Human-post dynamic matching system and method based on digital trajectory

Abstract

The invention discloses a human-post dynamic matching system and method based on a digital trajectory, and relates to the technical field of human resources. According to the method, a behavior stability index Ks of an operation behavior is established based on a behavior trajectory data sequence, and the behavior stability index Ks is preliminarily compared and evaluated; and the system can automatically identify the rhythm fluctuation trend of the personnel in the post process execution process. When the behavior stability index Ks exceeds a preset threshold value, it can be judged that the behavior process is abnormal in real time, and a process structure depth detection module is triggered. The mechanism breaks through the limitation of judging the flow abnormality by depending on single residence time or the number of events in the prior art, realizes statistical stability monitoring of the time interval of the operation event, and enables the system to recognize the problems of operation rhythm abnormality, behavior jitter, node delay and the like in advance, thereby improving the sensitivity and accuracy of flow behavior monitoring.

Description

Technical Field

[0001] This invention relates to the field of human resources technology, specifically to a dynamic matching system and method for people and jobs based on digital trajectories. Background Technology

[0002] With the development of human resources, job matching process detection systems have gradually become an important tool for enterprises to assess job matching degree, job stability, and job adaptability. As enterprises continuously improve their process digitization and automation, behavioral data-driven job process detection has evolved from traditional manual experience-based judgment to a modern detection method that relies on intelligent analysis of operational behavior data, process structure data, and matching result data. In this type of detection method, the system needs to collect, model, and analyze the operational behaviors of personnel during job process execution to identify the rhythm characteristics, structural deviations, and process matching degree of personnel during process execution. These data constitute the core content of "job matching process detection data."

[0003] Currently, in existing technologies, personnel-job matching process detection systems typically store all the aforementioned different types of data in the same data structure or mixed data table, lacking a clear data hierarchical management mechanism. Digital trajectory data, process structure mapping data, intermediate matching judgment indicator data, and final detection result data are often stored together, making subsequent data calculations, structure comparisons, behavioral analysis, and result verification complex and even chaotic. Especially when it is necessary to trace the reasons for a person's mismatch or execution deviation, the lack of data type classification and hierarchical management makes it difficult for engineers or the system to quickly locate key data sources, increasing the cost of data cleaning, data reconstruction, and model iteration. In addition, the lack of a hierarchical mechanism also makes the data logic between different detection processes unclear, resulting in high coupling between detection modules and poor scalability.

[0004] The root cause of these problems lies in the fact that existing human-job matching detection systems generally adopt a "results-oriented" data recording method, which only records the final data and lacks structured management of data sources, data types, and data flow. This mixed storage method may be sustainable in the early stages when the data volume is small or the detection process is limited, but in large enterprises with multiple positions and processes, it leads to a series of abnormal technical effects. For example, due to the lack of a clear data hierarchy structure, the system cannot accurately distinguish between "behavioral source data" and "structural mapping intermediate data" when performing process mapping analysis, resulting in unstable matching judgment results. During process backtracking, because the data is not stored hierarchically, it is impossible to quickly identify whether the anomaly is caused by behavioral fluctuations, process irregularities, or algorithmic errors, significantly reducing analysis efficiency. Furthermore, when the system performs aggregation calculations on large amounts of heterogeneous data, the mixed storage of different types of data leads to decreased computational performance, increased read latency, and erroneous data dependencies, seriously affecting the reliability and real-time performance of the entire process detection system. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides a dynamic matching system and method for people and positions based on digital trajectories, which solves the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution, comprising the following steps: S1. Collect digital trajectory data of personnel during their job operation process and input the digital trajectory data into the central processing system in real time; the central processing system then forms a behavioral trajectory data sequence based on the digital trajectory data. S2. Establish a behavior stability index Ks based on the behavior trajectory data sequence; conduct a preliminary comparative evaluation of the behavior stability index Ks to determine the fluctuation trend of the process behavior; when the stability index Ks determines that the behavior process is abnormal, trigger the process structure depth detection module to perform process structure mapping processing on the behavior trajectory data sequence and calculate the process structure mapping deviation vector Fm. S3. Based on the process structure mapping deviation vector Fm, construct the process structure consistency index Cp, and use it as input to calculate the process matching index Rm. Perform process matching judgment on the process matching index Rm, and output the detection result of the person-job matching process. When the detection result of the person-job matching process determines that the job process is not suitable, execute the adaptation control mechanism.

[0007] Preferably, S1 includes S11; S11. Set up multiple data collection points on the personnel operation terminals of the enterprise's multi-channel customer service center, and set up corresponding data collection plug-ins in each data collection point to extract digital trajectory data summarized by customer service personnel during their job process operations in real time, and transmit it to the central processing system. The data collection points include operation rhythm data collection points, process node dwell data collection points, operation continuity stability data collection points, system process template data collection points, and node behavior mapping data collection points. The digital trajectory data includes the operation time interval parameter Te, the process dwell time parameter Ts, the rhythm stability parameter So, the template time parameter Tp, and the mapping deviation parameter △m; The operation rhythm acquisition point is set in the event trigger listening module of the operation terminal, and an operation timestamp acquisition plugin is embedded in the operation rhythm acquisition point to automatically record the event occurrence time when each operation event is triggered, with the operation time interval parameter Te between adjacent operation events. The process node dwell time collection point is set in the node status listening interface of the process control module, and the process node dwell time monitoring plugin is embedded in the process node dwell time collection point. It is used to record the node timestamp when personnel enter and exit the process node, so as to obtain the process dwell time parameter Ts of personnel at the process node. Operational continuous stability acquisition points are set within the behavior sampling module of the behavior sequence buffer, and a rhythm stability analysis plugin is embedded in the operational continuous stability acquisition points. The plugin generates the rhythm stability parameter So by performing sliding window calculation on the operation event sequence within multiple consecutive sampling windows. The system process template collection point is set at the node time parameter interface of the job process template management module, and a template time parameter retrieval plugin is embedded in the system process template collection point to extract the template time parameter Tp of each process node from the job process template; The node behavior mapping collection point is set within the behavior structure mapping interface of the process analysis module, and a node behavior deviation collection plugin is embedded in the node behavior mapping collection point. The mapping deviation parameter △m is calculated based on the time difference between personnel operation events and template nodes.

