarity compensation coefficient; θ2 is the number of abnormal sub-components in the first-level comparison model; β i is the influence factor of the i-th abnormal sub-component in the first-level comparison model; k 1i is the fit value between the i-th abnormal sub-component and the target response package in the first-level comparison model of the target response package; U2 is a preset second conversion coefficient; is the number of associated sub-components in the first-level comparison model; k 2iThis represents the fit value between the i-th associated sub-component and the target response package in the first-level comparison model.

[0056] In some embodiments of this application, a system for reviewing technical parameters for bidding on medical device products is provided, including:

[0057] The monitoring unit is used to obtain the tender documents and all bid documents;

[0058] The review unit includes:

[0059] The first review module is used to set multiple technical sub-components based on the preprocessing results of the tender documents;

[0060] Establish a sequence of technical components A, A=(a1, a2, ..., a3) i …a n ), where a i Let i be the i-th technical sub-component; n is the number of technical sub-components;

[0061] The second review module is used to generate abnormal parameter values ​​for each technical component based on a preset anomaly assessment model.

[0062] The third review module is used to set a review strategy based on a preset correlation review model and all parameter anomalies. The third review module is also used to generate review results based on the review strategy.

[0063] The second review module is also used for:

[0064] Based on the sequence of technical sub-components A, a is set sequentially. i For target technology components;

[0065] The abnormal parameter value b of the target technical component is generated based on the abnormality assessment model;

[0066] b=[ η i *s i ];

[0067] Where θ1 represents the number of parameter evaluation indicators for the target technical component; η i Let s be the influence factor of the i-th evaluation indicator; i This is a reference value for the evaluation index of the i-th parameter in the target technical sub-component;

[0068] The abnormal values ​​of parameters for each technical sub-component are generated sequentially.

[0069] In some embodiments of this application, the third review module is further configured to:

[0070] Establish a sequence of outlier parameters B, B=(b1,b2…b…). i …b n ), where b iis the parameter outlier value of the i-th technical sub-component; n is the number of technical sub-components;

[0071] The preset parameter outlier threshold B1;

[0072] Select the maximum value b in the parameter outlier sequence B max ;

[0073] If b max < B1, generate a first-level review instruction;

[0074] If b max > B1, generate a second-level review instruction;

[0075] Among them, the second-level review instruction includes:

[0076] If b i > B1, set the i-th technical sub-component as an abnormal sub-component;

[0077] Judge whether each technical sub-component is an abnormal sub-component in turn;

[0078] Obtain all abnormal sub-components according to the judgment results, and construct a first-level comparison model according to all abnormal sub-components; <s

[0079] Judge whether to generate a first-level review instruction according to the first-level comparison model.

[0080] In some embodiments of the present application, the first-level review instruction includes:

[0081] Obtain all tender documents;

[0082] Generate multiple response parameter packages according to the preprocessing results of each tender document;

[0083] Establish a response parameter package sequence W, W=(w1, w2…w i …w m ), where w i [[ID=五十三]]is the response parameter package corresponding to the i-th tender document; m is the number of tender documents;

[0084] Set a<s i as the sub-component to be reviewed according to the technical sub-component sequence A;

[0085] Generate the compliance risk value f of the sub-component to be reviewed according to all response parameter packages;

[0086] f = U1 * e * k i ;

[0087] Among them, U1 is the preset first conversion coefficient; e is the risk compensation coefficient set according to the parameter outlier value of the sub-component to be reviewed; m is the number of tender documents; k iThe fit value between the sub-component to be reviewed and the i-th response parameter package;

[0088] The preset compliance risk value threshold F1;

[0089] If f > F1, set the sub-component to be reviewed as a risk sub-component;

[0090] If f < F1, set the sub-component to be reviewed as a compliant sub-component;

[0091] Generate the compliance risk values of each technical sub-component in sequence, and generate the review result based on all the compliance risk values.

