A crew post allocation method and device based on information matching

Abstract

The application discloses a crew post distribution method and device based on information matching, and the method comprises the following steps: obtaining a crew post distribution historical information set; the crew post distribution historical information set comprises a plurality of crew post distribution historical vectors and corresponding crew post equipment operation state information sets; the crew post equipment operation state information set comprises an operation state value sub-set of each crew post equipment; the operation state value sub-set comprises an operation value sequence of each technical index of the crew post equipment; the crew post distribution historical information set is subjected to modeling processing to obtain a crew post information matching model; the crew post information matching model is solved to obtain a crew post distribution vector; and the crew post distribution vector comprises crew information engaged in each post.

Description

Technical Field

[0001] This invention relates to the fields of industrial data processing, strategy optimization, and data processing technology, specifically to a method and apparatus for assigning crew positions based on information matching. Background Technology

[0002] Crew assignment is a core aspect of ship operation and management, and its rationality directly impacts ship operational efficiency, equipment safety, and crew work effectiveness. In traditional ship operations, crew assignment largely relies on the experience of management personnel, a method with significant limitations: Firstly, human experience is heavily influenced by subjective factors, making it difficult to comprehensively assess the precise match between crew members' skills, experience, and job requirements, easily leading to mismatches between crew capabilities and job requirements, resulting in low work efficiency. Secondly, traditional assignment methods often ignore long-term operational status feedback from equipment at crew positions, failing to correlate key information such as equipment stability and failure rates with job assignments. Improper assignment may lead to excessive equipment wear and tear or frequent malfunctions, increasing ship operating costs and safety risks.

[0003] With the development of data-driven technologies, some ship operations have begun to try using simple data models to assist in job allocation, but significant shortcomings still exist. First, existing data-driven methods lack a systematic approach to processing raw data, failing to address the characteristics of crew job allocation-related data (such as missing values, noise in sensor-collected equipment operation data, and interference from occasional outliers), resulting in inconsistent data quality input to the model and directly affecting the reliability of the model's output. Second, existing models mostly focus on one-way matching between crew members and positions, failing to construct a correlation model between crew allocation, job requirements, and equipment operating status, thus failing to form a closed loop of "allocation-operational feedback-optimized allocation," making it difficult to continuously improve the adaptability of job allocation schemes. Finally, existing models have a single dimension for evaluating equipment operating status, judging equipment status only through a few technical indicators, failing to comprehensively reflect the actual operating conditions of the equipment, resulting in insufficient accuracy in capturing the correlation between crew allocation and equipment status, making it difficult to output the optimal allocation scheme. Summary of the Invention

[0004] This invention primarily addresses the problem that existing crew position allocation methods rely on a single factor, making it difficult to maximize the operational status and efficiency of ship equipment. This invention discloses a crew position allocation method and apparatus based on information matching.

[0005] In a first aspect, this invention discloses a method for assigning crew positions based on information matching, comprising:

[0006] S1, Obtain the set of historical information on crew member job assignments; the set of historical information on crew member job assignments includes several historical vectors of crew member job assignments and a set of operating status information of the corresponding crew member job equipment; the set of operating status information of the crew member job equipment includes a subset of operating status values ​​for each crew member job equipment; the subset of operating status values ​​includes a sequence of operating values ​​for each technical indicator of the crew member job equipment; in the set of historical information on crew member job assignments, each historical vector of crew member job assignments has a corresponding set of operating status information of the crew member job equipment;

[0007] S2, Model the set of historical information on crew member job assignments to obtain a crew member job information matching model;

[0008] S3, Solve the crew position information matching model to obtain the crew position allocation vector; the crew position allocation vector includes the crew information allocated to each position.

[0009] The process of modeling the historical information set of crew member job assignments to obtain a crew member job information matching model includes:

[0010] S21, preprocess the set of historical information on crew member job assignments to obtain a preprocessed information set;

[0011] S22, The preprocessed information set is modeled to obtain a crew member job information matching model.

[0012] The preprocessing of the historical information set of crew member job assignments yields a preprocessed information set, including:

[0013] S211, perform data cleaning processing on the set of historical information on crew member job assignments to obtain a first data set;

[0014] S212, Perform data category checking on the first data set to obtain the second data set;

[0015] S213, perform pattern discrimination processing on the second data set to obtain a preprocessed information set.