[0008] Preferably, S1 further includes S12; S12. The central processing system manages the acquisition scheduling strategy of multiple acquisition points in a unified manner and acquires digital trajectory data periodically at a unified sampling frequency. After acquiring the digital trajectory data, the extreme value normalization method is used to perform dimensionless processing on all parameters in the digital trajectory data to eliminate the influence of the unit dimension of all parameters in the digital trajectory data. The dimensionless digital trajectory data is then sorted according to the acquisition time to obtain the behavioral trajectory data sequence.

[0009] Preferably, S2 includes S21; S21. In the central processing system, based on all operation time interval parameters Te in the behavior trajectory data sequence, calculate the standard deviation and mean respectively to obtain the standard deviation and mean of operation time interval parameter Te; then calculate the ratio of the standard deviation and mean of operation time interval parameter Te to obtain the behavior stability index Ks. A preliminary comparative evaluation is conducted based on the output results of the behavioral stability index Ks to determine the fluctuation trend of process behavior. The specific evaluation content is as follows: When the behavioral stability index Ks ≤ 0.32, it indicates that the behavioral process is stable; When the behavioral stability index Ks > 0.32, it indicates abnormal behavioral processes.

[0010] Preferably, S2 further includes S22; S22. When the preliminary comparison and evaluation determines that the process behavior fluctuation trend is abnormal, and the process behavior is abnormal for three consecutive time windows, the process structure depth detection module is triggered. The process structure depth detection module performs process structure mapping processing on the behavior trajectory data sequence, takes the node time parameter of the job process template as a reference, performs node-level matching calculation on the operation time interval parameter Te of the digital trajectory data and the process dwell time parameter Ts, and generates a process structure mapping deviation vector Fm based on the process structure mapping deviation formula. The process structure mapping deviation vector Fm is calculated and output using the following process structure mapping deviation formula: In the formula, Fm i Te represents the process structure mapping deviation vector of the i-th node. i Tp represents the operation time interval parameter of the i-th node. i Ts represents the template time parameter of the i-th node. i Tp represents the process dwell time parameter at node i. s,i The template time parameter represents the recommended node stay duration (s) in the job process template for node i.

[0011] Preferably, S3 includes S31; S31. The central processing system performs a discrete analysis on the deviation values ​​of each node in the process structure mapping deviation vector Fm, obtains the standard deviation and the average value of the process structure mapping deviation vector Fm, calculates the ratio, obtains the process structure consistency index Cp, and forms intermediate data for matching judgment. Based on the output of the process structure consistency index Cp, the fluctuation of behavioral deviations is then assessed. The specific assessment criteria are as follows: When the process structure consistency index Cp≤0.3, it means that the process structure mapping deviation is concentrated among multiple process nodes, the degree of deviation of each process node is relatively consistent, and there is no obvious structural fluctuation trend when personnel execute job processes, which is a stable state of process structure. When the process structure consistency index Cp > 0.3, it indicates that the process structure mapping deviation is abnormal between different nodes, which is an unstable state of process structure.

[0012] Preferably, S3 further includes S32; S32. When it is determined that the process structure is unstable, the process structure consistency index Cp is used as an input item to calculate the process matching index Rm. The process matching index Rm is calculated and output using the following algorithm formula: Rm=1 / 1+Cp.

[0013] Preferably, S3 further includes S32; S32. The central processing system will use the calculated process matching index Rm as the output result to determine process matching. The specific determination is as follows: When the process matching index Rm≥0.68, it is determined to be a job process fit; When the process matching index Rm < 0.68, it is determined that the job process is not suitable.

[0014] Preferably, S3 further includes S33; S33. When a job process is determined to be incompatible, an adaptation control mechanism is executed. This adaptation control mechanism identifies the process structure mapping deviation vector Fm of the i-th node from the process structure mapping deviation vector Fm of all nodes through the central processing system. i Generate node deviation report results and send the node deviation report results to the training system, monitoring system or process optimization system; Meanwhile, after receiving the node deviation report results from the central processing system, the operator's terminal automatically triggers interface assistance strategies based on the degree of node deviation, including rhythm prompts, key node reminders, and process structure guidance prompts.