[0092] Compared with the prior art, the beneficial effects of the method and system for reviewing the technical parameters of a medical device product in an embodiment of the present application are as follows:

[0093] By screening the tender documents, dividing them into multiple technical sub-components, initially screening each technical sub-component according to the constructed abnormal evaluation model, and constructing a comparison model based on the screening results to screen all the tender documents, so as to determine whether there are abnormalities in each technical sub-component, and improve the review efficiency of the compliance of the technical parameters of medical device products.

[0094] After completing the abnormal screening of each technical sub-component, generate the compliance risk values of each technical sub-component through the tender documents, realize the double judgment of each technical sub-component, and improve the review accuracy of the compliance of the technical parameters of medical device products by constructing a double review model. Brief Description of the Drawings

[0095] Figure 1 It is a flowchart of a method for reviewing the technical parameters of a medical device product in a preferred embodiment of an embodiment of the present application. Detailed Embodiments

[0096] The following will further describe in detail the specific embodiments of the present application in conjunction with the drawings and embodiments. The following embodiments are used to illustrate the present application, but are not used to limit the scope of the present application.

[0097] In the description of the present application, it should be understood that the orientation or positional relationship indicated by the terms "center", "upper", "lower", "front", "rear", "left", "right", "vertical", "horizontal", "top", "bottom", "inner", "outer", etc. is based on the orientation or positional relationship shown in the drawings, and is only for the convenience of describing the present application and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be construed as limiting the present application.

[0098] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Therefore, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this application, unless otherwise stated, "a plurality of" means two or more.

[0099] In the description of this application, it should be noted that, unless otherwise expressly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection between two components. Those skilled in the art can understand the specific meaning of the above terms in this application based on the specific circumstances.

[0100] like Figure 1 As shown in the preferred embodiment of this application, a method for reviewing the technical parameters of a medical device product in a tender includes:

[0101] S101: Obtain the tender documents and all tender documents;

[0102] S102: Based on the preprocessing results of the tender documents, set up multiple technical sub-components and generate abnormal parameter values ​​for each technical sub-component based on the preset abnormality evaluation model;

[0103] S103: Set the review strategy based on the preset correlation review model and all parameter anomalies, and generate the review results according to the review strategy;

[0104] This involves generating multiple technical sub-components, including:

[0105] Establish a sequence of technical components A, A=(a1, a2, ..., a3) i …a n ), where a i Let be the i-th technical sub-component; n is the number of technical sub-components.

[0106] Specifically, by analyzing historical bidding data and combining it with semantic processing technology, a preprocessing model is constructed. This preprocessing model can extract various technical parameters and their corresponding related content (technical parameter categories, ranges, and expression methods) from the bidding documents.

[0107] Specifically, the current tender document is analyzed through a preprocessing model to generate multiple technical sub-components. Each technical sub-component represents a technical parameter package extracted from the tender document, which includes: technical parameter name, required parameter range, and name expression method.

[0108] Specifically, the technical parameters mentioned above refer to the technical parameters of various types of medical device products, including but not limited to: physical parameters, performance parameters, material property parameters, safety and electrical parameters, etc.

[0109] Specifically, it generates outlier values ​​for the parameters of each technical component, including:

[0110] Based on the sequence of technical sub-components A, a is set sequentially. i For target technology components;

[0111] The abnormal parameter value b of the target technical component is generated based on the abnormality assessment model;

[0112] b=[ η i *s i ];

[0113] Where θ1 represents the number of parameter evaluation indicators for the target technical component; η i Let s be the influence factor of the i-th evaluation indicator; i This is a reference value for the evaluation index of the i-th parameter in the target technical sub-component;

[0114] The abnormal values ​​of parameters for each technical sub-component are generated sequentially.