[0016] The process of performing pattern discrimination on the second data set to obtain a preprocessed information set includes:

[0017] S2131, Perform a first pattern discrimination process on the set of operating status information of crew member post equipment in the second data set to obtain a first set of operating status information;

[0018] S2132, perform second mode discrimination processing on the crew position allocation history vector and the first operating status information set in the second data set to obtain a preprocessed information set.

[0019] The second pattern discrimination processing is performed on the crew position allocation history vector and the first operational status information set in the second data set to obtain a preprocessed information set, including:

[0020] S21321, Based on the historical vector of each crew member position allocation in the second dataset, obtain the sequence of operating values ​​of the technical indicators of all crew member position equipment corresponding to the historical vector of crew member position allocation in the first operating status information set;

[0021] S21322, calculate the mean of the operating value sequence of each acquired technical indicator to obtain the corresponding technical indicator mean;

[0022] S21323, using the average of all technical indicators of the equipment for each crew member position, a row vector is constructed, and using the row vectors of all the equipment for each crew member position, a state matrix corresponding to the historical vector of crew member position allocation is constructed.

[0023] S21324, Based on the historical vector and state matrix of crew member job allocation, a crew member job equipment pattern matching model is constructed;

[0024] S21325, using all the historical vectors and state matrices of crew member job assignments, solve the crew member job equipment pattern matching model to obtain the matching model;

[0025] The expression for the matching model is: x is the historical vector of crew position assignments input to the matching model, and matrix A is obtained by solving the crew position equipment pattern matching model;

[0026] S21326, using a matching model, the historical vectors assigned to each crew member's position are calculated and processed to obtain the solution matrix;

[0027] S21327, Subtract the solution matrix from the state matrix corresponding to the crew position allocation history vector and calculate the matrix modulus to obtain the matrix modulus;

[0028] S21328, delete the crew position allocation history vector corresponding to the state matrix whose matrix modulus is greater than the set difference threshold from the second dataset, and delete the operating value sequence of the technical indicators of all crew position equipment corresponding to the crew position allocation history vector corresponding to the state matrix from the first operating status information set.

[0029] S21329, For each crew member position in the second dataset, assign a historical vector. After executing S21326 to S21328 respectively, use all the crew member position assignment historical vectors in the second dataset and the sequence of operating values ​​of the technical indicators of all crew member position equipment corresponding to the historical vectors in the first operating status information set to construct a preprocessed information set.

[0030] The step of modeling the preprocessed information set to obtain a crew member job information matching model includes:

[0031] S221, evaluate and process the set of operating status information of each crew member's equipment to obtain the corresponding comprehensive evaluation value;

[0032] S222, using the historical vector of crew position allocation as a multivariate independent variable and the comprehensive evaluation value of all crew position equipment as a multivariate dependent variable, a matching model is performed on the multivariate independent and multivariate dependent variables to obtain the crew position information matching model.

[0033] The evaluation and processing of the operational status information set of each crew member's equipment to obtain the corresponding comprehensive evaluation value includes:

[0034] Obtain the standard operating values ​​of all technical indicators for the equipment at each crew member's post;

[0035] For each crew member's equipment, a corresponding data vector is constructed using the sequence of operating values ​​for each technical indicator;

[0036] Subtract the corresponding standard operating value from the data vector of each technical indicator to obtain the corresponding difference vector;

[0037] Using all the difference vectors of the equipment for a crew member's post, a difference matrix for the equipment for that post is constructed.

[0038] Statistical processing is performed on each row vector of the difference matrix to obtain the mean, variance, kurtosis, and norm of each row vector;

[0039] The mean, variance, kurtosis, and norm of all row vectors of the crew member equipment are fused and calculated to obtain the comprehensive evaluation value of the crew member equipment.

[0040] The expression for the fusion calculation is:

[0041]

[0042] Where zp is the comprehensive evaluation value of the equipment for crew positions, and N1 is the row dimension of the difference matrix. , , and Let be the mean, variance, kurtosis, and norm of the i-th row vector of the difference matrix, respectively. and These are the preset weighting factors.

