Building material witness sampling, inspection, re-inspection whole process traceability control system

Abstract

The application discloses a building material witness sampling, inspection, reinspection whole-process traceability management system, and particularly relates to the technical field of traceability management; the same source relationship set of the parent sample and the child sample is constructed by acquiring the parent sample identification, witness information and segmentation and division sample preparation and sample reservation operation records, and the state drift index of the child sample relative to the parent sample is calculated based on the sample state transfer model to identify abnormal child samples; the operation abnormal contribution degree is obtained by further analyzing the segmentation and division granularity uniformity deviation and the sample preparation and sample reservation environment cumulative effect deviation of the abnormal child samples, and the result deviation representative degree is obtained by simultaneously analyzing the result dispersion of the normal child sample detection result and the parent sample detection result and the inspection transportation environment cumulative effect; the target traceability link is determined according to the operation abnormal contribution degree and the result deviation representative degree, and the whole-process traceability chain containing the parent sample identification, the child sample identification and the target traceability link is generated.

Description

Technical Field

[0001] This invention relates to the field of traceability and control technology, and more specifically, to a traceability and control system for the entire process of witness sampling, testing, and retesting of building materials. Background Technology

[0002] Building material testing typically includes witness sampling, sealing, sending for testing, testing, and retesting. The relevant process records are generally formed by the participating entities such as construction, supervision, construction, and testing. The records include information such as sample source, witness sampling, sample division, reduction, sample preparation, sample retention, transportation and handover, and test results.

[0003] Existing building material testing process management focuses more on the retention of records and result verification of individual links. The correlation between records of different links is limited. It is difficult to form a continuous and consistent traceability chain for the source relationship and process status of samples from parent samples to sub-samples, from sampling to delivery for testing, and from initial testing to retesting. When the initial testing results are inconsistent with the retesting results, it is difficult to make a unified traceability judgment on links such as sampling witnessing, segmentation and reduction, sample preparation and retention, and delivery and transportation for testing based on the same source relationship. This affects the accuracy of locating abnormal links and the reliability of traceability processing. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, embodiments of the present invention provide a traceability and control system for the entire process of witness sampling, testing, and re-testing of building materials to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] A comprehensive traceability and control system for building materials, encompassing witness sampling, testing, and re-testing, including:

[0007] Sample source construction module: acquires the parent sample identifier, witnessing information and segmentation and reduction sample preparation and retention operation records recorded during the witness sampling process of building materials, and constructs a set of homology relationships between the parent sample and the sub-samples;

[0008] Drift determination module: For each sub-sample in the homology set, the preset sample state transition model is used to calculate the state drift index of the sub-sample relative to the parent sample, and to determine whether there are abnormal sub-samples whose state drift index exceeds the preset threshold.

[0009] Anomaly Contribution Module: If there are abnormal sub-samples, extract the segmentation, reduction, sample preparation, and retention operation records of the abnormal sub-samples, analyze the deviation of the uniformity of the segmentation and reduction particle size of the abnormal sub-samples and the cumulative effect deviation of the sample preparation and retention environment, and obtain the operational anomaly contribution of the abnormal sub-samples.

[0010] Deviation Representativeness Module: Analyzes the dispersion of the test results of normal subsamples and the test results of the parent sample, as well as the cumulative effect of the delivery and transportation environment, to obtain the deviation representativeness of the results of normal subsamples;

[0011] Link identification module: Based on the contribution of abnormal sub-samples to operational abnormalities and the representativeness of normal sub-samples to result deviations, identify the target tracing link in the homology set that causes inconsistencies between the initial test results and the retest results;

[0012] Traceability Chain Module: Based on the target traceability link, trace back the parent sample identifier and sub-sample identifier associated with the target traceability link in the set of homologous relationships to generate a full-process traceability chain.

[0013] In a preferred embodiment, the parent sample identifier, witnessing information, and segmentation and reduction sample preparation and retention operation records recorded during the witness sampling process of building materials are obtained to construct a set of homology relationships between the parent sample and the sub-samples, specifically as follows:

[0014] Starting with the parent sample identifier, analyze the preceding and following sample identifiers generated by each segmentation, reduction, sample preparation, and retention operation in the segmentation, reduction, sample preparation, and retention operation record.

[0015] Establish a parent-child relationship between the pre-sample identifier and the post-sample identifier, and link the parent-child relationships step by step starting from the parent sample identifier. Bind the witness information to the parent sample identifier to generate a set of homologous relationships that includes the parent sample identifier, the child sample identifier and the corresponding parent-child relationship.

[0016] In a preferred embodiment, for each sub-sample in the homology set, a preset sample state transition model is used to calculate the state drift index of the sub-sample relative to the parent sample, and it is determined whether there are any abnormal sub-samples whose state drift index exceeds a preset threshold. Specifically:

[0017] Extract the operation time and operation environment information from the segmentation and reduction sample preparation and retention operation records corresponding to each sub-sample from the homology set, and input the operation time and operation environment information into the preset sample state transition model.

[0018] The preset sample state transition model calculates the cumulative effect of time based on the operation time, and the cumulative effect of environment based on the operation environment information. Then, it calculates the state drift index by combining the homology between the subsample and the parent sample.

[0019] The state drift index is compared with a preset threshold, and subsamples whose state drift index exceeds the preset threshold are marked as abnormal subsamples.

[0020] In a preferred embodiment, if abnormal subsamples exist, the segmentation, reduction, sample preparation, and retention operation records of the abnormal subsamples are extracted. The deviations in the uniformity of the segmentation and reduction particle size of the abnormal subsamples and the cumulative effects of the sample preparation and retention environment are analyzed to obtain the operational anomaly contribution of the abnormal subsamples. Specifically:

[0021] For sub-samples whose state drift index exceeds a preset threshold, extract the particle size distribution data of the sample before segmentation and the particle size distribution data of the sub-sample after segmentation from the segmentation and reduction sample preparation and retention operation record, and calculate the segmentation and reduction particle size uniformity deviation.

[0022] By comparing the operational environment information corresponding to the abnormal sub-samples in the operation environment information recorded in the segmentation and reduction sample preparation and retention operation record with the standard sample preparation and retention environment parameters, the cumulative effect deviation of the sample preparation and retention environment is obtained.

[0023] The operational anomaly contribution of abnormal subsamples is obtained by weighted summing of the deviation in uniformity of particle size reduction and the cumulative effect of the sample preparation and retention environment.

[0024] In a preferred embodiment, the dispersion of the test results of normal subsamples and the test results of the parent sample, as well as the cumulative effect of the delivery and transportation environment, are analyzed to obtain the representativeness of the deviation of the normal subsample results. Specifically:

[0025] Subsamples whose state drift index does not exceed a preset threshold are extracted from the set of homology relationships and designated as normal subsamples.

[0026] Obtain the test results of normal subsamples and parent samples, calculate the absolute value of the difference between the test results of normal subsamples and parent samples relative to the test results of parent samples, and obtain the result dispersion between the test results of normal subsamples and parent samples.

[0027] Extract the inspection and transportation time and environment information from the inspection and transportation records, and calculate the cumulative effect of the inspection and transportation environment based on the inspection and transportation time and environment information.

[0028] The deviation of the normal subsample from the parent sample is obtained by weighting and summing the dispersion of the test results between the normal subsample and the parent sample, as well as the cumulative effect of the transportation environment.

[0029] In a preferred embodiment, based on the operational anomaly contribution of the abnormal subsamples and the result deviation representativeness of the normal subsamples, the target tracing link in the homology set that causes the inconsistency between the initial test results and the retest results is identified, specifically:

[0030] Compare the operational anomaly contribution of abnormal subsamples with a preset contribution threshold, and compare the deviation of normal subsamples from representativeness with a preset deviation threshold.

[0031] When the contribution of abnormal operation of abnormal subsamples exceeds the preset contribution threshold, the segmentation and reduction process or the sample preparation and retention process is identified as the target traceability process based on the magnitude of the deviation in uniformity of segmentation and reduction particle size and the cumulative effect deviation of the sample preparation and retention environment.

[0032] When the contribution of abnormal operation to abnormal sub-samples is not greater than the preset contribution threshold and the deviation of normal sub-samples from representativeness is greater than the preset deviation threshold, the delivery and transportation link will be identified as the target traceability link.

[0033] When the contribution of abnormal operation to abnormal sub-samples is not greater than the preset contribution threshold and the deviation of normal sub-samples from representativeness is not greater than the preset deviation threshold, the witness sampling step will be identified as the target traceability step.

[0034] In a preferred embodiment, based on the target tracing step, the parent sample identifier and child sample identifier associated with the target tracing step in the homology relationship set are traced back to generate a full-process tracing chain, specifically as follows:

[0035] When the target tracing link is the segmentation and reduction link or the sample preparation and retention link, extract the abnormal sub-samples and the parent sample identifiers that have a parent-child relationship with the abnormal sub-samples from the homology set.

[0036] When the target traceability link is the delivery and transportation link, extract the normal sub-sample and the parent sample identifier that has a parent-child relationship with the normal sub-sample from the homology set;

[0037] When the target tracing step is the witness sampling step, extract the parent sample identifier that is bound to the witness information and all child sample identifiers that have a parent-child relationship with the parent sample identifier that is bound to the witness information from the same source relationship set;

[0038] The extracted parent sample identifier and child sample identifier are linked together according to the parent-child relationship, and the target traceability link is marked according to the linking result to generate a full-process traceability chain.