[0115] Specifically, the parameter evaluation indicators include, but are not limited to: historical deviation (the difference between historical and current required ranges), expression deviation (generating a general expression for each technical parameter and determining the difference between the current expression and the general expression), and scope requirement deviation (based on the actual requirements of the current tender document, determining the expected range of the technical parameters corresponding to the current technical sub-component and generating the difference between the current and expected required ranges). These parameters map out anomalies in the tender content. Through quantification, the reference values ​​of each parameter evaluation indicator are kept within the same range. Furthermore, the larger the reference value for each parameter evaluation indicator, the greater the likelihood that the tender content corresponding to the target technical sub-component has an anomaly risk.

[0116] Specifically, the impact factors of each parameter evaluation indicator can be set according to the degree of correlation with the anomalies in the bidding content. The greater the correlation, the larger the value of the corresponding impact factor.

[0117] Specifically, setting review strategies includes:

[0118] Establish a sequence of outlier parameters B, B=(b1,b2…b…). i …b n ), where b i represents the abnormal parameter value of the i-th technical sub-component; n is the number of technical sub-components;

[0119] Preset parameter outlier threshold B1;

[0120] Select the maximum value b in the parameter outlier sequence B max ;

[0121] If b max < B1, generate a first-level review instruction;

[0122] If b max > B1, generate a second-level review instruction;

[0123] Among them, the second-level review instruction includes:

[0124] If b i > B1, set the i-th technical sub-component as an abnormal sub-component;

[0125] Judge whether each technical sub-component is an abnormal sub-component in turn;

[0126] Obtain all abnormal sub-components according to the judgment results, and construct a first-level comparison model according to all abnormal sub-components;

[0127] Judge whether to generate a first-level review instruction according to the first-level comparison model.

[0128] Specifically, when generating a first-level review instruction, it means that there is no potential abnormal risk in the parameter content of the current bidding document.

[0129] Specifically, the parameter outlier threshold can be set according to historical parameters. If the parameter outlier of a technical sub-component is greater than the preset parameter outlier threshold, it means that there is a risk of directional setting (i.e., setting for a certain bidding document) for this technical sub-component. It is necessary to screen and check in time.

[0130] Specifically, construct a first-level comparison model according to all the selected abnormal sub-components and all the associated sub-components corresponding to each abnormal sub-component, and use the first-level comparison model to screen each bidding document to judge whether there is an abnormal risk.

[0131] Specifically, by analyzing all technical sub-components, judge whether there is an association relationship between each technical sub-component. (If the parameters corresponding to two technologies are different parameters of the same device, there is an association relationship, otherwise there is no association relationship). If there is an association relationship, the two technical sub-components are associated sub-components with each other.

[0132] It can be understood that in the above embodiments, by screening the bidding document, splitting multiple technical sub-components, initially screening each technical sub-component according to the constructed abnormal evaluation model, and constructing a comparison model based on the screening results to screen all bidding documents, so as to determine whether there is an abnormality in each technical sub-component, and improve the review efficiency of the compliance of the technical parameters of medical device products.

[0133] In the preferred embodiment of the embodiment of the present application, the first-level review instruction includes:

[0134] Obtain all the tender documents;

[0135] Generate multiple response parameter packets according to the preprocessing results of each tender document;

[0136] Establish a sequence of response parameter packets W, W = (w1, w2... w i …w m ), where w i is the response parameter packet corresponding to the i-th tender document; m is the number of tender documents;

[0137] Set a i as the sub-component to be reviewed according to the technical sub-component sequence A;

[0138] Generate a compliance risk value f for the sub-component to be reviewed according to all the response parameter packets;

[0139] Preset a compliance risk value threshold F1;

[0140] If f > F1, set the sub-component to be reviewed as a risk sub-component;

[0141] If f < F1, set the sub-component to be reviewed as a compliant sub-component;

[0142] Generate the compliance risk values of each technical sub-component in sequence, and generate a review result according to all the compliance risk values.

[0143] Specifically, the compliance risk value threshold can be set according to historical parameters. If the compliance risk value of the current technical sub-component is greater than the preset compliance risk value threshold, it means that the response degree of the tender document to this technical sub-component is insufficient, and there is a compliance risk in the formulation of this technical sub-component (that is, the formulation is unreasonable).