[0043] A second aspect of the present invention discloses a crew position allocation device based on information matching, the device comprising:

[0044] Memory containing executable program code;

[0045] A processor coupled to the memory;

[0046] The processor calls the executable program code stored in the memory to execute the information matching-based crew assignment method.

[0047] In a third aspect, the present invention discloses a computer-readable storage medium storing computer instructions, which, when invoked by a computer, are used to execute the information matching-based crew assignment method.

[0048] In a fourth aspect of this invention, an information data processing terminal is disclosed, which is used to implement the aforementioned information matching-based crew position allocation method.

[0049] The beneficial effects of this invention are as follows:

[0050] This invention integrates historical information on crew member job assignments with corresponding information on the operational status of equipment for those crew member positions to construct a matching model. This breaks away from the limitations of traditional job assignments that rely on human experience. By using equipment operational status as a key reference for job assignments, the invention ensures that the assignment scheme not only matches the crew members' skills and experience but also meets the requirements for long-term stable operation of equipment. This effectively reduces the risk of equipment failure due to improper assignments and improves the overall operational efficiency and safety of the vessel.

[0051] This invention performs multi-step preprocessing on the historical information set of crew member job assignments. By cleaning up missing values, smoothing noisy data, and deleting outliers, it ensures that data attributes meet preset requirements through data category checks and eliminates abnormal information that differs too much from the overall data pattern through pattern discrimination. This comprehensively improves the data quality of the input model, provides reliable data support for the construction and solution of the subsequent job information matching model, avoids model bias caused by poor data, and significantly enhances the credibility and stability of the model output results.

[0052] In the modeling process, this invention comprehensively evaluates the operational status information of each crew member's equipment. By calculating the difference between the equipment's technical indicators and standard operating values, and integrating multi-dimensional statistical features such as the mean, variance, kurtosis, and norm of the difference matrix row vectors, a comprehensive evaluation value is obtained. This comprehensive and accurate reflection of the actual operating status of the equipment enables the job information matching model to more clearly capture the intrinsic relationship between crew member allocation and equipment operating status, thereby outputting a job allocation scheme that better meets the optimal operating requirements of the equipment, extending the service life of the equipment, and reducing the cost of ship operation and maintenance. Attached Figure Description

[0053] Figure 1 This is a flowchart illustrating the implementation of the method of the present invention. Detailed Implementation

[0054] To better understand the content of this invention, an embodiment is provided here.

[0055] Figure 1 This is a flowchart illustrating the implementation of the method of the present invention.

[0056] In a first aspect, this invention discloses a method for assigning crew positions based on information matching, comprising:

[0057] S1, Obtain the historical information set of crew member job assignments; the historical information set of crew member job assignments includes several historical vectors of crew member job assignments and the corresponding set of operating status information of crew member job equipment; the set of operating status information of crew member job equipment includes a subset of operating status values ​​for each crew member job equipment. The subset of operating status values ​​includes a sequence of operating values ​​for each technical indicator of the crew member job equipment;

[0058] S2, Model the set of historical information on crew member job assignments to obtain a crew member job information matching model;

[0059] S3, Solve the crew position information matching model to obtain the crew position allocation vector; the crew position allocation vector includes crew information for each position.

[0060] The process of modeling the historical information set of crew member job assignments to obtain a crew member job information matching model includes:

[0061] S21, preprocess the set of historical information on crew member job assignments to obtain a preprocessed information set;

[0062] S22, The preprocessed information set is modeled to obtain a crew member job information matching model;

[0063] The preprocessing of the historical information set of crew member job assignments yields a preprocessed information set, including:

[0064] S211, perform data cleaning processing on the set of historical information on crew member job assignments to obtain a first data set;

[0065] S212, Perform data category checking on the first data set to obtain the second data set;

[0066] S213, perform pattern discrimination processing on the second data set to obtain a preprocessed information set.

[0067] The data cleaning process includes: filling in missing values, smoothing noisy data, and smoothing or deleting outliers. Smoothing noisy data involves first identifying the noisy data, and then smoothing it based on the data preceding and following it. The noisy data refers to values ​​whose values ​​are less than the sensor's detection sensitivity for the observed data, or greater than the sensor's measurement limit for the observed data. Outlier identification can be performed using a Kalman filter. The filling value for missing values ​​can be determined by averaging the measurements within a certain sampling interval before and after the missing value.