[0039] The technical effects and advantages of the present invention's traceability and control system for the entire process of witness sampling, testing, and re-testing of building materials are as follows:

[0040] By constructing a set of homology relationships between the parent sample and sub-samples, the sample identification relationships during witnessed sampling, segmentation and reduction, sample preparation and retention, delivery and transportation, initial inspection, and re-inspection are continuously expressed, providing a traceable basis for the sample origin relationship. By calculating the state drift index of sub-samples relative to the parent sample, abnormal sub-samples that may have state shifts can be identified, improving the targeting of abnormal sample screening. By analyzing the uniformity deviation of segmentation and reduction granularity and the cumulative effect deviation of the sample preparation and retention environment of abnormal sub-samples, the impact of segmentation and reduction or sample preparation and retention on sample abnormalities can be quantified. By analyzing the result dispersion between the test results of normal sub-samples and the test results of the parent sample, as well as the cumulative effect of the delivery and transportation environment, the impact of the delivery and transportation process on the deviation of test results can be evaluated. By comprehensively considering the contribution of operational anomalies and the representativeness of result deviations, the target traceability link is identified, and a full-process traceability chain including the parent sample identifier, sub-sample identifier, and target traceability link is generated, which can improve the accuracy of anomaly location and the reliability of traceability processing when the initial inspection results and re-inspection results are inconsistent. Attached Figure Description

[0041] Figure 1 This is a schematic diagram of the structure of the traceability and control system for the entire process of witness sampling, testing, and re-testing of building materials according to the present invention. Detailed Implementation

[0042] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0043] Example 1

[0044] Figure 1 This invention presents a traceability and control system for the entire process of witness sampling, testing, and re-testing of building materials, including:

[0045] Sample source construction module: acquires the parent sample identifier, witnessing information and segmentation and reduction sample preparation and retention operation records recorded during the witness sampling process of building materials, and constructs a set of homology relationships between the parent sample and the sub-samples;

[0046] Drift determination module: For each sub-sample in the homology set, the preset sample state transition model is used to calculate the state drift index of the sub-sample relative to the parent sample, and to determine whether there are abnormal sub-samples whose state drift index exceeds the preset threshold.

[0047] Anomaly Contribution Module: If there are abnormal sub-samples, extract the segmentation, reduction, sample preparation, and retention operation records of the abnormal sub-samples, analyze the deviation of the uniformity of the segmentation and reduction particle size of the abnormal sub-samples and the cumulative effect deviation of the sample preparation and retention environment, and obtain the operational anomaly contribution of the abnormal sub-samples.

[0048] Deviation Representativeness Module: Analyzes the dispersion of the test results of normal subsamples and the test results of the parent sample, as well as the cumulative effect of the delivery and transportation environment, to obtain the deviation representativeness of the results of normal subsamples;

[0049] Link identification module: Based on the contribution of abnormal sub-samples to operational abnormalities and the representativeness of results deviations in normal sub-samples, identify the target tracing link in the homology set that causes inconsistencies between the initial test results and the retest results;

[0050] Traceability Chain Module: Based on the target traceability link, trace back the parent sample identifier and sub-sample identifier associated with the target traceability link in the set of homologous relationships to generate a full-process traceability chain.

[0051] Obtain the parent sample identifier, witnessing information, and segmentation and reduction sample preparation and retention operation records recorded during the witness sampling process of building materials, and construct a set of homology relationships between the parent sample and the sub-samples, including:

[0052] In one implementation, the parent sample identifier, witnessing information, and segmentation and reduction sample retention operation record generated during the witnessing sampling of building materials are obtained. The parent sample identifier is used to uniquely indicate the source of the original sample obtained from the same batch of materials during witnessing sampling. The parent sample identifier is formed by combining the project code, material batch code, sampling date, sampling location code, and sampling sequence number in a fixed order. The fixed order can be the project code, material batch code, sampling date, sampling location code, and sampling sequence number arranged sequentially. The fields are connected by horizontal lines. For example, the parent sample identifier can be set as "P23001-B20230801-20230815-A01-0001".

[0053] Witnessing information is used to record the sampling supervision facts during the witnessed sampling process of building materials, but is not limited to the identity information of witnessing personnel, the identity information of sampling personnel, sampling time, sampling location, material name, material batch, witnessing method, sealing confirmation information, and witnessing image index information. Witnessing personnel identity information includes the witnessing personnel's name, document type, document number, job title, and contact number. Sampling personnel identity information includes the sampling personnel's name, document type, document number, job title, and contact number. Witnessing method is used to indicate the way the witnessing personnel participate in the sampling supervision. The witnessing method can be on-site witnessing, video witnessing, or verification witnessing. Sealing confirmation information includes the sealing label number, sealing time, sealing photo index, and sealing integrity confirmation value. Sealing integrity confirmation value is used to indicate the sealing status of the sample. The sealing integrity confirmation value can use 1 to indicate that the sealing is complete, 0 to indicate that the sealing is incomplete, and -1 to indicate that the sealing status cannot be confirmed. Witnessing image index information is used to locate the image files or video files of the sampling process. Witnessing image index information includes the image file name, image formation time, image storage location number, and image acquisition personnel's name.

[0054] The segmentation, reduction, sample preparation, and retention operation record is used to record every sample identification change that occurs during the segmentation, reduction, sample preparation, and retention process of the sample corresponding to the parent sample identification. The segmentation, reduction, sample preparation, and retention operation record includes at least the operation sequence number, operation type, previous sample identification, subsequent sample identification, operation time, operator identity information, operation location, sample quality before operation, sample quality after operation, sample container identification, and operation result confirmation information. The operation sequence number indicates the order of segmentation, reduction, sample preparation, and retention operations in the record formation process. The operation type indicates segmentation, reduction, sample preparation, or retention. The previous sample identification indicates the sample identification that participated in the operation before the start of a segmentation, reduction, sample preparation, and retention operation. The subsequent sample identification indicates the sample identification formed after the completion of a segmentation, reduction, sample preparation, and retention operation. There can be one or more subsequent sample identifications. When there are multiple subsequent sample identifications, each subsequent sample identification represents multiple sub-samples generated by the same segmentation, reduction, sample preparation, and retention operation. The sub-sample identification is used to uniquely identify the sample directly formed from the parent sample or indirectly formed from the parent sample through at least one segmentation, reduction, sample preparation, and retention operation.

[0055] Before parsing the segmentation and reduction sample preparation and retention operation records, a field integrity check is performed on the operation records. The field integrity check includes checking whether the operation sequence number and operation time are empty. For segmentation and reduction sample preparation and retention operation records with empty operation time but not empty operation sequence number, they are sorted according to the operation sequence number. For segmentation and reduction sample preparation and retention operation records with not empty operation time and not empty operation sequence number, they are sorted from earliest to latest operation time, and then sorted from smallest to largest operation sequence number. The sorting result is used as the parsing order of the preceding sample identifier and the following sample identifier.

[0056] Starting with the parent sample identifier, the operation type, previous sample identifier, and subsequent sample identifier in the segmentation and reduction sample preparation and retention operation record are read in the parsing order. When the previous sample identifier is equal to the parent sample identifier, the segmentation and reduction sample preparation and retention operation record corresponding to the previous sample identifier is determined as a serializable record. When the previous sample identifier is not equal to the parent sample identifier, but the previous sample identifier has already appeared as a subsequent sample identifier in the segmentation and reduction sample preparation and retention operation record that is earlier in the parsing order, the segmentation and reduction sample preparation and retention operation record corresponding to the previous sample identifier is determined as a serializable record.

[0057] When establishing parent-child relationships, the preceding sample identifier in each concatenable record is used as the parent sample identifier, and each subsequent sample identifier in the same concatenable record is used as the child sample identifier, thus forming a parent-child relationship from the parent sample identifier to the child sample identifier. The parent-child relationship is a directional data relationship, with the direction of the parent-child relationship pointing from the parent sample identifier to the child sample identifier. The content of the parent-child relationship record includes the parent sample identifier, child sample identifier, operation type, operation sequence number, operation time, and relationship generation time. For example, if the parent sample identifier is M001, when the first reduction operation generates subsequent sample identifiers S001 and S002, a parent-child relationship is established from M001 to S001 and from M001 to S002. When the subsequent sample preparation operation generates the subsequent sample identifier T001 from the preceding sample identifier S001, a parent-child relationship is established from S001 to T001.

[0058] When linking parent and child relationships step by step starting from the parent sample identifier, the parent sample identifier is used as the root node. All child sample identifiers with the parent sample identifier as their parent sample identifier are identified as the first-level child sample identifiers. Then, each of the first-level child sample identifiers is used as a new parent sample identifier to search for the next-level child sample identifier. This search is repeated layer by layer until there are no next-level child sample identifiers with the current-level child sample identifier as their parent sample identifier. During the step-by-step linking process, a level number and path code are generated for each child sample identifier. The level number is used to indicate the number of segmentation, reduction, sample preparation, and retention operations that the child sample identifier has undergone relative to the parent sample identifier. The level number of the first-level child sample identifier is 1, and the level number of the second-level child sample identifier is 2. The path code is used to indicate the sequence of sample identifiers that are passed from the parent sample identifier to the child sample identifier. The path code can be composed of the parent sample identifier, each level of child sample identifier, and a slash separator. For example, the path code can be set to "M001 / S001 / T001".