[0144] Specifically, extract all the technical sub-components in each tender document through a preprocessing model, and generate corresponding response parameter packets according to the technical sub-components of the current tender document.

[0145] Specifically, generating the compliance risk value f includes:

[0146] f = U1 * e * k i ;

[0147] where U1 is a preset first conversion coefficient; e is a risk compensation coefficient set according to the parameter anomaly value of the sub-component to be reviewed; m is the number of tender documents; k i is the fit value between the sub-component to be reviewed and the i-th response parameter packet.

[0148] Specifically, by presetting the first fixed coefficient, the compliance risk value is within the preset value range, and k i The larger the value, the smaller the corresponding compliance risk value, and the mapping relationship between the two can be set according to historical parameters.

[0149] Specifically, the larger the parameter outlier, the larger the value of the corresponding risk compensation coefficient, and the mapping relationship between the two can be set according to historical parameters.

[0150] Specifically, by analyzing the difference between the requirement range of the sub-piece to be reviewed and the response range in the response parameter package, the corresponding fit value is set. The smaller the difference, the larger the corresponding fit value, and the mapping relationship between the two can be set according to historical parameters.

[0151] Specifically, the larger the compliance risk value, the greater the possibility that the sub-piece to be reviewed has a compliance risk.

[0152] It can be understood that in the above embodiments, after the abnormal screening of each technical sub-piece is completed, the compliance risk value of each technical sub-piece is generated through the bidding document, realizing the dual judgment of each technical sub-piece, and by constructing a dual review model, the review accuracy of the compliance of the technical parameters of medical device products is improved.

[0153] In the preferred embodiment of this application, determining whether to generate a first-level review instruction includes:

[0154] Set w i as the target response package in sequence according to the response parameter package sequence W;

[0155] Generate the parameter similarity value g of the target response package according to the first-level comparison model;

[0156] Generate the parameter similarity values of each response parameter package in sequence;

[0157] Establish a parameter similarity value sequence G, G = (g1, g2... g i …g m ), where g i is the i-th parameter similarity value; m is the number of response parameter packages;

[0158] Preset the parameter similarity value threshold G1;

[0159] Select the maximum value g max in the parameter similarity value sequence G;

[0160] If g max > G1, generate a first-level warning instruction;

[0161] If g max < G1, generate a first-level review instruction.

[0162] Specifically, the parameter similarity threshold G1 can be set based on historical parameters. If the parameter similarity value of the target response package is greater than the preset parameter similarity threshold, it indicates a risk of targeted manipulation between the tender document corresponding to the target response package and the current tender document. A warning instruction should be generated promptly for verification to ensure the fairness of the bidding process.

[0163] Specifically, the generation of parameter similarity values ​​g includes:

[0164] g=r*[ β i *k 1i ];

[0165] r=U2*[ k 2i ];

[0166] Where r is the similarity compensation coefficient; θ2 is the number of anomalous sub-components in the first-level comparison model; β i k represents the influence factor of the i-th anomalous sub-component in the first-level comparison model. 1i U1 represents the matching value between the i-th abnormal sub-component and the target response package in the first-level comparison model of the target response package; U2 is the preset second conversion coefficient. k represents the number of associated sub-components in the first-level comparison model. 2i This represents the fit value between the i-th associated sub-component and the target response package in the first-level comparison model.

[0167] Specifically, by presetting a second conversion coefficient, the similarity compensation coefficient r is made to be within a preset value range, and [ k 2i The larger the value of ], the larger the corresponding similarity compensation coefficient r will be. The mapping relationship between the two can be set according to historical parameters.

[0168] Specifically, the impact factor of each abnormal sub-component can be set according to its corresponding abnormal parameter value. The larger the abnormal parameter value, the larger the value of the corresponding impact factor. The mapping relationship between the two can be set according to historical parameters.