[0015] A dynamic matching system for people and positions based on digital trajectories includes a process data acquisition module, a process structure mapping module, and a matching determination module; The process data acquisition module collects digital trajectory data of personnel during job process operations and inputs the digital trajectory data into the central processing system in real time; the central processing system then forms a behavioral trajectory data sequence based on the digital trajectory data. The process structure mapping module establishes a behavior stability index Ks based on the behavior trajectory data sequence; it performs a preliminary comparative evaluation of the behavior stability index Ks to determine the fluctuation trend of the process behavior; when the stability index Ks determines that the behavior process is abnormal, it triggers the process structure deep detection module to perform process structure mapping processing on the behavior trajectory data sequence and calculates the process structure mapping deviation vector Fm. The matching judgment module constructs a process structure consistency index Cp based on the process structure mapping deviation vector Fm, and uses it as input to calculate the process matching index Rm. It then performs process matching judgment on the process matching index Rm and outputs the detection result of the person-job matching process. When the detection result of the person-job matching process determines that the job process is not compatible, the adaptation control mechanism is executed.

[0016] This invention provides a dynamic matching system and method for people and positions based on digital trajectories. It has the following beneficial effects: (1) This method establishes a behavioral stability index Ks based on behavioral trajectory data sequences and performs preliminary comparative evaluation of the behavioral stability index, enabling the system to automatically identify the rhythm fluctuation trend of personnel during job process execution. When the behavioral stability index Ks exceeds a preset threshold, it can be determined in real time as a behavioral process abnormality and trigger the process structure depth detection module. This mechanism breaks through the limitations of existing technologies that rely on a single dwell time or number of events to judge process abnormalities, and realizes the statistical stability monitoring of the time interval of operational events, enabling the system to identify problems such as abnormal operation rhythm, behavioral jitter, and node delay in advance, thereby improving the sensitivity and accuracy of process behavior monitoring.

[0017] (2) This method divides the human-job matching process detection data into a digital trajectory data layer, a process structure mapping data layer, a matching judgment intermediate data layer, and a final detection result layer, thereby achieving structured hierarchical management of multiple types of detection data. This solves the problems of mixed storage, difficulty in traceability, and difficulty in analysis of process detection data in existing technologies. Each layer of data is maintained with an independent logical structure, making the source of behavioral deviation clearer, the calculation of structural deviation more interpretable, and the matching judgment results more transparent. This significantly reduces the difficulty of subsequent analysis and processing, and improves the scalability, data processing efficiency, and system reliability of the detection system. In addition, the hierarchical structure provides accurate data sources for process backtracking, personnel training, and automatic correction strategies, avoiding misjudgments and delays caused by data chaos in traditional systems.

[0018] (3) This method enables the system to form a complete closed-loop linkage process of "behavioral fluctuation detection, structural deviation analysis, matching judgment output, and adaptation control mechanism". When the behavioral stability index Ks is determined to be abnormal, the system immediately enters deep structural analysis based on the process structure mapping deviation vector Fm; when the structural consistency index Cp is determined to be unstable, the system completes the person-job matching judgment based on the process matching index Rm; when the process matching index Rm is determined to be incompatible, the adaptation control mechanism is triggered, and the source of deviation is quickly located by relying on the hierarchical data storage system. This linkage mechanism of behavioral layer, structural layer and decision layer significantly improves the accuracy, interpretability and response efficiency of person-job matching detection, and realizes intelligent management of the whole process from data collection, behavioral modeling, structural analysis to final matching judgment. It provides enterprises with a more refined, real-time and closed-loop person-job matching process detection capability, which is a comprehensive effect that is difficult to achieve with existing technologies. Attached Figure Description

[0019] Figure 1 This is a schematic diagram illustrating the steps of a dynamic matching method for people and positions based on digital trajectories according to the present invention. Figure 2 This is a schematic diagram of the process of a dynamic matching system for people and jobs based on digital trajectories according to the present invention; Figure 3 Process matching index Rm judgment curve. Detailed Implementation

[0020] 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 skilled in the art without creative effort are within the scope of protection of the present invention.

[0021] Example 1; please refer to Figure 1 This invention provides a method for dynamic matching of people and positions based on digital trajectories. To achieve the above objectives, this invention is implemented through the following technical solution, including the following steps: S1. Collect digital trajectory data of personnel during their job operation process and input the digital trajectory data into the central processing system in real time; the central processing system then forms a behavioral trajectory data sequence based on the digital trajectory data. S2. Establish a behavior stability index Ks based on the behavior trajectory data sequence; conduct a preliminary comparative evaluation of the behavior stability index Ks to determine the fluctuation trend of the process behavior; when the stability index Ks determines that the behavior process is abnormal, trigger the process structure depth detection module to perform process structure mapping processing on the behavior trajectory data sequence and calculate the process structure mapping deviation vector Fm. S3. Based on the process structure mapping deviation vector Fm, construct the process structure consistency index Cp, and use it as input to calculate the process matching index Rm. Perform process matching judgment on the process matching index Rm, and output the detection result of the person-job matching process. When the detection result of the person-job matching process determines that the job process is not suitable, execute the adaptation control mechanism.