[0059] When binding witnessing information to the parent sample identifier, the witnessing information is written as the root witnessing attribute of the parent sample identifier into the same-origin relationship set, and a binding operation record is generated. The binding operation record includes at least the parent sample identifier, witnessing personnel name, sampling personnel name, sampling time, sampling location, material name, material batch, binding time, binding personnel identity information, and binding operation verification value. The binding operation verification value is used to verify whether the binding content between the witnessing information and the parent sample identifier has changed. The binding operation verification value is generated through a fixed field connection method. The fixed field connection method arranges the field content in the order of parent sample identifier, witnessing personnel name, sampling personnel name, sampling time, sampling location, material name, material batch, and binding time. The fields are connected by a vertical bar separator. When the field content is empty, it is filled with the null character NULL. The string formed after the connection is completed is used as the binding operation verification value.

[0060] Witness information is indirectly linked to each child sample identifier through the parent sample identifier and all parent-child relationships. The indirect linking method is that each child sample identifier traces back to the parent sample identifier through path coding, and then the parent sample identifier reads the witness information. The witness information is not written to each child sample identifier separately.

[0061] When generating a set of common origin relationships, the parent sample identifier, child sample identifier, parent-child relationship, witness information, hierarchical number, path code, and binding operation record are saved as a set of common origin relationships. The set of common origin relationships can be implemented using a tabular data structure, a tree data structure, or a graph data structure. In a tabular data structure, each row corresponds to a parent-child relationship. In a tree data structure, the parent sample identifier is the root node and the child sample identifier is the lower-level node. In a graph data structure, the parent sample identifier and the child sample identifier are nodes and the parent-child relationship is a directed edge. After the set of common origin relationships is generated, it should at least support querying the corresponding parent sample identifier by any child sample identifier, querying all child sample identifiers by the parent sample identifier, querying the path code from the parent sample identifier to the child sample identifier by any child sample identifier, and querying the bound witness information by the parent sample identifier.

[0062] For each sub-sample in the homology set, a preset sample state transition model is used to calculate the state drift index of the sub-sample relative to the parent sample. The system then determines whether there are any anomalous sub-samples whose state drift index exceeds a preset threshold, including:

[0063] The system reads the following from the homology set: each subsample identifier, the parent sample identifier corresponding to each subsample identifier, the path code corresponding to each subsample identifier, the hierarchical number corresponding to each subsample identifier, and the segmentation and reduction sample preparation and retention operation record corresponding to each parent-child relationship in the path code. The operation time in the segmentation and reduction sample preparation and retention operation record indicates the time when the segmentation and reduction sample preparation and retention operation is completed and the subsequent sample identifier is formed. The operation environment information in the segmentation and reduction sample preparation and retention operation record includes the operation environment temperature, operation environment relative humidity, operation environment dust concentration, operation environment vibration intensity, and sample exposure status. The unit of operation environment temperature is degrees Celsius, the unit of operation environment relative humidity is percentage, the unit of operation environment dust concentration is milligrams per cubic meter, the operation environment vibration intensity is represented by integers from 0 to 5, where 0 indicates no vibration was recorded and 5 indicates continuous high-intensity vibration, and the sample exposure status is represented by 0 or 1, where 0 indicates that the sample is in a closed container state during the operation interval and 1 indicates that the sample is in an unclosed exposed state during the operation interval.

[0064] The path code corresponding to each sub-sample identifier is parsed into a sequence of sample identifiers that pass sequentially from the parent sample identifier to the current sub-sample identifier. Then, according to the sample identifier sequence, the segmentation and sample preparation operation record corresponding to each parent-child relationship is read to form the path operation record sequence corresponding to the current sub-sample identifier. The first segmentation and sample preparation operation record in the path operation record sequence takes the sampling time contained in the witness information as the starting point of the time interval. The second and subsequent segmentation and sample preparation operation records in the path operation record sequence take the operation time of the previous segmentation and sample preparation operation record as the starting point of the time interval. The operation time of each segmentation and sample preparation operation record is taken as the ending point of the time interval.

[0065] The preset sample state transition model is a deterministic calculation process consisting of time cumulative action calculation rules, environmental cumulative action calculation rules, homologous hierarchical action calculation rules, and weighted synthesis rules. Time cumulative action represents the total time influence experienced by the current sub-sample identifier from the formation of the parent sample identifier to the formation of the current sub-sample identifier. Environmental cumulative action represents the total environmental influence experienced by the current sub-sample identifier from the formation of the parent sample identifier to the formation of the current sub-sample identifier. Homologous hierarchical action represents the influence of the number of parent-child relationship layers experienced by the current sub-sample identifier relative to the parent sample identifier on the sample state difference. The state drift index represents the comprehensive degree of state shift of the current sub-sample identifier relative to the parent sample identifier. The state drift index ranges from 0 to 1. The closer the state drift index is to 1, the higher the probability that the current sub-sample identifier has shifted its state relative to the parent sample identifier.

[0066] The cumulative effect over time is calculated using the following formula: ;

[0067] in, Indicates the first The cumulative time effect corresponding to each subsample identifier is expressed in minutes. Indicates the index of the subsample identifier in the set of homology relations; Indicates the first The number of segmentation, reduction, sample preparation, and retention operation records in the path operation record sequence corresponding to each sub-sample identifier; Indicates the first The sequence number of the segmentation, reduction, sample preparation, and retention operation record in the path operation record sequence corresponding to each sub-sample identifier; Indicates the first The path operation record sequence corresponding to each sub-sample identifier is the first one. The time interval corresponding to the segmentation and reduction sample preparation and retention operation record is in minutes; Indicates the first The path operation record sequence corresponding to each sub-sample identifier is the first one. The operation type coefficient corresponding to the operation record of segmented and reduced sample preparation and retention.

[0068] The operation type coefficient is set according to the relative degree of influence of the operation type on the sample state. The operation type coefficient is determined through historical sample records. Historical sample records come from the same engineering project, the same material category, or the same testing management scope, and have completed witnessed sampling, submission for testing, retesting, and formed complete test results. The method for determining the operation type coefficient is to calculate the average correlation strength between the time interval corresponding to segmentation, reduction, sample preparation, and sample retention and the difference ratio between the initial test result and the retest result. The operation type coefficient corresponding to the minimum average correlation strength is set to 1, and different operation type coefficients are determined according to the ratio of the average correlation strength to the minimum average correlation strength of different operation types. For example, the operation type coefficient corresponding to segmentation can be set to 1.1, the operation type coefficient corresponding to reduction can be set to 1.2, the operation type coefficient corresponding to sample preparation can be set to 1.4, and the operation type coefficient corresponding to sample retention can be set to 1.0.

[0069] The cumulative environmental impact is calculated using the following formula:

[0070] ;

[0071] in, Indicates the first The cumulative environmental impact corresponding to each subsample is indicated in minutes. Indicates the first The path operation record sequence corresponding to each sub-sample identifier is the first one. The ambient temperature corresponding to the segmented and reduced sample preparation and retention operation record, in degrees Celsius; This indicates the reference operating temperature for the corresponding material type, in degrees Celsius. This represents the normalized reference value for temperature deviation, in degrees Celsius. This indicates the relative humidity of the operating environment, expressed as a percentage. The relative humidity of the baseline operating environment for the corresponding material category is expressed as a percentage. This represents the normalized baseline value for relative humidity deviation, expressed as a percentage. This indicates the dust concentration in the operating environment, expressed in milligrams per cubic meter. This represents the normalized baseline value for dust concentration, expressed in milligrams per cubic meter. This represents the vibration intensity of the operating environment, with values ​​ranging from 0 to 5 (integers). This represents the normalized reference value for vibration intensity, and its value is greater than 0. Indicates the sample exposure status, with a value of 0 or 1; , , , , These represent the weights of temperature, relative humidity, dust concentration, vibration intensity, and sample exposure status, respectively, with the sum of these weights being 1.

[0072] The weights for the influence of temperature, relative humidity, dust concentration, vibration intensity, and sample exposure status are determined by the correlation between the proportions of differences between the various operating environment information in historical sample records and the initial and retest results. The higher the correlation, the greater the influence weight of the operating environment information. For example, the influence weight of temperature can be set to 0.25, the influence weight of relative humidity can be set to 0.30, the influence weight of dust concentration can be set to 0.15, the influence weight of vibration intensity can be set to 0.10, and the influence weight of sample exposure status can be set to 0.20.

[0073] The reference operating environment temperature, reference operating environment relative humidity, normalized reference value for temperature deviation, normalized reference value for relative humidity deviation, normalized reference value for dust concentration, and normalized reference value for vibration intensity are determined separately according to material category. The determination method involves selecting sample records from historical sample records where the difference between the initial test result and the retest result is less than a preset stable ratio. The preset stable ratio is determined based on the allowable ratio of test differences pre-set in the application document. The allowable ratio of test differences can be determined based on the 75th percentile or 90th percentile of the difference ratio between the initial test result and the retest result in historical sample records. For example, the preset stable ratio can be... The baseline operating environment temperature is set to 2%. The median of the operating environment temperature corresponding to the filtered sample records is then calculated as the baseline operating environment temperature. The median of the relative humidity corresponding to the operating environment of the filtered sample records is then calculated as the baseline operating environment relative humidity. The 95th percentile of the absolute value of the temperature deviation in the filtered sample records is then calculated as the normalized baseline value for temperature deviation. The 95th percentile of the absolute value of the relative humidity deviation in the filtered sample records is then calculated as the normalized baseline value for relative humidity deviation. The 95th percentile of the dust concentration in the filtered sample records is then calculated as the normalized baseline value for dust concentration. The normalized baseline value for vibration intensity can be set to 5.