[0169] It is understandable that in the above embodiments, after anomaly screening of each technical sub-component, compliance risk values ​​of each technical sub-component are generated through the tender documents, thereby achieving dual judgment of each technical sub-component. By constructing a dual review model, the accuracy of reviewing the compliance of technical parameters of medical device products is improved.

[0170] In another preferred embodiment of the method for reviewing technical parameters for bidding on medical device products based on any of the above preferred embodiments, this preferred embodiment provides a system for reviewing technical parameters for bidding on medical device products, including:

[0171] The monitoring unit is used to obtain the tender documents and all bid documents;

[0172] The review unit includes:

[0173] The first review module is used to set multiple technical sub-components based on the preprocessing results of the tender documents;

[0174] Establish a sequence of technical components A, A=(a1, a2, ..., a3) i …a n ), where a i Let i be the i-th technical sub-component; n is the number of technical sub-components;

[0175] The second review module is used to generate abnormal parameter values ​​for each technical component based on a preset anomaly assessment model.

[0176] The third review module is used to set review strategies based on the preset correlation review model and all parameter anomalies. The third review module is also used to generate review results based on the review strategies.

[0177] The second review module is also used for:

[0178] Based on the sequence of technical sub-components A, a is set sequentially. i For target technology components;

[0179] The abnormal parameter value b of the target technical component is generated based on the abnormality assessment model;

[0180] b=[ η i *s i ];

[0181] Where θ1 represents the number of parameter evaluation indicators for the target technical component; η i Let s be the influence factor of the i-th evaluation indicator; i This is a reference value for the evaluation index of the i-th parameter in the target technical sub-component;

[0182] The abnormal values ​​of parameters for each technical sub-component are generated sequentially.

[0183] In a preferred embodiment of this application, the third review module is further configured to:

[0184] Establish a sequence of outlier parameters B, B=(b1,b2…b…). i …b n ), where b i represents the abnormal parameter value of the i-th technical sub-component; n is the number of technical sub-components;

[0185] Preset parameter outlier threshold B1;

[0186] Select the maximum value b from the parameter outlier sequence B.max ;

[0187] If b max <B1, generate a first-level review instruction;

[0188] If b max >B1, generate a second-level review instruction;

[0189] Among them, the second-level review instruction includes:

[0190] If b i >B1, set the i-th technical sub-component as an abnormal sub-component;

[0191] Judge whether each technical sub-component is an abnormal sub-component in turn;

[0192] Obtain all abnormal sub-components according to the judgment result, and construct a first-level comparison model according to all abnormal sub-components;

[0193] Judge whether to generate a first-level review instruction according to the first-level comparison model.

[0194] In the preferred embodiment of this application embodiment, the first-level review instruction includes:

[0195] Obtain all tender documents;

[0196] Generate multiple response parameter packages according to the preprocessing results of each tender document;<00°0526>

[0197] Establish a response parameter package sequence W, W=(w1, w2…w i …w m ), where w i ​​​​​​​​​​​​​​​​​​​​​​​​​​​​

[0204] If f < F1, set the to-be-reviewed sub-component as a compliant sub-component;

[0205] Generate the compliance risk values of each technical sub-component in sequence, and generate a review result based on all the compliance risk values.

[0206] According to the first concept of the present application, by screening the tender documents, dividing multiple technical sub-components, initially screening each technical sub-component according to the constructed abnormal evaluation model, and constructing a comparison model based on the screening results to screen all the tender documents, so as to determine whether there are abnormalities in each technical sub-component, and improve the review efficiency of the compliance of the technical parameters of medical device products.

[0207] According to the second concept of the present application, after completing the abnormal screening of each technical sub-component, generate the compliance risk values of each technical sub-component through the tender documents, achieve a double judgment on each technical sub-component, and improve the review accuracy of the compliance of the technical parameters of medical device products by constructing a double review model.

[0208] The above are only the preferred embodiments of the present application. It should be noted that for those of ordinary skill in the art, without departing from the technical principle of the present application, several improvements and replacements can still be made, and these improvements and replacements should also be regarded as the protection scope of the present application.