[0022] In this embodiment, the method collects digital trajectory data such as operation time intervals and process dwell times in real time in S1 because the anomalies that first appear in the job process often manifest as rhythm drift or dwell anomalies. If these subtle changes cannot be captured in real time, behavioral fluctuations will be masked, making subsequent structural analysis lack accurate premises. The purpose of introducing the behavioral stability index Ks in S2 is to statistically quantify the above-mentioned behavioral changes and identify rhythm fluctuation trends through the relationship between standard deviation and mean, so as to avoid misjudgment caused by traditional methods that rely solely on the number of actions or dwell time. When the behavioral stability index Ks exceeds the threshold, structural depth detection is triggered to intervene in time before behavioral anomalies spread to all process nodes, so as to avoid the accumulation of deviations and systemic process instability. Continuing in S2, the process structure mapping deviation vector Fm is calculated, which is used to characterize the deviation patterns of personnel at each process node, because it is impossible to determine whether the deviation originates from individual nodes or global behavior based solely on overall behavioral fluctuations. In S3, a process structure consistency index Cp is constructed by mapping the process structure to a deviation vector Fm, and then the process matching index Rm is calculated. This is to determine the concentration and acceptability of deviations at the structural level. For example, when deviations are concentrated at key nodes, the process structure consistency index Cp will increase significantly. At this time, the decrease in the process matching index Rm has a clear physical meaning: personnel may have insufficient understanding, disordered rhythm, or operational deviations when executing key steps of the template, resulting in overall process mismatch. Finally, when the process matching index Rm is less than a threshold triggering the adaptation control mechanism, the system can immediately locate abnormal nodes based on the node deviation values ​​of the process structure mapping deviation vector Fm, avoiding the problem of "detecting mismatch but not finding the cause." Through the above-mentioned layer-by-layer detection method, the method can not only identify process anomalies in advance but also accurately locate the source of deviations and guide process correction in real time, thereby improving process stability, the accuracy of personnel-job matching judgment, and the quality of job operations. Example 2; please refer to Figure 2 Specifically: S1 includes S11; S11. Set up multiple data collection points on the personnel operation terminals of the enterprise's multi-channel customer service center, and set up corresponding data collection plug-ins in each data collection point to extract digital trajectory data summarized by customer service personnel during their job process operations in real time, and transmit it to the central processing system. The data collection points include operation rhythm data collection points, process node dwell data collection points, operation continuity and stability data collection points, system process template data collection points, and node behavior mapping data collection points; Digital trajectory data includes operation time interval parameter Te, process dwell time parameter Ts, rhythm stability parameter So, template time parameter Tp, and mapping deviation parameter △m; The operation rhythm acquisition point is set in the event trigger listening module of the operation terminal, and an operation timestamp acquisition plugin is embedded in the operation rhythm acquisition point to automatically record the event occurrence time when each operation event is triggered, with the operation time interval parameter Te between adjacent operation events. The process node dwell time collection point is set in the node status listening interface of the process control module, and the process node dwell time monitoring plugin is embedded in the process node dwell time collection point. It is used to record the node timestamp when personnel enter and exit the process node, so as to obtain the process dwell time parameter Ts of personnel at the process node. The continuous stability sampling points are set within the behavior sampling module of the behavior sequence buffer, and a rhythm stability analysis plugin is embedded in these sampling points. This plugin generates the rhythm stability parameter So by performing sliding window calculations on the sequence of operation events within multiple consecutive sampling windows. It should be noted that the rhythm stability analysis plugin first obtains the timestamps of operation events recorded in chronological order from the behavior sequence buffer. It automatically calculates the intervals between adjacent operation events and uses these time intervals as the basis for subsequent rhythm analysis. Since these time intervals reflect the speed of personnel's actions, they can directly characterize the rhythm of the personnel's workflow. Subsequently, the rhythm stability analysis plugin segments these time intervals according to a preset window method. Each window represents the personnel's operation within a continuous time period. For each window, it analyzes whether there are significant speed variations between the operation intervals within the window. For example, when multiple operation intervals are relatively close and without significant fluctuations, it indicates that the personnel's operation rhythm is relatively uniform; if the operation intervals within a window differ significantly, it indicates that the personnel's operation rhythm fluctuates within that segment of the workflow. The rhythm stability analysis plugin comprehensively evaluates the rhythm fluctuations of multiple windows to form the final rhythm stability parameter So. The rhythm stability parameter So is usually represented by a value within a limited range. The higher the value, the more stable the operation rhythm of the personnel when performing the job process. Conversely, when the rhythm stability parameter So decreases, it indicates that the personnel have inconsistent rhythms and fluctuate in speed when performing the process, and there is a trend of fluctuation in process behavior. The system process template collection point is set at the node time parameter interface of the job process template management module, and a template time parameter retrieval plugin is embedded in the system process template collection point to extract the template time parameter Tp of each process node from the job process template, so as to serve as the correlation data for comparing digital trajectory data with the job process structure. The node behavior mapping collection point is set within the behavior structure mapping interface of the process analysis module, and a node behavior deviation collection plugin is embedded in the node behavior mapping collection point. The mapping deviation parameter △m is calculated based on the time difference between personnel operation events and template nodes, so as to provide basic data for subsequent process structure deviation calculation.

[0023] S1 also includes S12; S12. The central processing system manages the acquisition scheduling strategy of multiple acquisition points in a unified manner and acquires digital trajectory data periodically at a unified sampling frequency. After acquiring the digital trajectory data, the extreme value normalization method is used to perform dimensionless processing on all parameters in the digital trajectory data to eliminate the influence of the unit dimension of all parameters in the digital trajectory data. The dimensionless digital trajectory data is then sorted according to the acquisition time to obtain the behavioral trajectory data sequence.