[0074] After converting the cumulative effect over time, the cumulative effect of the environment, and the hierarchical number into dimensionless data, the state drift index is calculated using the following formula:

[0075] ;

[0076] in, Indicates the first The state drift index corresponding to each subsample identifier ranges from 0 to 1; This represents the upper limit of the time-cumulative effect, in minutes, and is determined by the 95th percentile of the time-cumulative effect in historical sample records. This represents the upper limit of the cumulative environmental impact, expressed in minutes. The upper limit of the cumulative environmental impact is determined by the 95th percentile of the cumulative environmental impact in historical sample records. Indicates the first Each subsample is identified by its corresponding hierarchical number; This represents the upper limit of the reference hierarchy, determined by the maximum value or the 95th percentile of the hierarchy number in the historical sample records; , , These represent the cumulative effect weight over time, the cumulative effect weight over the environment, and the effect weight at the same level, respectively, and the sum of these weights is 1.

[0077] The weights of time cumulative effect, environmental cumulative effect, and homologous hierarchy effect are determined based on the degree of influence of the proportion of differences between the time cumulative effect, environmental cumulative effect, and hierarchy number and the initial and retest results in the historical sample records. The degree of influence is calculated by the change in the proportion of differences in the test results after single-factor grouping. The larger the change, the greater the weight. For example, the weight of time cumulative effect can be set to 0.35, the weight of environmental cumulative effect can be set to 0.45, and the weight of homologous hierarchy effect can be set to 0.20.

[0078] The preset threshold is determined through historical sample records. The state drift index corresponding to each historical sample record is calculated according to the preset sample state transition model. Then, the candidate thresholds are traversed in a manner that increases by 0.01 each time from 0.10 to 0.90. The proportion of sample records with known initial test results and retest results that are consistent under each candidate threshold is mistakenly marked as abnormal sub-samples, and the proportion of sample records with known initial test results and retest results that are inconsistent is identified as abnormal sub-samples. The candidate threshold with the highest mislabeling ratio and the highest identification ratio is selected as the preset threshold. The preset mislabeling ratio is determined according to the abnormal false alarm tolerance ratio preset in the application document. The abnormal false alarm tolerance ratio can be determined according to the false alarm management requirements of normal sub-samples in the historical sample records. For example, the preset mislabeling ratio can be set to 5%.

[0079] The state drift index corresponding to each subsample identifier is compared with a preset threshold. Subsamples with a state drift index exceeding the preset threshold are marked as abnormal subsamples, while subsamples with a state drift index not exceeding the preset threshold are marked as normal subsamples.

[0080] If abnormal subsamples exist, extract the segmentation, reduction, sample preparation, and retention operation records of the abnormal subsamples. Analyze the deviation in particle size uniformity of the abnormal subsample segmentation and the cumulative effect deviation of the sample preparation and retention environment to obtain the contribution of operational abnormality to the abnormal subsamples, including:

[0081] In one implementation, the system reads the abnormal sub-sample identifier whose state drift index exceeds a preset threshold, the parent sample identifier corresponding to the abnormal sub-sample identifier, the path code corresponding to the abnormal sub-sample identifier, the hierarchical number corresponding to the abnormal sub-sample identifier, and the segmentation and reduction sample preparation and retention operation record corresponding to the abnormal sub-sample identifier. Following the path code, the system sequentially locates the abnormal sub-sample identifier starting from the parent sample identifier. It then filters the segmentation and reduction sample preparation and retention operation records from the location path, selecting those with the operation type of segmentation or reduction. These filtered operation records are used as particle size deviation analysis records. The particle size deviation analysis records include sample particle size distribution data before segmentation and reduction and sub-sample particle size distribution data after segmentation and reduction. The sample particle size distribution data before and after segmentation and reduction are saved as extended fields of the segmentation and reduction sample preparation and retention operation records or associated with the segmentation and reduction particle size weighing record via the operation sequence number. The segmentation and reduction particle size weighing record includes the operation sequence number, the preceding sample identifier, the following sample identifier, the particle size range, the sample mass corresponding to the particle size range, the total sample mass, and the weighing time.

[0082] The particle size distribution data of the sample before segmentation and reduction is used to represent the mass ratio of the particle size range of the sample corresponding to the previous sample identifier before segmentation or reduction. The particle size distribution data of the sub-sample after segmentation and reduction is used to represent the mass ratio of the particle size range of the sub-sample corresponding to the subsequent sample identifier after segmentation or reduction. The mass ratio of the particle size range is the proportion of the sample mass within the same particle size range to the total sample mass, and the value range is 0 to 1. The sum of the mass ratios of all particle size ranges corresponding to the same sample identifier is 1. The particle size distribution data of the sample before segmentation and reduction and the particle size distribution data of the sub-sample after segmentation and reduction are recorded using the same particle size range template. The particle size range template is preset according to the material category. The particle size range template includes multiple continuous and non-overlapping particle size ranges. For example, the particle size range template can be set to less than 0.075 mm, 0.075 mm to 0.150 mm, 0.150 mm to 0.300 mm, 0.300 mm to 0.600 mm, 0.600 mm to 1.180 mm, and greater than 1.180 mm.

[0083] When the particle size distribution data of the sample before segmentation and the particle size distribution data of the sub-samples after segmentation are recorded using different particle size interval templates, all particle size boundaries in the particle size distribution data of the sample before segmentation and the sub-samples after segmentation are extracted. Then, a unified particle size interval template is formed according to the order of particle size boundaries from small to large. The mass ratio of the original particle size interval is allocated to the unified particle size interval template according to the length overlap ratio between the original particle size interval and the unified particle size interval. The length overlap ratio is the overlap length between the original particle size interval and the unified particle size interval divided by the length of the original particle size interval.

[0084] When the number of particle size deviation analysis records is greater than or equal to 1, the particle size distribution difference value is calculated for each particle size deviation analysis record, and then the average of all particle size distribution difference values ​​is obtained to obtain the particle size uniformity deviation of the segmentation and reduction. When calculating the particle size distribution difference value of a single particle size deviation analysis record, each uniform particle size interval in the uniform particle size interval template is read, and then the absolute value of the difference between the mass ratio of the sample in the same uniform particle size interval before segmentation and reduction and the mass ratio of the sub-sample in the same uniform particle size interval after segmentation and reduction is calculated. The absolute values ​​of the differences corresponding to all uniform particle size intervals are summed, and finally the summation result is multiplied by 0.5 to obtain the particle size distribution difference value of a single particle size deviation analysis record. The value range of the particle size uniformity deviation of segmentation and reduction is 0 to 1. The larger the particle size uniformity deviation of segmentation and reduction, the greater the difference in particle size distribution before and after segmentation or reduction.

[0085] From the segmentation and reduction sample preparation and retention operation records corresponding to the abnormal sub-sample identifiers, those operation types of segmentation and reduction sample preparation and retention operation records are selected as environmental deviation analysis records. The operation environment information corresponding to the abnormal sub-samples is extracted from the environmental deviation analysis records. The operation environment information includes the operation environment temperature, operation environment relative humidity, operation environment dust concentration, operation environment vibration intensity, and sample exposure status. The standard sample preparation and retention environment parameters are a set of environmental parameters pre-established according to material category and operation type. The standard sample preparation and retention environment parameters include the standard sample preparation and retention environment temperature, standard sample preparation and retention environment relative humidity, standard sample preparation and retention environment dust concentration, standard sample preparation and retention environment vibration intensity, standard sample preparation and retention sample exposure status, temperature deviation normalized reference value, relative humidity deviation normalized reference value, dust concentration deviation normalized reference value, vibration intensity deviation normalized reference value, and sample exposure status deviation normalized reference value.

[0086] The standard sample preparation and retention environmental parameters are determined through historical sample records. These historical sample records are from the same engineering project, the same material category, or the same testing management scope, and have completed witnessed sampling, submission for testing, retesting, and formed complete test results. Sample records with a difference ratio between the initial test result and the retest result less than a preset stability ratio are selected from the historical sample records. The preset stability ratio is determined according to the allowable ratio of test difference set in the application document. The allowable ratio of test difference can be determined based on the 75th percentile or the 90th percentile of the difference ratio between the initial test result and the retest result in the historical sample records. For example, the preset stability ratio can be set to 2%. Then, the median of the operating environment temperature, the median of the operating environment relative humidity, the median of the operating environment dust concentration, the median of the operating environment vibration intensity, and the mode of the sample exposure state corresponding to the sample preparation and retention operations in the selected sample records are calculated separately. These parameters are used as the standard sample preparation and retention environment temperature, standard sample preparation and retention environment relative humidity, standard sample preparation and retention environment dust concentration, standard sample preparation and retention environment vibration intensity, and standard sample preparation and retention sample exposure state for the corresponding material category and corresponding operation type.

[0087] The normalized reference values ​​for temperature deviation, relative humidity deviation, dust concentration deviation, vibration intensity deviation, and sample exposure state deviation were determined through historical sample records. The normalized reference value for temperature deviation was the 95th percentile of the absolute value of the difference between the operating environment temperature and the standard sample preparation environment temperature in the selected sample records. The normalized reference value for relative humidity deviation was the 95th percentile of the absolute value of the difference between the operating environment relative humidity and the standard sample preparation environment in the selected sample records. The normalized reference value for dust concentration deviation was the 95th percentile of the absolute value of the difference between the operating environment dust concentration and the standard sample preparation environment dust concentration in the selected sample records. The normalized reference value for vibration intensity deviation was the 95th percentile of the absolute value of the difference between the operating environment vibration intensity and the standard sample preparation environment vibration intensity in the selected sample records. The normalized reference value for sample exposure state deviation could be set to 1. When the calculated result of any deviation normalized reference value was 0, the corresponding deviation normalized reference value was set to 1.