Claims

1. A method for reviewing technical parameters in bidding for medical device products, characterized in that, Including: Obtain the tender documents and all bid documents; Set multiple technical sub-components according to the preprocessing results of the tender documents, and generate parameter anomaly values for each technical sub-component according to a preset anomaly evaluation model; Set a review strategy according to a preset association review model and all parameter anomaly values, and generate a review result according to the review strategy; Among them, generating multiple technical sub-components includes: Establish a sequence of technical components A, A=(a1, a2, ..., a3) i …a n ), where a i Let i be the i-th technical sub-component; n is the number of technical sub-components; Setting a review strategy includes: Establish a sequence of outlier parameters B, B=(b1,b2…b…). i …b n ), where b i represents the abnormal parameter value of the i-th technical sub-component; n is the number of technical sub-components; Preset parameter anomaly value threshold B1; Select the maximum value b from the parameter outlier sequence B. max ; If b max <B1, generate a first-level review instruction; If b max >B1, Generate a Level 2 Review Instruction; Among them, the secondary review instruction includes: If b i >B1, set the i-th technical sub-component as an abnormal sub-component; Sequentially determine whether each technical sub-component is an abnormal sub-component; Obtain all abnormal sub-components according to the judgment results, and construct a primary comparison model based on all abnormal sub-components; Judge whether to generate a primary review instruction according to the primary comparison model; The primary review instruction includes: Obtain all bid documents; Establish a response parameter packet sequence W, W=(w1,w2…w i …w m ), where w i This represents the response parameter package corresponding to the i-th bid document; m is the number of bid documents. Based on the sequence of technical sub-components A, a is set sequentially. i Sub-component pending review; Generate multiple response parameter packages according to the preprocessing results of each bid document; Generate the compliance risk value f of the sub-component to be reviewed according to all response parameter packages; Preset compliance risk value threshold F1; If f > F1, set the sub-component to be reviewed as a risk sub-component; If f < F1, set the sub-component to be reviewed as a compliant sub-component; Sequentially generate the compliance risk values of each technical sub-component, and generate a review result according to all compliance risk values; Based on the response parameter packet sequence W, set w sequentially. i For the target response packet; Judging whether to generate a primary review instruction includes: Generate the parameter similarity value g of the target response package according to the primary comparison model; Establish a parameter similarity value sequence G, G = (g1, g2, ..., g...) i …g m ), where g i The similarity value of the i-th parameter; m is the number of response parameter packets; Sequentially generate the parameter similarity values of each response parameter package; Select the maximum value g from the parameter similarity value sequence G. max ; If g max >G1, generate a Level 1 warning instruction; If g max <G1, generate a first-level review instruction; Preset parameter similarity value threshold G1; g=r [ b i k 1i ]; r=U2 [ k 2i ]; Where r is the similarity compensation coefficient; θ2 is the number of anomalous sub-components in the first-level comparison model; β i k represents the influence factor of the i-th anomalous sub-component in the first-level comparison model. 1i U1 represents the matching value between the i-th abnormal sub-component and the target response packet in the first-level comparison model; U2 is the preset second conversion coefficient. k represents the number of associated sub-components in the first-level comparison model. 2i This represents the fit value between the i-th associated sub-component and the target response package in the first-level comparison model; Generating the parameter similarity value g includes: Construct a primary comparison model according to all selected abnormal sub-components and all associated sub-components corresponding to each abnormal sub-component; 2. The method for reviewing the technical parameters of medical device products in bidding as described in claim 1, characterized in that, By analyzing all technical sub-components, judge whether there is an association relationship between each technical sub-component. If there is an association relationship, the two technical sub-components are mutually associated sub-components. Based on the sequence of technical sub-components A, a is set sequentially. i For target technology components; The generating of the parameter anomaly values of each technical sub-component includes: b=[ or i s i ]; Where θ1 represents the number of parameter evaluation indicators for the target technical component; η i Let s be the influence factor of the i-th evaluation indicator; i This is a reference value for the evaluation index of the i-th parameter in the target technical sub-component; Generate the parameter anomaly value b of the target technical sub-component according to the anomaly evaluation model; 3. The method for reviewing the technical parameters of medical device products in bidding as described in claim 2, characterized in that, Sequentially generate the parameter anomaly values of each technical sub-component. f=U1 of [ k i ]; Where U1 is the preset first conversion coefficient; e is the risk compensation coefficient set according to the abnormal values ​​of the parameters of the sub-component to be reviewed; m is the number of tender documents; k i This represents the match value between the component to be reviewed and the i-th response parameter packet.