[0068] The data category checking process includes: for each type of data in the first data set, determining whether its data attributes match the preset data attributes of the data type, deleting the data that does not match from the first data set, and obtaining the second data set.

[0069] The process of performing pattern discrimination on the second data set to obtain a preprocessed information set includes:

[0070] S2131, Perform a first pattern discrimination process on the set of operating status information of crew member post equipment in the second data set to obtain a first set of operating status information;

[0071] S2132, perform second mode discrimination processing on the crew position allocation history vector and the first operating status information set in the second data set to obtain a preprocessed information set.

[0072] The first mode discrimination process is performed on the set of operating status information of crew member positions and equipment in the second data set to obtain the first set of operating status information, including:

[0073] S21321, Based on the historical vector of each crew member position allocation in the second dataset, obtain the sequence of operating values ​​of the technical indicators of all crew member position equipment corresponding to the historical vector of crew member position allocation in the first operating status information set;

[0074] S21322, calculate the mean of the operating value sequence of each acquired technical indicator to obtain the corresponding technical indicator mean;

[0075] S21323, using the average of all technical indicators of the equipment for each crew member position, a row vector is constructed, and using the row vectors of all the equipment for each crew member position, a state matrix corresponding to the historical vector of crew member position allocation is constructed.

[0076] S21324, Based on the historical vector and state matrix of crew member job allocation, a crew member job equipment pattern matching model is constructed;

[0077] S21325, using all the historical vectors and state matrices of crew member job assignments, solve the crew member job equipment pattern matching model to obtain the matching model;

[0078] The expression for the matching model is: x is the job assignment vector input to the matching model, and matrix A is obtained by solving the crew job equipment pattern matching model;

[0079] S21326, using a matching model, the historical vectors assigned to each crew member's position are calculated and processed to obtain the solution matrix;

[0080] S21327, Subtract the solution matrix from the state matrix corresponding to the crew position allocation history vector and calculate the matrix modulus to obtain the matrix modulus;

[0081] S21328, delete the crew position allocation history vector corresponding to the state matrix whose matrix modulus is greater than the set difference threshold from the second dataset, and delete the operating value sequence of the technical indicators of all crew position equipment corresponding to the crew position allocation history vector corresponding to the state matrix from the first operating status information set.

[0082] S21329, For each crew member position in the second dataset, assign a historical vector. After executing S21326 to S21328 respectively, use all the crew member position assignment historical vectors in the second dataset and the sequence of operating values ​​of the technical indicators of all crew member position equipment corresponding to the historical vectors in the first operating status information set to construct a preprocessed information set.

[0083] The expression for the crew member job equipment pattern matching model is as follows:

[0084]

[0085] ,

[0086] in, Let A represent the identity matrix with row dimensions as its dimension, where A represents the matrix to be solved. R represents the historical vector of crew member job assignments, and R represents the state matrix of the historical vector of crew member job assignments.

[0087] The matching model's expression, by establishing a correlation between the input job assignment vector and a specific matrix, can directly map the corresponding equipment status-related results. This expression simplifies the calculation process of the correlation between job assignment and equipment status, enabling the rapid acquisition of corresponding equipment status characteristics based on job assignment. This provides an efficient calculation method for subsequently determining whether data conforms to the overall pattern, helps to accurately filter historical data consistent with the overall pattern, eliminates abnormal information, and ensures that the data used for modeling has good consistency and representativeness.

[0088] The expression of the crew position equipment pattern matching model aims to minimize the difference between the correlation result of the position allocation vector and the matrix and the state matrix, and ensures the normalization of the matrix through constraints. This setting can accurately capture the inherent correlation between position allocation and equipment operating status, optimize the matching relationship between the two by minimizing the difference, and at the same time, the constraints avoid unreasonable values ​​of matrix parameters, prevent the model from overfitting local data, enhance the stability and generalization ability of the model, and enable the obtained matching model to more accurately reflect the overall data pattern, providing a reliable basis for subsequent data screening.

[0089] The algorithm for solving the crew member position equipment pattern matching model can be a genetic algorithm or a particle filter algorithm.