[0088] When calculating the cumulative effect deviation of the sample preparation and retention environment, the deviation values ​​for operating environment temperature, relative humidity, dust concentration, vibration intensity, and sample exposure state are calculated separately for each environmental deviation analysis record. The operating environment temperature deviation is calculated as the absolute value of the difference between the actual operating environment temperature and the standard sample preparation and retention environment temperature, divided by the normalized reference value for temperature deviation. The operating environment relative humidity deviation is calculated as the absolute value of the difference between the actual operating environment relative humidity and the standard sample preparation and retention environment relative humidity, divided by the normalized reference value for relative humidity deviation. The operating environment dust concentration deviation is calculated as the absolute value of the difference between the actual operating environment dust concentration and the standard sample preparation and retention environment dust concentration, divided by the normalized reference value for dust concentration deviation. The operating environment vibration intensity deviation is calculated as the absolute value of the difference between the actual operating environment vibration intensity and the standard sample preparation and retention environment vibration intensity, divided by the normalized reference value for vibration intensity deviation. The sample exposure state deviation is calculated as the absolute value of the difference between the actual sample exposure state and the standard sample preparation and retention sample exposure state, divided by the normalized reference value for sample exposure state deviation.

[0089] When calculating the weighted environmental deviation value for each environmental deviation analysis record, the following steps are taken: multiplying the operating environment temperature deviation value by its temperature deviation weight; multiplying the operating environment relative humidity deviation value by its relative humidity deviation weight; multiplying the operating environment dust concentration deviation value by its dust concentration deviation weight; multiplying the operating environment vibration intensity deviation value by its vibration intensity deviation weight; and multiplying the sample exposure state deviation value by its sample exposure state deviation weight. These five products are then summed to obtain the weighted environmental deviation value for each record. Finally, the weighted environmental deviation value for each record is multiplied by the corresponding time interval to obtain the weighted environmental deviation value for the individual record. The cumulative environmental deviation value of each environmental deviation analysis record is calculated; then, the cumulative environmental deviation values ​​corresponding to all environmental deviation analysis records are summed, and the summation result is divided by the upper limit of the environmental deviation reference value to obtain the initial sample preparation and retention environmental cumulative effect deviation; when the initial sample preparation and retention environmental cumulative effect deviation is greater than 1, the sample preparation and retention environmental cumulative effect deviation is set to 1; when the initial sample preparation and retention environmental cumulative effect deviation is less than or equal to 1, the initial sample preparation and retention environmental cumulative effect deviation is used as the sample preparation and retention environmental cumulative effect deviation; the upper limit of the environmental deviation reference value is determined by the 95th percentile of the summation result of the cumulative environmental deviation values ​​in the historical sample records.

[0090] The weights for temperature deviation, relative humidity deviation, dust concentration deviation, vibration intensity deviation, and sample exposure state deviation are determined by the correlation between the degree to which each operational environmental information in historical sample records deviates from the standard sample preparation and retention environmental parameters and the proportion of difference between the initial test results and the retest results. The correlation is expressed using absolute correlation coefficients. The absolute correlation coefficients between the five environmental deviation values ​​and the proportion of difference between the initial test results and the retest results are calculated separately. Then, the five absolute correlation coefficients are summed. Finally, the absolute correlation coefficient corresponding to each environmental deviation value is divided by the sum of the five absolute correlation coefficients to obtain the corresponding deviation weight. When the sum of the five absolute correlation coefficients is 0, the weights for all five deviations are set to 0.20. For example, the weight for temperature deviation can be set to 0.25, the weight for relative humidity deviation can be set to 0.30, the weight for dust concentration deviation can be set to 0.15, the weight for vibration intensity deviation can be set to 0.10, and the weight for sample exposure state deviation can be set to 0.20.

[0091] The operational anomaly contribution of the abnormal subsample is obtained by weighted summing of the uniformity deviation of the segmentation and reduction particle size and the cumulative effect deviation of the sample preparation and retention environment. Specifically, the uniformity deviation of the segmentation and reduction particle size is multiplied by the weight of the uniformity deviation of the segmentation and reduction particle size, and the cumulative effect deviation of the sample preparation and retention environment is multiplied by the weight of the cumulative effect deviation of the sample preparation and retention environment. The two products are then summed to obtain the operational anomaly contribution of the abnormal subsample. The value of the operational anomaly contribution of the abnormal subsample ranges from 0 to 1. The larger the operational anomaly contribution of the abnormal subsample, the higher the degree of contribution of the segmentation, reduction, sample preparation and retention operation to the abnormal state of the abnormal subsample.

[0092] The weights for both the uniformity deviation of the particle size distribution and the cumulative effect deviation of the sample preparation and retention environment are greater than or equal to 0, and the sum of these two weights is 1. These weights are determined from historical sample records that have undergone anomaly cause verification. The method involves dividing the historical sample records into a particle size distribution anomaly cause record group, a sample preparation and retention environment anomaly cause record group, and a non-operational anomaly cause record group based on the anomaly cause verification results. The absolute value of the difference between the average particle size uniformity deviation of the particle size distribution anomaly cause record group and the non-operational anomaly cause record group is calculated as the particle size distribution uniformity deviation. The deviation discrimination value is calculated by taking the absolute value of the difference between the average value of the cumulative effect deviation of the sample preparation and retention environment between the abnormal sample preparation and retention environment cause record group and the non-operational abnormal sample preparation and retention environment cause record group. Then, the deviation discrimination value of the segmentation and reduction particle size uniformity deviation and the deviation discrimination value of the cumulative effect deviation of the sample preparation and retention environment are normalized into the segmentation and reduction particle size uniformity deviation weight and the cumulative effect deviation weight of the sample preparation and retention environment. When the sum of the two discrimination values ​​is 0, the segmentation and reduction particle size uniformity deviation weight and the cumulative effect deviation weight of the sample preparation and retention environment are both set to 0.50. For example, the segmentation and reduction particle size uniformity deviation weight can be set to 0.55 and the cumulative effect deviation weight of the sample preparation and retention environment can be set to 0.45.

[0093] The dispersion of test results between normal subsamples and parent samples, along with the cumulative effect of the delivery and transportation environment, was analyzed to determine the representativeness of the deviation of normal subsample results, including:

[0094] In one implementation, each subsample identifier, the parent sample identifier corresponding to each subsample identifier, the path code corresponding to each subsample identifier, and the hierarchical number corresponding to each subsample identifier are read from the homology set. Then, the state drift index and preset threshold corresponding to each subsample identifier are read. Subsample identifiers whose state drift index does not exceed the preset threshold are identified as normal subsample identifiers. Subsamples corresponding to normal subsample identifiers are identified as normal subsamples. Normal subsample identifiers are used as query conditions for obtaining normal subsample test results and delivery and transportation records. Parent sample identifiers are used as query conditions for obtaining parent sample test results.

[0095] The test results of normal subsamples and parent samples are read from the test result records respectively. The test result records are associated with the homology set through the sample identifier. The test results of normal subsamples are obtained by querying the normal subsample identifier, and the test results of parent samples are obtained by querying the parent sample identifier. The test result record includes the sample identifier, test item name, test item value, testing unit, testing time, testing personnel identity information, test result record number, and test result entry time. The test results of normal subsamples are used to represent the numerical test data obtained by normal subsamples under the test items, and the test results of parent samples are used to represent the test data obtained by parent samples under the same test items. The test items are preset according to the material category. For example, the test items may include compressive strength, moisture content, density, fineness, setting time, or chloride ion content.

[0096] After obtaining the test results of the normal subsample and the parent sample, the test results of the normal subsample and the parent sample are matched according to the test item name and testing unit. The dispersion of the results is calculated only for test items with the same test item name and testing unit. When calculating the dispersion of an individual result, the absolute value of the difference between the test item value in the normal subsample test result and the test item value in the parent sample test result is calculated. Then, the absolute value of the difference is divided by the absolute value of the test item value in the parent sample test result to obtain the dispersion of the individual result. When the test item value in the parent sample test result is equal to 0, the 5th percentile of the absolute value of the same test item value in historical sample records of the same material category is used as the substitute denominator. When the substitute denominator is equal to 0, the substitute denominator is set to 1 to avoid division by zero. When two or more test items are successfully matched in the same normal subsample, the dispersion of all individual results is weighted and summed according to the test item weights to obtain the dispersion of the normal subsample test results and the parent sample test results. The test item weights are determined by the absolute correlation coefficient between the proportion of differences between each test item and the initial test result and the retest result in the historical sample records. The larger the absolute correlation coefficient, the larger the corresponding test item weight, and the sum of the weights of all test items is 1. For example, the test item weight corresponding to compressive strength can be set to 0.40, the test item weight corresponding to moisture content can be set to 0.25, the test item weight corresponding to density can be set to 0.20, and the test item weight corresponding to fineness can be set to 0.15.

[0097] The sample transport record is used to document the transportation of normal subsamples from the sampling site, sample preparation location, or sample retention location to the testing location. The transport record is associated with the normal subsample identifier, sample container identifier, or batch number. The transport record includes the normal subsample identifier, parent sample identifier, sample container identifier, batch number, start time, arrival time, transport personnel identification information, transport route number, transport environment information, and receipt confirmation information. When multiple transport records are associated with the same normal subsample identifier, they are sorted from earliest to latest according to their start time. Transport records with consecutive times or identical sample container identifiers are merged into a transport record sequence. The impact of transport time and transport environment information is then calculated segment by segment within the transport record sequence.