[0024] In this embodiment, the method sets up operation rhythm collection points, process node dwell collection points, operation continuity and stability collection points, system process template collection points, and node behavior mapping collection points on the personnel operation terminal. This enables the system to extract digital trajectory data such as operation time intervals, node dwell times, operation rhythm changes, template reference times, and node behavior deviations in real time when each key action occurs during process execution. The reason for using distributed collection from multiple collection points instead of a single collection pipeline is that process anomalies are often not caused by a single parameter, but by a "combination of offsets" of different process characteristics. For example, relying solely on node dwell time cannot identify rhythm changes, and relying solely on event time intervals cannot determine node jumps. Therefore, multi-point collaborative monitoring is necessary to capture the complete chain of anomaly formation. Simultaneously, by uniformly managing the collection scheduling strategy in the central processing system and performing extreme value normalization on all digital trajectory data, dimensionless data from different sources and with different dimensions are made available, allowing them to be compared and calculated under the same algorithm system. This avoids data distance distortion or deviation amplification caused by different units such as "seconds, counts, and deviation values." For example, without dimensionless processing, short operation intervals coupled with long node dwell times can lead to a single dimension dominating overall deviation judgments in subsequent algorithms, causing the system to misjudge process anomalies. By normalizing and reconstructing the behavioral trajectory data sequence in chronological order, the acquired behavioral trajectories are made continuous and comparable. This not only ensures the accuracy of subsequent behavioral stability calculations but also provides a clean and unified input foundation for in-depth analysis such as structural mapping deviations. Ultimately, this chain-like implementation of acquisition-cleaning-unified processing enables the system to avoid noise interference in the data dimension, capture abnormal trends in advance in the behavioral dimension, and build a reliable deviation baseline in the structural dimension. This significantly improves the sensitivity of process deviation identification, the accuracy of process anomaly localization, and the reliability of overall human-job matching detection results.

[0025] Example 3; please refer to Figure 2 Specifically: S2 includes S21; S21. In the central processing system, based on all operation time interval parameters Te in the behavior trajectory data sequence, calculate the standard deviation and mean respectively to obtain the standard deviation and mean of operation time interval parameter Te; then calculate the ratio of the standard deviation and mean of operation time interval parameter Te to obtain the behavior stability index Ks. A preliminary comparative evaluation is conducted based on the output results of the behavioral stability index Ks to determine the fluctuation trend of process behavior. The specific evaluation content is as follows: When the behavioral stability index Ks ≤ 0.32, it indicates that the behavioral process is stable; When the behavioral stability index Ks > 0.32, it indicates abnormal behavioral processes.

[0026] S2 also includes S22; S22. When the preliminary comparison and evaluation determines that the process behavior fluctuation trend is abnormal, and the process behavior is abnormal for three consecutive time windows, the process structure depth detection module is triggered. The process structure depth detection module performs process structure mapping processing on the behavior trajectory data sequence. Taking the node time parameter of the job process template as a reference, it performs node-level matching calculation by matching the operation time interval parameter Te of the digital trajectory data with the process dwell time parameter Ts, and generates the process structure mapping deviation vector Fm based on the process structure mapping deviation formula. The process structure mapping deviation vector Fm is calculated and output using the following process structure mapping deviation formula: In the formula, Fm i Te represents the process structure mapping deviation vector of the i-th node. i Tp represents the operation time interval parameter of the i-th node. i Ts represents the template time parameter of the i-th node. i Tp represents the process dwell time parameter at node i. s,i The template time parameter represents the duration (s) of the recommended node stay in the job process template of node i. The process structure mapping deviation formula used in this formula is mathematically derived from the Euclidean distance formula. This formula belongs to the "basic formula for n-dimensional spatial distance measurement" in mathematics and is a classic algorithm widely used in statistics, machine learning and signal analysis. This formula extends it to "measure the difference between operational behavior and job process template", which is an application and derivation of the classic formula in a specific engineering scenario. In this formula, the behavior of personnel at process nodes can be characterized by two continuous quantitative parameters: the operation event time interval Te. i Duration of process node Ts i Treat it as a two-dimensional vector ; The job template is a standard behavioral pattern, including: node recommendation event time interval Tp i Recommended dwell time Tp at nodes s,i The template vector is: ; The deviation in behavior is represented by the distance between two vectors, which is the formula used here; the purpose of using Euclidean distance is: Because behavioral deviations have the following characteristics: multiple indicators (event interval + dwell time) simultaneously affect process performance; all parameters must change simultaneously to reflect the "true deviation"; it is necessary to calculate the "true distance between behavior and standard"; Euclidean distance is the most natural "comprehensive deviation measure" that can fully reflect the degree of deviation; square: amplifies the impact of deviation; summation: reflects the simultaneous effect of two types of deviation; square root: restores the original dimension "second", which is convenient for physical interpretation; Physical meaning: When the process structure mapping deviation vector Fm is small, the actual behavior is close to the template, the rhythm is normal, and the process is stable; When the process structure mapping deviation vector Fm is large, the behavior deviates from the template, resulting in problems such as process stalling, skipping steps, and delays.