[0090] The step of modeling the preprocessed information set to obtain a crew member job information matching model includes:

[0091] S221, evaluate and process the set of operating status information of each crew member's equipment to obtain the corresponding comprehensive evaluation value;

[0092] S222, using the historical vector of crew position allocation as a multivariate independent variable and the comprehensive evaluation value of all crew position equipment as a multivariate dependent variable, a matching model is performed on the multivariate independent and multivariate dependent variables to obtain the crew position information matching model.

[0093] The evaluation and processing of the operational status information set of each crew member's equipment to obtain the corresponding comprehensive evaluation value includes:

[0094] Obtain the standard operating values ​​of all technical indicators for the equipment at each crew member's post;

[0095] For each crew member's equipment, a corresponding data vector is constructed using the sequence of operating values ​​for each technical indicator;

[0096] Subtract the corresponding standard operating value from the data vector of each technical indicator to obtain the corresponding difference vector;

[0097] Using all the difference vectors of the equipment for a crew member's post, a difference matrix for the equipment for that post is constructed.

[0098] Statistical processing is performed on each row vector of the difference matrix to obtain the mean, variance, kurtosis, and norm of each row vector;

[0099] The mean, variance, kurtosis, and norm of all row vectors of the crew member equipment are fused and calculated to obtain the comprehensive evaluation value of the crew member equipment.

[0100] The expression for the fusion calculation is:

[0101] ,

[0102] Where zp is the comprehensive evaluation value of the equipment for crew positions, and N1 is the row dimension of the difference matrix. , , and Let be the mean, variance, kurtosis, and norm of the i-th row vector of the difference matrix, respectively. and These are the preset weighting factors.

[0103] The fusion calculation expression integrates multi-dimensional statistical features such as the mean, variance, kurtosis, and norm of the row vectors of the difference matrix, and introduces weighting factors for calculation, achieving a comprehensive assessment of equipment operating status. This expression comprehensively considers the characteristics of the differences between equipment technical indicators and standard values ​​in terms of central tendency, dispersion, distribution pattern, and overall deviation, avoiding the one-sidedness of single-dimensional assessment. Simultaneously, the introduction of weighting factors allows for flexible adjustment of the proportion of different technical indicators in the comprehensive assessment based on their importance, making the assessment results more closely aligned with the actual focus of equipment operation, thereby accurately reflecting the true operating status of the equipment and providing reliable dependent variable data for the job matching model.

[0104] The norm value can be adopted. The norm is calculated.

[0105] The process involves using the historical vector of crew member job allocation as a multivariate independent variable and the comprehensive evaluation value of all crew member job equipment as a multivariate dependent variable. Matching and modeling these multivariate independent and dependent variables yields a crew member job information matching model, including:

[0106] Using the historical vector of crew position allocation as the multivariate independent variable and the comprehensive evaluation value of all crew position equipment as the multivariate dependent variable, a multivariate function was fitted to obtain the multivariate fitting function.

[0107] Based on the aforementioned multivariate fitting function, a crew member job information matching model is constructed.

[0108] The expression for the crew member job information matching model is:

[0109] ,

[0110] ,

[0111] in, Represents a multivariate fitting function. This represents the vector magnitude of the multivariate variables output by the multivariate fitting function. Let be the crew position classification vector for the input crew position information matching model, and be the variable to be solved, whose elements take the crew member ID information for each position. This represents the crew position allocation vector obtained from the solution, where E is the preset upper limit threshold for allocation.

[0112] The elements of the crew position allocation vector represent the crew information assigned to each position.

[0113] The expression of the crew position information matching model aims to maximize the vector magnitude of the multivariate fitting function output, and ensures the rationality of the solution results by setting an upper limit threshold. This setting, under the premise of conforming to actual allocation constraints, can find the position allocation scheme that optimizes the overall performance of the comprehensive evaluation value of all crew positions and equipment. It ensures that the allocation scheme maximizes the comprehensive benefits of equipment operation, and avoids the allocation scheme from exceeding the actual operability range through threshold constraints. This allows it to adapt to the position configuration scale and actual operational needs of different ships, effectively improving the scientificity and feasibility of position allocation.