[0098] The delivery and transportation time is the time difference between the delivery arrival time and the delivery start time, expressed in minutes. Delivery and transportation environment information includes ambient temperature, relative humidity, vibration intensity, container opening status, and delivery delay status. Ambient temperature is expressed in degrees Celsius, relative humidity in percentages, vibration intensity using integers from 0 to 5, container opening status (0 for container not open, 1 for open), and delivery delay status (0 for delivery time not exceeding the preset delivery time, 1 for exceeding the preset delivery time). The preset delivery time is determined by historical sample records of the same material category where the difference between the initial and retest results is less than a preset stable ratio. The preset delivery time is taken as the 90th or 95th percentile of the delivery time in the selected sample records; for example, the preset delivery time can be set to 240 minutes.

[0099] When calculating the cumulative effect of the transport environment for inspection, standard transport environment parameters are established. These parameters include standard transport environment temperature, standard transport environment relative humidity, standard transport environment vibration intensity, standard transport container open state, standard transport delay state, normalized reference value for transport temperature deviation, normalized reference value for transport relative humidity deviation, normalized reference value for transport vibration intensity deviation, normalized reference value for transport container open state deviation, and normalized reference value for transport delay state deviation. The standard transport environment parameters for testing are determined through historical sample records. These historical sample records are from the same project, the same material category, or the same testing management scope, and have completed witnessed sampling, testing, retesting, and formed complete test results. First, historical sample records with a difference ratio of less than the preset stable ratio for initial and retest results are selected. Then, the median of transport environment temperature, the median of transport environment relative humidity, the median of transport environment vibration intensity, the mode of transport container open state, and the mode of transport delay state in the selected historical sample records are taken as the corresponding standard values. The normalized reference value for each deviation is taken as the 95th percentile of the absolute value of the difference between the corresponding transport environment field and the corresponding standard value in the selected historical sample records. When any deviation normalized reference value is equal to 0, the corresponding deviation normalized reference value is set to 1.

[0100] Based on the submitted transportation environment information, calculate the following deviation values: transportation temperature deviation, transportation relative humidity deviation, transportation vibration intensity deviation, transportation container opening status deviation, and transportation delay status deviation. The transportation temperature deviation is calculated as the absolute value of the difference between the transportation ambient temperature and the standard transportation ambient temperature, divided by the normalized reference value. Similarly, the transportation relative humidity deviation is calculated as the absolute value of the difference between the transportation ambient relative humidity and the standard transportation ambient relative humidity, divided by the normalized reference value. The transportation vibration intensity deviation is calculated as the absolute value of the difference between the transportation ambient vibration intensity and the standard transportation ambient vibration intensity, divided by the normalized reference value. The transportation container opening status deviation is calculated as the absolute value of the difference between the transportation container opening status and the standard transportation container opening status, divided by the normalized reference value. Finally, the transportation delay status deviation is calculated as the difference between the transportation delay status and the standard transportation delay status. The absolute value of the difference between delay states is divided by the normalized baseline value of the transport delay state deviation. Then, the five deviation values ​​are multiplied by the corresponding weights of transport temperature, transport relative humidity, transport vibration intensity, transport container opening status, and transport delay state, and summed to obtain the weighted value of the transport environment deviation. The weighted value of the transport environment deviation is then multiplied by the delivery transport time to obtain the cumulative value of the transport environment. The cumulative value of the transport environment is then divided by the upper limit of the transport environment reference value to obtain the initial cumulative effect of the transport environment. If the initial cumulative effect of the transport environment is greater than 1, the cumulative effect of the transport environment is set to 1. If the initial cumulative effect of the transport environment is less than or equal to 1, the initial cumulative effect of the transport environment is used as the cumulative effect of the transport environment. The upper limit of the transport environment reference value is determined by the 95th percentile of the cumulative value of the transport environment in the historical sample records.

[0101] The weights for the influence of transportation temperature, relative humidity, vibration intensity, container opening status, and delay status are determined by the absolute correlation coefficients between the proportions of differences between the initial and re-inspection results of each transportation environmental deviation value in historical sample records. First, the absolute correlation coefficients between the five transportation environmental deviation values ​​and the proportions of differences between the initial and re-inspection results are calculated separately. Then, the five absolute correlation coefficients are summed. Finally, the absolute correlation coefficient corresponding to each transportation environmental deviation value is divided by the sum of the five absolute correlation coefficients to obtain the corresponding transportation environmental influence weight. When the sum of the five absolute correlation coefficients is 0, the weights for all five transportation environmental influences are set to 0.20. For example, the weight for transportation temperature can be set to 0.25, the weight for relative humidity can be set to 0.25, the weight for vibration intensity can be set to 0.20, the weight for container opening status can be set to 0.15, and the weight for delay status can be set to 0.15.

[0102] The deviation of the normal subsample from the representativeness of the parent sample is obtained by weighted summing of the dispersion of the test results of the normal subsample and the cumulative effect of the transport environment. Specifically, the result dispersion is multiplied by the result dispersion weight, and the cumulative effect of the transport environment is multiplied by the cumulative effect weight of the transport environment. The two products are then summed to obtain the deviation of the normal subsample from the representativeness of ... The result dispersion weight and the cumulative effect weight of the transportation environment are determined through historical sample records. The determination method is as follows: historical sample records are divided into a group of records with consistent initial and re-inspection results and a group of records with inconsistent initial and re-inspection results. The absolute value of the difference between the average result dispersion values ​​of the consistent and inconsistent record groups is used as the result dispersion discrimination value. The absolute value of the difference between the average value of the cumulative effect of the transportation environment of the initial and re-inspection results is used as the cumulative effect discrimination value of the transportation environment of the transportation environment. Then, the result dispersion discrimination value and the cumulative effect discrimination value of the transportation environment of the transportation environment are normalized to the result dispersion weight and the cumulative effect weight of the transportation environment of the transportation environment. When the sum of the two discrimination values ​​is 0, the result dispersion weight and the cumulative effect weight of the transportation environment of the transportation environment are both set to 0.50. For example, the result dispersion weight can be set to 0.60 and the cumulative effect weight of the transportation environment of the transportation environment can be set to 0.40.

[0103] Based on the contribution of operational anomalies to abnormal subsamples and the representativeness of result deviations in normal subsamples, identify the target tracing links in the homology set that cause inconsistencies between the initial and retest results, including:

[0104] In one implementation, the parent sample identifier, all sub-sample identifiers, path codes, and hierarchical numbers corresponding to the homology set are read. Then, the uniformity deviation of the segmentation and reduction granularity, the cumulative effect deviation of the sample preparation and retention environment, and the contribution of operational anomalies corresponding to each abnormal sub-sample identifier are read. The result dispersion, the cumulative effect of the delivery and transportation environment, and the result deviation representativeness corresponding to each normal sub-sample identifier are also read. The target traceability link is used to represent the process link within the homology set that has priority traceability value for inconsistencies between the initial test results and the retest results. The value of the target traceability link includes the segmentation and reduction link, the sample preparation and retention link, the delivery and transportation link, and the witness sampling link.

[0105] When there are two or more anomalous sub-sample identifiers in the homology set, compare the operational anomaly contributions corresponding to all anomalous sub-sample identifiers, and take the largest operational anomaly contribution as the set-level operational anomaly contribution, and take the anomalous sub-sample identifier corresponding to the set-level operational anomaly contribution as the dominant anomalous sub-sample identifier; when two or more anomalous sub-sample identifiers correspond to the same maximum operational anomaly contribution, select the anomalous sub-sample identifier with the larger state drift index as the dominant anomalous sub-sample identifier. If the state drift indices are still the same, select the anomalous sub-sample identifier with the smaller hierarchical number as the dominant anomalous sub-sample identifier. If the hierarchical numbers are still the same, select the anomalous sub-sample identifier whose path code is arranged first in character order as the dominant anomalous sub-sample identifier; when there is only one anomalous sub-sample identifier in the homology set, take the operational anomaly contribution corresponding to that one anomalous sub-sample identifier as the set-level operational anomaly contribution, and take that one anomalous sub-sample identifier as the dominant anomalous sub-sample identifier; when there are no anomalous sub-sample identifiers in the homology set, set-level operational anomaly contribution is set to 0.

[0106] When there are two or more normal subsample identifiers in the homology set, the result deviation representativeness corresponding to all normal subsample identifiers is compared. The result deviation representativeness with the largest value is taken as the set-level result deviation representativeness, and the normal subsample identifier corresponding to the set-level result deviation representativeness is taken as the dominant normal subsample identifier. When two or more normal subsample identifiers correspond to the same maximum result deviation representativeness, the normal subsample identifier with the larger result dispersion is selected as the dominant normal subsample identifier. If the result dispersion is still the same, the normal subsample identifier with the greater cumulative effect of the delivery and transportation environment is selected as the dominant normal subsample identifier. If the cumulative effect of the delivery and transportation environment is still the same, the normal subsample identifier with the path code arranged first in character order is selected as the dominant normal subsample identifier. When there is only one normal subsample identifier in the homology set, the result deviation representativeness corresponding to the one normal subsample identifier is taken as the set-level result deviation representativeness, and the one normal subsample identifier is taken as the dominant normal subsample identifier. When there are no normal subsample identifiers in the homology set, the set-level result deviation representativeness is set to 0.