[0027] In this embodiment, the behavioral stability index Ks is calculated by dividing the standard deviation of the operation time interval parameter Te by its mean. The purpose is to quickly measure whether the rhythm of personnel operations is abnormal using the volatility of time series data. Observing only the absolute length of the time interval is easily affected by differences in process nodes. However, using the ratio of standard deviation to mean directly reflects the degree of rhythm fluctuation. For example, when personnel operate erratically, the standard deviation increases significantly, while the mean does not change significantly. In this case, the increase in the behavioral stability index Ks corresponds to the actual physical phenomenon of rhythm instability. Setting the threshold to 0.32 ensures that the system only triggers subsequent analysis when there is a significant rhythm drift, avoiding false alarms caused by individual operational fluctuations. Furthermore, requiring three consecutive windows to show anomalies before triggering deep detection prevents the instantaneous anomalies caused by brief operational pauses or accidental clicks from spreading to the structural analysis layer, thus stabilizing the system's judgment. Further, by comparing the behavioral trajectory data sequence with the job process template at the node level and calculating the process structure mapping deviation vector Fm, the system can determine whether personnel behavior deviates from the template at the structural level. The deviation calculation method based on Euclidean distance is adopted because behavioral deviations often stem from the simultaneous effects of two dimensions: "changes in operational rhythm" and "abnormal node dwell times." For example, prolonged dwell time alone cannot explain rhythm abnormalities, and rhythm abnormalities alone cannot determine node skipping. Euclidean distance can equivalently integrate these two types of deviations, truly reflecting the comprehensive distance relationship between behavior and the template. When the process structure mapping deviation vector Fm is small, it indicates that the behavior is highly consistent with the template, and the process is continuous and stable. When the process structure mapping deviation vector Fm is large, it indicates that changes in personnel rhythm, dwell times, or comprehension deviations have accumulated into considerable structural offsets, such as node stalls or skipped execution. By linking the behavioral stability index Ks at the behavioral layer with the process structure mapping deviation vector Fm at the structural layer, this implementation method can not only promptly identify behavioral fluctuations but also accurately locate structural deviations, thereby significantly improving the accuracy of process anomaly detection and the interpretability of process diagnosis.

[0028] Example 4; please refer to Figure 2 and Figure 3 Specifically: S3 includes S31; S31. The central processing system performs dispersion analysis on the deviation values ​​of each node in the process structure mapping deviation vector Fm, obtains the standard deviation and average value of the process structure mapping deviation vector Fm, and calculates the ratio to obtain the process structure consistency index Cp, forming intermediate data for matching judgment. This index constitutes the core data object of the intermediate data layer for matching judgment. The smaller the value, the more stable the personnel behavior and the more closely it fits the job process structure. The standard deviation of the process structure mapping deviation vector Fm represents the overall "fluctuation level" of the deviation values; the average value of the process structure mapping deviation vector Fm represents the average level of the deviation, that is, the "overall deviation level"; by calculating the ratio, we can obtain the proportion of the deviation fluctuation amplitude to the overall deviation level; that is, whether the deviation is uniform, stable, or has serious fluctuation points. Based on the output of the process structure consistency index Cp, the fluctuation of behavioral deviations is then assessed. The specific assessment criteria are as follows: When the process structure consistency index Cp≤0.3, it means that the process structure mapping deviation is concentrated among multiple process nodes, the degree of deviation of each process node is relatively consistent, and there is no obvious structural fluctuation trend when personnel execute job processes, which is a stable state of process structure. When the process structure consistency index Cp > 0.3, it indicates that the process structure mapping deviation is abnormal between different nodes. That is, personnel deviate excessively from the template at some nodes, while deviating less at other nodes. The overall behavior shows an uneven distribution of deviation, which is an unstable state of process structure.

[0029] S3 also includes S32; S32. When it is determined that the process structure is unstable, the process structure consistency index Cp is used as an input item to calculate the process matching index Rm. The process matching index Rm is calculated and output using the following algorithm formula: Rm=1 / 1+Cp.

[0030] S3 also includes S32; S32. The central processing system will use the calculated process matching index Rm as the output result to determine process matching. The specific determination is as follows: When the process matching index Rm≥0.68, it is determined to be a job process fit; When the process matching index Rm < 0.68, it is determined that the job process is not suitable.

[0031] S3 also includes S33; S33. When a job process is determined to be incompatible, an adaptation control mechanism is executed. This mechanism uses the central processing system to identify the process structure mapping deviation vector Fm of the i-th node from the process structure mapping deviation vector Fm of all nodes. i Generate node deviation report results and send the node deviation report results to the training system, monitoring system or process optimization system; Meanwhile, after receiving the node deviation report results from the central processing system, the operator's terminal automatically triggers interface assistance strategies based on the degree of node deviation, including rhythm prompts, key node reminders, and process structure guidance prompts, to correct process deviations and optimize operation rhythm in real time, thereby forming a closed-loop improvement mechanism for process execution.

[0032] In this embodiment, the method performs dispersion analysis on the node deviation values ​​of the process structure mapping deviation vector Fm and calculates the process structure consistency index Cp. Its purpose is to identify the "distribution pattern" of deviations among process nodes, rather than just the "magnitude" of the deviations. If only the average deviation is considered, situations may arise where the overall deviation is small, but individual key nodes deviate significantly, leading to distorted process judgment. The ratio of the standard deviation to the mean clearly reflects whether the deviation is concentrated in a few nodes, which is a crucial basis for assessors to determine if there is insufficient understanding or operational errors in key steps. When the process structure consistency index Cp is less than 0.3, it indicates a uniform deviation distribution; even if overall deviation exists, it indicates a problem of low proficiency. When the process structure consistency index Cp is greater than 0.3, it means that the deviation is distributed in local bursts, truly reflecting the physical phenomenon of structural anomalies at certain nodes. Furthermore, the process structure consistency index Cp is used to calculate the process matching index Rm, with 0.68 as the adaptation judgment threshold. This is to transform the continuous deviation distribution into a more easily interpreted matching degree expression; this threshold can effectively distinguish between two different types of problems: "process unfamiliarity" and "process incompatibility." For example, some personnel may have a slightly slower overall pace but their workflow sequence is correct, and they should not be judged as incompatible. The calculation of Rm is precisely to avoid such misjudgments. When a job workflow mismatch is ultimately determined, the system can directly locate the specific source of the anomaly by finding the node with the largest deviation and generating a node deviation report, avoiding the problem of manual node-by-node troubleshooting required in traditional methods. At the same time, the operator's terminal automatically triggers interface assistance strategies such as pace prompts and key node reminders based on the deviation report, so that the corrective action takes effect immediately during the workflow execution, preventing the anomaly from spreading further, thus forming a closed-loop mechanism of "automatic discovery, accurate location, and real-time correction". Through the above logical links, this implementation method significantly improves the accuracy of workflow anomaly identification and the ability to locate abnormal nodes, while enhancing the interpretability of workflow matching results and the real-time nature of operator correction.