[0114] This invention constructs a multivariate fitting function using the historical vector of crew member job allocation as a multivariate independent variable and the comprehensive equipment evaluation value as a multivariate dependent variable, forming a crew member job information matching model. Through a scientific solution algorithm, the optimal job allocation vector is determined. This not only efficiently avoids the tediousness and inefficiency of manual selection of allocation schemes and significantly improves job allocation efficiency, but also ensures that the allocation scheme is within a reasonable and feasible range by setting a preset allocation upper limit threshold. It can flexibly adapt to the job allocation needs of ships of different sizes and with different equipment configurations, and has strong practicality and promotion value.

[0115] The process involves using the historical vector of crew member job allocation as a multivariate independent variable and the comprehensive evaluation value of all crew member job equipment as a multivariate dependent variable. Matching and modeling these multivariate independent and dependent variables yields a crew member job information matching model, including:

[0116] S2221, the preprocessed information set is represented as a structured data matrix. Each row of the matrix corresponds to a historical vector of crew position allocation and a comprehensive evaluation value of all crew position equipment. The columns are divided into two parts: the first part is the historical vector of crew position allocation. , represents the historical vector of the i-th crew position allocation, and d is the total number of position categories; the latter part is the vector of the comprehensive evaluation value of the crew position equipment. , of which elements Let m represent the comprehensive evaluation value of the j-th device corresponding to the historical vector of the i-th employee's job assignment, where m is the number of devices.

[0117] S2222, Perform vector serialization processing on the structured data matrix to obtain a job allocation vector sequence, a comprehensive evaluation value vector sequence, and vector pair sequence samples, including: arranging all historical records in chronological order to form the job allocation vector sequence. and the corresponding comprehensive evaluation value vector sequence Extract a continuous subsequence of length L using a sliding window method, and generate a set of vector pairs of sequence samples. ,in , This is used to capture the temporal coupling relationship between job allocation dynamics and equipment state evolution;

[0118] S2223, Perform first-type frequent pattern mining on the job allocation vector sequence to obtain a first transition pattern model, including: processing the job allocation vector sequence... Vector clustering and discretization are performed. A density-based clustering algorithm is used to divide similar job classification vectors into several clusters, with the center of each cluster denoted as . and each Mapped to the discrete label of its cluster Convert the job assignment vector sequence into a symbolic sequence. Frequent sequence pattern mining is performed on this symbol sequence to identify recurring high-support job allocation state transition patterns. These high-support job allocation state transition patterns are used as the first transition pattern model, which takes the form of: This indicates that a certain type of job allocation state sequence has frequently occurred in history; the first transition mode model includes several job allocation state sequences;

[0119] S2224, Perform second-type frequent pattern mining on the comprehensive evaluation value vector sequence to obtain a second transition pattern model, including: for the comprehensive evaluation value vector sequence Multidimensional state partitioning is performed. For each device's evaluation dimension j, the total value range is divided according to its historical value distribution. Each performance level range (e.g., "low / medium / high stability") will be used to determine the performance level of each individual performance level. Based on the performance level range to which its value belongs, it is mapped to a multidimensional discrete state label. Where S is the set of all possible state combinations; based on the set of state sequences Frequent sequence pattern mining is performed to identify frequently occurring equipment performance evolution path patterns. The high-support job allocation state transition pattern is used as the second transition pattern model, which takes the form of: The support of a path in the second transition mode model is the frequency of that path appearing in the sequence; the historical value distribution can be obtained from the set of operational status information of crew member equipment. The second transition mode model includes several equipment performance evolution paths;

[0120] S2225, For the vector pair sequence samples, perform the third type of cross-domain association pattern mining to obtain strong association rules and corresponding confidence values, including: based on the vector pair sequence samples, construct a transaction database of job-equipment co-occurrence, where each transaction in the transaction database corresponds to a sliding window sample. Encode the sliding window samples into a pair of patterns: ,in Assign a status sequence (from S2223) to the job positions appearing in this window. For the corresponding equipment performance evolution path (from S2224); perform bimodal association rule mining on this transaction database to find strong association rules in the form of "if job state sequence P occurs, then equipment state sequence D has a high probability of occurring". The confidence level of the strong association rule is defined as:

[0121]

[0122] in Let P be the number of transactions containing the job pattern P. The number of transactions containing the device state sequence D. The number of transactions that simultaneously contain both P and D; only transactions with a confidence level greater than the threshold are retained. (e.g., 0.7) and the lift The strong association rules are then used to obtain the final strong association rules.