4. A system for reviewing technical parameters for bidding on medical device products, employing the method for reviewing technical parameters for bidding on medical device products as described in any one of claims 1-3, characterized in that, Generating the compliance risk value f includes: [[ID=3O]]Including: A monitoring unit for obtaining the tender documents and all bid documents; A review unit includes: Establish a sequence of technical components A, A=(a1, a2, ..., a3) i …a n ), where a i Let i be the i-th technical sub-component; n is the number of technical sub-components; A first review module for setting multiple technical sub-components according to the preprocessing results of the tender documents; A second review module for generating the parameter anomaly values of each technical sub-component according to a preset anomaly evaluation model; A third review module for setting a review strategy according to a preset association review model and all parameter anomaly values, and the third review module is also used to generate a review result according to the review strategy; Based on the sequence of technical sub-components A, a is set sequentially. i For target technology components; Among them, the second review module is also used for: b=[ or i s i ]; Where θ1 represents the number of parameter evaluation indicators for the target technical component; η i Let s be the influence factor of the i-th evaluation indicator; i This is a reference value for the evaluation index of the i-th parameter in the target technical sub-component; Generate the parameter anomaly value b of the target technical sub-component according to the anomaly evaluation model; 5. The review system for technical parameters of medical device products in bidding as described in claim 4, characterized in that, Sequentially generate the parameter anomaly values of each technical sub-component. Establish a sequence of outlier parameters B, B=(b1,b2…b…). i …b n ), where b i represents the abnormal parameter value of the i-th technical sub-component; n is the number of technical sub-components; The third review module is also used for: Select the maximum value b from the parameter outlier sequence B. max ; If b max <B1, generate a first-level review instruction; If b max >B1, Generate a Level 2 Review Instruction; Preset parameter anomaly value threshold B1; If b i >B1, set the i-th technical sub-component as an abnormal sub-component; Among them, the secondary review instruction includes: Sequentially determine whether each technical sub-component is an abnormal sub-component; Obtain all abnormal sub-components according to the judgment results, and construct a primary comparison model based on all abnormal sub-components; 6. The review system for technical parameters of medical device products in bidding as described in claim 5, characterized in that, Judge whether to generate a primary review instruction according to the primary comparison model. The primary review instruction includes: Obtain all bid documents; Establish a response parameter packet sequence W, W=(w1,w2…w i …w m ), where w i This represents the response parameter package corresponding to the i-th bid document; m is the number of bid documents. Based on the sequence of technical sub-components A, a is set sequentially. i Sub-component pending review; Generate multiple response parameter packages according to the preprocessing results of each bid document; Generate the compliance risk value f of the sub-component to be reviewed according to all response parameter packages; f=U1 of [ k i ]; Where U1 is the preset first conversion coefficient; e is the risk compensation coefficient set according to the abnormal values ​​of the parameters of the sub-component to be reviewed; m is the number of tender documents; k i The matching value between the sub-component to be reviewed and the i-th response parameter packet; Preset the compliance risk value threshold F1; If f > F1, set the sub-component to be reviewed as a risk sub-component; If f < F1, set the sub-component to be reviewed as a compliant sub-component; Generate the compliance risk values of each technical sub-component in sequence, and generate the review result based on all the compliance risk values.

Images