[0033] Example 5; please refer to Figure 1 and Figure 2 A dynamic matching system for people and positions based on digital trajectories includes a process data acquisition module, a process structure mapping module, and a matching determination module; The process data acquisition module collects digital trajectory data of personnel during job process operations and inputs the digital trajectory data into the central processing system in real time; the central processing system then forms a behavioral trajectory data sequence based on the digital trajectory data. The process structure mapping module establishes a behavior stability index Ks based on behavior trajectory data sequences; it performs a preliminary comparative evaluation of the behavior stability index Ks to determine the fluctuation trend of process behavior; when the stability index Ks determines that the behavior process is abnormal, it triggers the process structure depth detection module to perform process structure mapping processing on the behavior trajectory data sequences and calculates the process structure mapping deviation vector Fm. The matching judgment module constructs a process structure consistency index Cp based on the process structure mapping deviation vector Fm, and uses it as input to calculate the process matching index Rm. It then performs process matching judgment on the process matching index Rm and outputs the detection result of the person-job matching process. When the detection result of the person-job matching process determines that the job process is not compatible, the adaptation control mechanism is executed.

[0034] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention.

Claims

1. A method for dynamic matching of people and positions based on digital trajectories, characterized in that: Includes the following steps: S1. Collect digital trajectory data of personnel during their job operation process and input the digital trajectory data into the central processing system in real time; the central processing system then forms a behavioral trajectory data sequence based on the digital trajectory data. S2. Establish the behavioral stability index Ks of operational behavior based on behavioral trajectory data sequences; A preliminary comparative evaluation of the behavior stability index Ks is conducted to determine the fluctuation trend of process behavior. When the stability index Ks indicates that the behavior process is abnormal, the process structure depth detection module is triggered to perform process structure mapping processing on the behavior trajectory data sequence and calculate the process structure mapping deviation vector Fm. S3. Based on the process structure mapping deviation vector Fm, construct the process structure consistency index Cp, and use it as input to calculate the process matching index Rm. Perform process matching judgment on the process matching index Rm, and output the detection result of the person-job matching process. When the detection result of the person-job matching process determines that the job process is not suitable, execute the adaptation control mechanism.

2. The method for dynamic matching of people and positions based on digital trajectories according to claim 1, characterized in that: S1 includes S11; S11. Set up multiple data collection points on the personnel operation terminals of the enterprise's multi-channel customer service center, and set up corresponding data collection plug-ins in each data collection point to extract digital trajectory data summarized by customer service personnel during their job process operations in real time, and transmit it to the central processing system. The data collection points include operation rhythm data collection points, process node dwell data collection points, operation continuity stability data collection points, system process template data collection points, and node behavior mapping data collection points. The digital trajectory data includes the operation time interval parameter Te, the process dwell time parameter Ts, the rhythm stability parameter So, the template time parameter Tp, and the mapping deviation parameter △m; The operation rhythm acquisition point is set in the event trigger listening module of the operation terminal, and an operation timestamp acquisition plugin is embedded in the operation rhythm acquisition point to automatically record the event occurrence time when each operation event is triggered, with the operation time interval parameter Te between adjacent operation events. The process node dwell time collection point is set in the node status listening interface of the process control module, and the process node dwell time monitoring plugin is embedded in the process node dwell time collection point. It is used to record the node timestamp when personnel enter and exit the process node, so as to obtain the process dwell time parameter Ts of personnel at the process node. Operational continuous stability acquisition points are set within the behavior sampling module of the behavior sequence buffer, and a rhythm stability analysis plugin is embedded in the operational continuous stability acquisition points. The plugin generates the rhythm stability parameter So by performing sliding window calculation on the operation event sequence within multiple consecutive sampling windows. The system process template collection point is set at the node time parameter interface of the job process template management module, and a template time parameter retrieval plugin is embedded in the system process template collection point to extract the template time parameter Tp of each process node from the job process template; The node behavior mapping collection point is set within the behavior structure mapping interface of the process analysis module, and a node behavior deviation collection plugin is embedded in the node behavior mapping collection point. The mapping deviation parameter △m is calculated based on the time difference between personnel operation events and template nodes.

3. The method for dynamic matching of people and positions based on digital trajectories according to claim 2, characterized in that: S1 further includes S12; S12. The central processing system manages the acquisition scheduling strategy of multiple acquisition points in a unified manner and acquires digital trajectory data periodically at a unified sampling frequency. After acquiring the digital trajectory data, the extreme value normalization method is used to perform dimensionless processing on all parameters in the digital trajectory data to eliminate the influence of the unit dimension of all parameters in the digital trajectory data. The dimensionless digital trajectory data is then sorted according to the acquisition time to obtain the behavioral trajectory data sequence.