[0123] S2226. Based on the above three types of patterns, a heterogeneous relationship graph model is constructed. This model includes two types of nodes: job allocation state sequence nodes and equipment performance evolution path nodes. Nodes of the same type are connected by co-occurrence frequency (to represent pattern similarity), while nodes of different types are connected by bimodal association rules. Edge weights consist of support, confidence, and time decay factor. The graph model is obtained through weighted summation. Ultimately, this graph model serves as the core structure of the crew position information matching model, supporting the deduction of possible equipment operation trends from the new position allocation vector. Nodes of the same type are connected by co-occurrence frequency, which involves calculating the frequency of co-occurrence of nodes of the same type in the transaction database and connecting nodes of the same type whose frequency exceeds a preset frequency threshold through co-occurrence frequency.

[0124] S2227, Confirm that the heterogeneous relationship graph model is a crew member job information matching model.

[0125] The process of solving the equipment pattern matching model for crew positions yields a matching model, including:

[0126] With the objective of maximizing the modulus of the vector corresponding to the equipment performance evolution path, the heterogeneous relationship graph model is solved to obtain the crew position allocation vector.

[0127] The heterogeneous relation graph model can be solved using the P-SSOR algorithm or a method based on a directed acyclic graph (DAG).

[0128] In all embodiments of the present invention, the variables involved in all computational expressions or mathematical functions have been dimensionlessized before computation.

[0129] In all embodiments of the present invention, the values ​​of the independent variables in the input of all computational expressions or mathematical functions meet the reasonable requirements of the input range of the computational expressions or mathematical functions, and can ensure that the computational expressions or mathematical functions can be calculated smoothly without violating physical laws or mathematical rules.

[0130] A second aspect of the present invention discloses a crew position allocation device based on information matching, the device comprising:

[0131] Memory containing executable program code;

[0132] A processor coupled to the memory;

[0133] The processor calls the executable program code stored in the memory to execute the information matching-based crew assignment method.

[0134] In a third aspect, the present invention discloses a computer-readable storage medium storing computer instructions, which, when invoked by a computer, are used to execute the information matching-based crew assignment method.

[0135] In a fourth aspect of this invention, an information data processing terminal is disclosed, which is used to implement the aforementioned information matching-based crew position allocation method.

[0136] The above description is merely an embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of the present invention should be included within the scope of the claims of the present invention.

Claims

1. A method for assigning crew positions based on information matching, characterized in that, include: S1, Obtain the set of historical information on crew member job assignments; the set of historical information on crew member job assignments includes several historical vectors of crew member job assignments and a set of operating status information of the corresponding crew member job equipment; the set of operating status information of the crew member job equipment includes a subset of operating status values ​​for each crew member job equipment; the subset of operating status values ​​includes a sequence of operating values ​​for each technical indicator of the crew member job equipment; in the set of historical information on crew member job assignments, each historical vector of crew member job assignments has a corresponding set of operating status information of the crew member job equipment; S2, Modeling the set of historical information on crew member job assignments to obtain a crew member job information matching model, including: S21, preprocess the set of historical information on crew member job assignments to obtain a preprocessed information set; S22, Model the preprocessed information set to obtain a crew member job information matching model, including: S221, evaluate and process the set of operating status information of each crew member's equipment to obtain the corresponding comprehensive evaluation value; S222, using the historical vector of crew position allocation as a multivariate independent variable and the comprehensive evaluation value of all crew position equipment as a multivariate dependent variable, a matching model is performed on the multivariate independent and multivariate dependent variables to obtain the crew position information matching model; S3, Solve the crew position information matching model to obtain the crew position allocation vector; the crew position allocation vector includes the crew information allocated to each position.

2. The crew position allocation method based on information matching as described in claim 1, characterized in that, The preprocessing of the historical information set of crew member job assignments yields a preprocessed information set, including: S211, perform data cleaning processing on the set of historical information on crew member job assignments to obtain a first data set; S212, Perform data category checking on the first data set to obtain the second data set; S213, perform pattern discrimination processing on the second data set to obtain a preprocessed information set.