[0107] The preset contribution threshold is determined through historical sample records that have completed the target traceability process. These historical sample records are from the same project, the same material category, or the same testing management scope, and have completed witnessed sampling, testing, and retesting, forming complete traceability confirmation conclusions. The complete traceability confirmation conclusions are used to mark the steps in the historical sample records that lead to inconsistencies between the initial and retest results. These conclusions are formed by comparing the verification records, testing and re-inspection records, transportation and handover records, segmentation and reduction sample preparation and retention operation records, and witnessed sampling records in the historical sample records, and are saved as fields in the historical sample records. When determining the preset contribution threshold, the operational anomaly contribution corresponding to each abnormal sub-sample identifier in the historical sample records is calculated. Then, according to the set of homologous relationships, the set-level operational anomaly contribution corresponding to the historical sample records is determined, confirming the target traceability step as a segmentation and reduction loop. Historical sample records from the sample preparation or retention stage are designated as the operation anomaly record group, while historical sample records from the target traceability stage, which is identified as the delivery and transportation stage or the witnessed sampling stage, are designated as the non-operation anomaly record group. Candidate contribution thresholds are iterated through from 0.10 to 0.90, increasing by 0.01 each time. The recognition rate of each candidate contribution threshold for the operation anomaly record group and the false recognition rate for the non-operation anomaly record group are statistically analyzed. The candidate contribution threshold with the highest recognition rate and a false recognition rate not exceeding the preset contribution threshold is selected as the preset contribution threshold. The preset contribution threshold is determined based on the permissible false alarm ratio for operation anomalies pre-set in the application document. For example, the preset contribution threshold can be set to 5%. When multiple candidate contribution thresholds correspond to the same recognition ratio, the candidate contribution threshold with the smaller value is selected as the preset contribution threshold.

[0108] The preset deviation threshold is determined through historical sample records that have completed the target traceability process. These historical sample records are from the same project, material category, or testing management scope, and include records of witnessed sampling, delivery for testing, re-testing, and complete traceability confirmation conclusions. The complete traceability confirmation conclusions are formed by comparing the verification records, testing and re-testing records, transportation handover records, segmentation and reduction sample preparation and retention operation records, and witnessed sampling records from the historical sample records, and are saved as fields in the historical sample records. When determining the preset deviation threshold, the representativeness of the result deviation corresponding to each normal sub-sample identifier in the historical sample records is calculated. Then, according to the set of homologous relationships, the set-level representativeness of the result deviation corresponding to the historical sample records is determined separately. Historical sample records where the target traceability process is confirmed as the delivery and transportation process are used as the transportation deviation threshold. For the non-transportation deviation record group, historical sample records from the target traceability stage, such as the witness sampling stage, the segmentation and reduction stage, or the sample preparation and retention stage, are designated as the non-transportation deviation record group. Candidate deviation thresholds are iterated through from 0.10 to 0.90, increasing by 0.01 each time. The recognition rate of each candidate deviation threshold for the transportation deviation record group and the false recognition rate for the non-transportation deviation record group are calculated. The candidate deviation threshold with the highest recognition rate and a false recognition rate not exceeding the preset deviation false recognition rate is selected as the preset deviation threshold. The preset deviation false recognition rate is determined based on the transportation deviation false alarm tolerance rate pre-set in the application document. For example, the preset deviation false recognition rate can be set to 5%. When multiple candidate deviation thresholds correspond to the same recognition rate, the smaller candidate deviation threshold is selected as the preset deviation threshold.

[0109] The contribution of ensemble-level operational anomalies is compared with a preset contribution threshold, and the deviation of ensemble-level results from representativeness is compared with a preset deviation threshold. When the contribution of ensemble-level operational anomalies is greater than the preset contribution threshold, the uniformity deviation of segmentation granularity and the cumulative effect deviation of sample preparation and retention environment corresponding to the dominant anomaly subsample identifier are read, and the magnitudes of the uniformity deviation of segmentation granularity and the cumulative effect deviation of sample preparation and retention environment are compared. When the uniformity deviation of segmentation granularity is greater than the cumulative effect deviation of sample preparation and retention environment, the segmentation step is identified as the target traceability step. When the cumulative effect deviation of sample preparation and retention environment is greater than the uniformity deviation of segmentation granularity, the sample preparation and retention step is identified as the target traceability step.

[0110] When the contribution of abnormal operation at the set level is not greater than the preset contribution threshold and the deviation of the set level result from the representativeness is greater than the preset deviation threshold, the inspection and transportation link will be identified as the target traceability link.

[0111] When the contribution of the aggregate-level operation anomaly is not greater than the preset contribution threshold and the deviation of the aggregate-level result from the representativeness threshold is not greater than the preset deviation threshold, the witness sampling step will be identified as the target tracing step.

[0112] Based on the target tracing stage, the parent sample identifier and child sample identifier associated with the target tracing stage are traced back in the homology set to generate a full-process tracing chain, including:

[0113] In one implementation, the end-to-end traceability chain is used to express the source path from the parent sample identifier to the target sub-sample identifier according to the parent-child relationship, and to mark the target traceability link that causes the inconsistency between the initial test result and the retest result. The parent-child relationship includes direct parent-child relationship and indirect parent-child relationship. The direct parent-child relationship means that the parent sample identifier and the sub-sample identifier are adjacent to each other in the same segmentation and reduction sample preparation and retention operation record. The indirect parent-child relationship means that the parent sample identifier is traced back to the target sub-sample identifier through one or more intermediate sub-sample identifiers.

[0114] When the target tracing step is a segmentation and reduction step or a sample preparation and retention step, the dominant abnormal sub-sample identifier is read from the target tracing step judgment result. The dominant abnormal sub-sample identifier is used as the abnormal sub-sample identifier. Then, the path code corresponding to the abnormal sub-sample identifier is queried from the homology relationship set according to the abnormal sub-sample identifier. According to the path code, the parent sample identifier located at the starting point of the path code, the abnormal sub-sample identifier located at the ending point of the path code, the intermediate sub-sample identifier located between the parent sample identifier and the abnormal sub-sample identifier, and the parent-child relationship between adjacent sample identifiers are extracted. The extracted parent sample identifier, abnormal sub-sample identifier, intermediate sub-sample identifier, and parent-child relationship together constitute the chain basic data corresponding to the segmentation and reduction step or the sample preparation and retention step.

[0115] When the target traceability step is the segmentation and reduction step, the segmentation and reduction particle size uniformity deviation and particle size deviation analysis record number are read from the target traceability step judgment result. Then, the parent-child relationship with operation type segmentation or reduction and matching particle size deviation analysis record number is extracted from the chain basic data corresponding to the segmentation and reduction step. The matched parent-child relationship is marked as the key parent-child relationship, and the operation sequence number, operation time, pre-sample identifier, and post-sample identifier corresponding to the key marked parent-child relationship are read. When the target traceability step is the sample preparation and retention step, the sample preparation and retention environment cumulative effect deviation and environmental deviation analysis record number are read from the target traceability step judgment result. Then, the parent-child relationship with operation type sample preparation or retention and matching environmental deviation analysis record number is extracted from the chain basic data corresponding to the sample preparation and retention step. The matched parent-child relationship is marked as the key parent-child relationship, and the operation sequence number, operation time, pre-sample identifier, and post-sample identifier corresponding to the key marked parent-child relationship are read.

[0116] When the target traceability link is the delivery and transportation link, the dominant normal sub-sample identifier is read from the target traceability link determination result. The dominant normal sub-sample identifier is used as the normal sub-sample identifier. Then, the path code corresponding to the normal sub-sample identifier is queried from the homology relationship set according to the normal sub-sample identifier. According to the path code, the parent sample identifier at the starting point of the path code, the normal sub-sample identifier at the ending point of the path code, the intermediate sub-sample identifier between the parent sample identifier and the normal sub-sample identifier, and the parent-child relationship between adjacent sample identifiers are extracted. The delivery and transportation record number, delivery and transportation time, cumulative effect of delivery and transportation environment, and result deviation representativeness corresponding to the normal sub-sample identifier are read. The extracted parent sample identifier, normal sub-sample identifier, intermediate sub-sample identifier, parent-child relationship, and delivery and transportation record number together constitute the basic chain data corresponding to the delivery and transportation link.

[0117] When the target tracing stage is the witness sampling stage, the parent sample identifier is read from the target tracing stage judgment result. Then, the parent sample identifier, witness information, and binding operation record bound to the witness information are queried from the same source relationship set according to the parent sample identifier. All child sample identifiers that can be reached step by step through the parent-child relationship starting from the parent sample identifier are extracted from the same source relationship set. The extraction method is to first extract the first layer of child sample identifiers with the parent sample identifier as the parent sample identifier, and then use the first layer of child sample identifiers as the new parent sample identifier to continue extracting the next layer of child sample identifiers. The extraction is carried out layer by layer in the same way until there are no more next layer of child sample identifiers. The extracted parent sample identifier, witness information, binding operation record, all child sample identifiers, and all parent-child relationship relationships together constitute the basic chain data corresponding to the witness sampling stage.

[0118] When concatenating the extracted parent sample identifier and child sample identifier according to the parent-child relationship, the parent sample identifier is set as the starting point of the chain. Then, child sample identifiers are added sequentially from the parent sample identifier to the child sample identifier according to the direction of the parent-child relationship. When adding child sample identifiers, they are arranged in ascending order of hierarchical number. When there are two or more child sample identifiers at the same level, they are arranged in ascending order of operation time corresponding to the parent-child relationship. If the operation time is the same, they are arranged in ascending order of operation sequence number. If the operation sequence number is the same, they are arranged in the character order of the child sample identifier. When there are branches in the concatenation result, a sub-chain is generated for each path from the parent sample identifier to the end child sample identifier. All sub-chains are combined to form the full-process traceability chain. The end child sample identifier indicates the child sample identifier that no longer appears as the parent sample identifier in the parent-child relationship.