4. The method for dynamic matching of people and positions based on digital trajectories according to claim 3, characterized in that: S2 includes S21; S21. In the central processing system, based on all operation time interval parameters Te in the behavior trajectory data sequence, calculate the standard deviation and mean respectively to obtain the standard deviation and mean of operation time interval parameter Te; then calculate the ratio of the standard deviation and mean of operation time interval parameter Te to obtain the behavior stability index Ks. A preliminary comparative evaluation is conducted based on the output results of the behavioral stability index Ks to determine the fluctuation trend of process behavior. The specific evaluation content is as follows: When the behavioral stability index Ks ≤ 0.32, it indicates that the behavioral process is stable; When the behavioral stability index Ks > 0.32, it indicates abnormal behavioral processes.

5. The method for dynamic matching of people and positions based on digital trajectories according to claim 4, characterized in that: S2 further includes S22; S22. When the preliminary comparison and evaluation determines that the process behavior fluctuation trend is abnormal, and the process behavior is abnormal for three consecutive time windows, the process structure depth detection module is triggered. The process structure depth detection module performs process structure mapping processing on the behavior trajectory data sequence, takes the node time parameter of the job process template as a reference, performs node-level matching calculation on the operation time interval parameter Te of the digital trajectory data and the process dwell time parameter Ts, and generates a process structure mapping deviation vector Fm based on the process structure mapping deviation formula. The process structure mapping deviation vector Fm is calculated and output using the following process structure mapping deviation formula: In the formula, Fm i Te represents the process structure mapping deviation vector of the i-th node. i Tp represents the operation time interval parameter of the i-th node. i Ts represents the template time parameter of the i-th node. i Tp represents the process dwell time parameter at node i. s,i The template time parameter represents the recommended node stay duration (s) in the job process template for node i.

6. The method for dynamic matching of people and positions based on digital trajectories according to claim 1, characterized in that: S3 includes S31; S31. The central processing system performs a discrete analysis on the deviation values ​​of each node in the process structure mapping deviation vector Fm, obtains the standard deviation and the average value of the process structure mapping deviation vector Fm, calculates the ratio, obtains the process structure consistency index Cp, and forms intermediate data for matching judgment. Based on the output of the process structure consistency index Cp, the fluctuation of behavioral deviations is then assessed. The specific assessment criteria are as follows: When the process structure consistency index Cp≤0.3, it means that the process structure mapping deviation is concentrated among multiple process nodes, the degree of deviation of each process node is relatively consistent, and there is no obvious structural fluctuation trend when personnel execute job processes, which is a stable state of process structure. When the process structure consistency index Cp > 0.3, it indicates that the process structure mapping deviation is abnormal between different nodes, which is an unstable state of process structure.

7. The method for dynamic matching of people and positions based on digital trajectories according to claim 6, characterized in that: S3 further includes S32; S32. When it is determined that the process structure is unstable, the process structure consistency index Cp is used as an input item to calculate the process matching index Rm. The process matching index Rm is calculated and output using the following algorithm formula: Rm=1 / 1+Cp.

8. The method for dynamic matching of people and positions based on digital trajectories according to claim 7, characterized in that: S3 further includes S32; S32. The central processing system will use the calculated process matching index Rm as the output result to determine process matching. The specific determination is as follows: When the process matching index Rm≥0.68, it is determined to be a job process fit; When the process matching index Rm < 0.68, it is determined that the job process is not suitable.

9. The method for dynamic matching of people and positions based on digital trajectories according to claim 8, characterized in that: S3 also includes S33; S33. When a job process is determined to be incompatible, an adaptation control mechanism is executed. This adaptation control mechanism identifies the process structure mapping deviation vector Fm of the i-th node from the process structure mapping deviation vector Fm of all nodes through the central processing system. i Generate node deviation report results and send the node deviation report results to the training system, monitoring system or process optimization system; Meanwhile, after receiving the node deviation report results from the central processing system, the operator's terminal automatically triggers interface assistance strategies based on the degree of node deviation, including rhythm prompts, key node reminders, and process structure guidance prompts.

10. A human-job dynamic matching system based on digital trajectory, applied to the human-job dynamic matching method based on digital trajectory as described in any one of claims 1-9, characterized in that: It includes a process data acquisition module, a process structure mapping module, and a matching and determination module; The process data acquisition module collects digital trajectory data of personnel during job process operations and inputs the digital trajectory data into the central processing system in real time; the central processing system then forms a behavioral trajectory data sequence based on the digital trajectory data. The process structure mapping module establishes a behavioral stability index Ks for operational behaviors based on behavioral trajectory data sequences. A preliminary comparative evaluation of the behavior stability index Ks is conducted to determine the fluctuation trend of process behavior. When the stability index Ks indicates that the behavior process is abnormal, the process structure depth detection module is triggered to perform process structure mapping processing on the behavior trajectory data sequence and calculate the process structure mapping deviation vector Fm. The matching judgment module constructs a process structure consistency index Cp based on the process structure mapping deviation vector Fm, and uses it as input to calculate the process matching index Rm. It then performs process matching judgment on the process matching index Rm and outputs the detection result of the person-job matching process. When the detection result of the person-job matching process determines that the job process is not compatible, the adaptation control mechanism is executed.

Images