3. The crew position allocation method based on information matching as described in claim 2, characterized in that, The process of performing pattern discrimination on the second data set to obtain a preprocessed information set includes: S2131, Perform a first pattern discrimination process on the set of operating status information of crew member post equipment in the second data set to obtain a first set of operating status information; S2132, perform second mode discrimination processing on the crew position allocation history vector and the first operating status information set in the second data set to obtain a preprocessed information set.

4. The crew position allocation method based on information matching as described in claim 3, characterized in that, The second pattern discrimination processing is performed on the crew position allocation history vector and the first operational status information set in the second data set to obtain a preprocessed information set, including: S21321, Based on the historical vector of each crew member position allocation in the second dataset, obtain the sequence of operating values ​​of the technical indicators of all crew member position equipment corresponding to the historical vector of crew member position allocation in the first operating status information set; S21322, calculate the mean of the operating value sequence of each acquired technical indicator to obtain the corresponding technical indicator mean; S21323, using the average of all technical indicators of the equipment for each crew member position, a row vector is constructed, and using the row vectors of all the equipment for each crew member position, a state matrix corresponding to the historical vector of crew member position allocation is constructed. S21324, Based on the historical vector and state matrix of crew member job allocation, a crew member job equipment pattern matching model is constructed; S21325, using all the historical vectors and state matrices of crew member job assignments, solve the crew member job equipment pattern matching model to obtain the matching model; The expression for the matching model is f(x) = xA, where x is the historical vector of crew position allocation as input to the matching model, and matrix A is obtained by solving the crew position equipment pattern matching model. S21326, using a matching model, the historical vectors assigned to each crew member's position are calculated and processed to obtain the solution matrix; S21327, Subtract the solution matrix from the state matrix corresponding to the crew position allocation history vector and calculate the matrix modulus to obtain the matrix modulus; S21328, delete the crew position allocation history vector corresponding to the state matrix whose matrix modulus is greater than the set difference threshold from the second dataset, and delete the operating value sequence of the technical indicators of all crew position equipment corresponding to the crew position allocation history vector corresponding to the state matrix from the first operating status information set. S21329, For each crew member position in the second dataset, assign a historical vector. After executing S21326 to S21328 respectively, use all the crew member position assignment historical vectors in the second dataset and the sequence of operating values ​​of the technical indicators of all crew member position equipment corresponding to the historical vectors in the first operating status information set to construct a preprocessed information set.

5. The crew position allocation method based on information matching as described in claim 1, characterized in that, The evaluation and processing of the operational status information set of each crew member's equipment to obtain the corresponding comprehensive evaluation value includes: Obtain the standard operating values ​​of all technical indicators for the equipment at each crew member's post; For each crew member's equipment, a corresponding data vector is constructed using the sequence of operating values ​​for each technical indicator; Subtract the corresponding standard operating value from the data vector of each technical indicator to obtain the corresponding difference vector; Using all the difference vectors of the equipment for a crew member's post, a difference matrix for the equipment for that post is constructed. Statistical processing is performed on each row vector of the difference matrix to obtain the mean, variance, kurtosis, and norm of each row vector; The mean, variance, kurtosis, and norm of all row vectors of the crew member equipment are fused and calculated to obtain the comprehensive evaluation value of the crew member equipment. The expression for the fusion calculation is: Where zp is the comprehensive evaluation value of the equipment for crew positions, and N1 is the row dimension of the difference matrix. , , and Let be the mean, variance, kurtosis, and norm of the i-th row vector of the difference matrix, respectively. and These are the preset weighting factors.

6. A crew member job allocation device based on information matching, characterized in that, The device includes: Memory containing executable program code; A processor coupled to the memory; The processor calls the executable program code stored in the memory to execute the crew position allocation method based on information matching as described in any one of claims 1 to 5.

7. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions, which, when invoked by a computer, are used to execute the crew position allocation method based on information matching as described in any one of claims 1 to 5.

8. An information data processing terminal, characterized in that, The information data processing terminal is used to implement the crew position allocation method based on information matching as described in any one of claims 1 to 5.

Images