[0119] When labeling target traceability links based on the cascade results, the name of the target traceability link is written into the cascade results, and the labeling position is determined according to the name of the target traceability link. When the target traceability link is a segmentation and reduction link, the parent-child relationship is highlighted as the target traceability link relationship, and the segmentation and reduction particle size uniformity deviation and particle size deviation analysis record number are written into the target traceability link relationship. When the target traceability link is a sample preparation and retention link, the parent-child relationship is highlighted as the target traceability link relationship, and the cumulative effect deviation of the sample preparation and retention environment and the environmental deviation analysis record number are written into the target traceability link relationship. When the target traceability link is a delivery and transportation link, the delivery and transportation record number corresponding to the normal subsample identifier is marked as the target traceability link relationship record, and the delivery and transportation time, cumulative effect of the delivery and transportation environment, and result deviation representativeness are written into the target traceability link relationship record. When the target traceability link is a witness sampling link, the parent sample identifier bound to the witness information is marked as the target traceability link relationship node, and the witness information and binding operation record are written into the target traceability link relationship node.

[0120] When generating a full-process traceability chain, the concatenation results of the completed target traceability links are saved as the full-process traceability chain. The full-process traceability chain includes at least the parent sample identifier, child sample identifier, parent-child relationship, path code, hierarchical number, target traceability link, target traceability link labeling position, target traceability link associated data, witness information, binding operation record, and chain generation time. When the target traceability link is a segmentation and reduction link, the associated data includes the segmentation and reduction granularity uniformity deviation and granularity deviation analysis record number. When the target traceability link is a sample preparation and retention link, the associated data includes the cumulative effect deviation of the sample preparation and retention environment and the environmental deviation analysis record number. When the target traceability link is a delivery and transportation link, the associated data includes the delivery and transportation record number, delivery and transportation time, cumulative effect of the delivery and transportation environment, and result deviation representativeness. When the target traceability link is a witness sampling link, the associated data includes witness information and binding operation record.

[0121] The above embodiments are only used to illustrate the technical methods of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical methods of the present invention without departing from the spirit and scope of the technical methods of the present invention.

Claims

1. A traceability and control system for the entire process of witness sampling, testing, and re-testing of building materials, characterized in that: include: Sample source construction module: acquires the parent sample identifier, witnessing information and segmentation and reduction sample preparation and retention operation records recorded during the witness sampling process of building materials, and constructs a set of homology relationships between the parent sample and the sub-samples; Drift determination module: For each sub-sample in the homology set, the preset sample state transition model is used to calculate the state drift index of the sub-sample relative to the parent sample, and to determine whether there are abnormal sub-samples whose state drift index exceeds the preset threshold. Anomaly Contribution Module: If there are abnormal sub-samples, extract the segmentation, reduction, sample preparation, and retention operation records of the abnormal sub-samples, analyze the deviation of the uniformity of the segmentation and reduction particle size of the abnormal sub-samples and the cumulative effect deviation of the sample preparation and retention environment, and obtain the operational anomaly contribution of the abnormal sub-samples. Deviation Representativeness Module: Analyzes the dispersion of the test results of normal subsamples and the test results of the parent sample, as well as the cumulative effect of the delivery and transportation environment, to obtain the deviation representativeness of the results of normal subsamples; Link identification module: Based on the contribution of abnormal sub-samples to operational abnormalities and the representativeness of normal sub-samples to result deviations, identify the target tracing link in the homology set that causes inconsistencies between the initial test results and the retest results; Traceability Chain Module: Based on the target traceability link, trace back the parent sample identifier and sub-sample identifier associated with the target traceability link in the set of homologous relationships to generate a full-process traceability chain.

2. The full-process traceability and control system for witness sampling, testing, and re-testing of building materials according to claim 1, characterized in that, Obtain the parent sample identifier, witnessing information, and segmentation and reduction sample preparation and retention operation records recorded during the witnessed sampling process of building materials, and construct a set of homology relationships between the parent sample and the sub-samples, specifically: Starting with the parent sample identifier, analyze the preceding and following sample identifiers generated by each segmentation, reduction, sample preparation, and retention operation in the segmentation, reduction, sample preparation, and retention operation record. Establish a parent-child relationship between the pre-sample identifier and the post-sample identifier, and link the parent-child relationships step by step starting from the parent sample identifier. Bind the witness information to the parent sample identifier to generate a set of homologous relationships that includes the parent sample identifier, the child sample identifier and the corresponding parent-child relationship.

3. The full-process traceability and control system for witness sampling, testing, and re-testing of building materials according to claim 1, characterized in that, For each sub-sample in the homology set, a preset sample state transition model is used to calculate the state drift index of the sub-sample relative to the parent sample. This is used to determine whether there are any abnormal sub-samples whose state drift index exceeds a preset threshold. Specifically: Extract the operation time and operation environment information from the segmentation and reduction sample preparation and retention operation records corresponding to each sub-sample from the homology set, and input the operation time and operation environment information into the preset sample state transition model. The preset sample state transition model calculates the cumulative effect of time based on the operation time, and the cumulative effect of environment based on the operation environment information. Then, it calculates the state drift index by combining the homology between the subsample and the parent sample. The state drift index is compared with a preset threshold, and subsamples whose state drift index exceeds the preset threshold are marked as abnormal subsamples.

4. The full-process traceability and control system for witness sampling, testing, and re-testing of building materials according to claim 1, characterized in that, If abnormal subsamples exist, extract the segmentation, reduction, sample preparation, and retention operation records of the abnormal subsamples. Analyze the deviation in particle size uniformity of the abnormal subsamples and the cumulative effect of the sample preparation and retention environment to obtain the operational anomaly contribution of the abnormal subsamples. Specifically: For sub-samples whose state drift index exceeds a preset threshold, extract the particle size distribution data of the sample before segmentation and the particle size distribution data of the sub-sample after segmentation from the segmentation and reduction sample preparation and retention operation record, and calculate the segmentation and reduction particle size uniformity deviation. By comparing the operational environment information corresponding to the abnormal sub-samples in the operation environment information recorded in the segmentation and reduction sample preparation and retention operation record with the standard sample preparation and retention environment parameters, the cumulative effect deviation of the sample preparation and retention environment is obtained. The operational anomaly contribution of abnormal subsamples is obtained by weighted summing of the deviation in uniformity of particle size reduction and the cumulative effect of the sample preparation and retention environment.

5. The full-process traceability and control system for witness sampling, testing, and re-testing of building materials according to claim 1, characterized in that, The dispersion of the test results of normal subsamples compared with those of the parent sample, and the cumulative effect of the delivery and transportation environment, were analyzed to obtain the representativeness of the deviation of the normal subsample results, specifically: Subsamples whose state drift index does not exceed a preset threshold are extracted from the set of homology relationships and designated as normal subsamples. Obtain the test results of normal subsamples and the test results of the parent sample, calculate the ratio of the absolute value of the difference between the test results of normal subsamples and the test results of the parent sample to the test results of the parent sample, and obtain the result dispersion of the test results of normal subsamples and the test results of the parent sample. Extract the inspection and transportation time and environment information from the inspection and transportation records, and calculate the cumulative effect of the inspection and transportation environment based on the inspection and transportation time and environment information. The deviation of the normal subsample from the parent sample is obtained by weighting and summing the dispersion of the test results between the normal subsample and the parent sample, as well as the cumulative effect of the transportation environment.

6. The full-process traceability and control system for witness sampling, testing, and re-testing of building materials according to claim 1, characterized in that, Based on the operational anomaly contribution of abnormal subsamples and the representativeness of result deviations in normal subsamples, the target tracing link in the homology set that leads to inconsistencies between the initial and retest results is identified, specifically: Compare the operational anomaly contribution of abnormal subsamples with a preset contribution threshold, and compare the deviation of normal subsamples from representativeness with a preset deviation threshold. When the contribution of abnormal operation of abnormal subsamples exceeds the preset contribution threshold, the segmentation and reduction process or the sample preparation and retention process is identified as the target traceability process based on the magnitude of the deviation in uniformity of segmentation and reduction particle size and the cumulative effect deviation of the sample preparation and retention environment. When the contribution of abnormal operation to abnormal sub-samples is not greater than the preset contribution threshold and the deviation of normal sub-samples from representativeness is greater than the preset deviation threshold, the delivery and transportation link will be identified as the target traceability link. When the contribution of abnormal operation to abnormal sub-samples is not greater than the preset contribution threshold and the deviation of normal sub-samples from representativeness is not greater than the preset deviation threshold, the witness sampling step will be identified as the target traceability step.

7. The full-process traceability and control system for witness sampling, testing, and re-testing of building materials according to claim 1, characterized in that, Based on the target tracing step, the parent sample identifier and child sample identifier associated with the target tracing step are traced back in the homology set to generate a full-process tracing chain, specifically: When the target tracing link is the segmentation and reduction link or the sample preparation and retention link, extract the abnormal sub-samples and the parent sample identifiers that have a parent-child relationship with the abnormal sub-samples from the homology set. When the target traceability link is the delivery and transportation link, extract the normal sub-sample and the parent sample identifier that has a parent-child relationship with the normal sub-sample from the homology set; When the target tracing step is the witness sampling step, extract the parent sample identifier that is bound to the witness information and all child sample identifiers that have a parent-child relationship with the parent sample identifier that is bound to the witness information from the same source relationship set; The extracted parent sample identifier and child sample identifier are linked together according to the parent-child relationship, and the target traceability link is marked according to the linking result to generate a full-process traceability chain.

Images