Digital employee intelligent auxiliary task allocation engine work order monitoring method and system

Abstract

The invention discloses a digital employee intelligent auxiliary task allocation engine work order monitoring method and system, and relates to the technical field of work order anomaly detection, and the method comprises the steps: obtaining link time consumption data corresponding to each work order based on a preset work order state database, and carrying out the anomaly analysis of the link time consumption data corresponding to each work order, performing potential problem analysis on the time consumption data of each abnormal link based on a preset digital employee intelligent auxiliary task allocation engine tool to obtain a corresponding potential problem list, and performing abnormal problem screening on each problem record in the potential problem list based on a rule classification algorithm to obtain abnormal link time consumption data; and obtaining a corresponding abnormal problem set, and performing risk assessment on the abnormal problem set to obtain a corresponding work order monitoring result. The technical problems that according to a traditional work order monitoring method, the focal length is mostly smaller than the total consumed time of the work order, subtle state changes of all links of the work order cannot be accurately captured, potential risks are likely to be omitted, and the accuracy of work order monitoring is reduced are solved.

Description

Technical Field

[0001] This invention relates to the field of work order anomaly detection technology, and in particular to a work order monitoring method and system for a digital employee intelligent auxiliary task allocation engine. Background Technology

[0002] In modern enterprise management, task allocation and work order processing are core links to improve operational efficiency and ensure service quality. Especially in the context of digital transformation, how to monitor each work order is not only related to the efficient allocation of internal resources, but also directly affects customer satisfaction and business continuity.

[0003] Currently, most traditional work order monitoring methods focus on the total time spent on a work order, failing to accurately capture subtle changes in the status of each stage of the work order, easily overlooking potential risks and reducing the accuracy of work order monitoring. Summary of the Invention

[0004] This invention provides a digital employee intelligent auxiliary task allocation engine work order monitoring method and system, which solves the technical problem that traditional work order monitoring methods mostly focus on the total time of the work order, cannot accurately capture the subtle status changes of each stage of the work order, are prone to missing potential risks, and reduce the accuracy of work order monitoring.

[0005] The first aspect of this invention provides a work order monitoring method for a digital employee intelligent auxiliary task allocation engine, comprising:

[0006] Based on a pre-set work order status database, obtain the time consumption data of each step corresponding to each work order;

[0007] Anomaly analysis was performed on the time consumption data of each step corresponding to the work order to obtain the time consumption data of the corresponding abnormal steps.

[0008] Based on the preset digital employee intelligent auxiliary task allocation engine tool, the time consumption data of each abnormal link is analyzed for potential problems to obtain the corresponding potential problem list.

[0009] Based on a rule-based classification algorithm, abnormal problems are filtered out from each problem record in the potential problem list to obtain the corresponding abnormal problem set.

[0010] A risk assessment is performed on the set of abnormal issues to obtain the corresponding work order monitoring results.

[0011] Optionally, the step of performing anomaly analysis on the time consumption data of each work order to obtain the corresponding abnormal time consumption data includes:

[0012] A grouping and aggregation algorithm is used to classify the time consumption data of each stage and generate statistical results of the time consumption of each stage.

[0013] The kernel density estimation algorithm is used to extract features from the time consumption statistics of each stage to obtain the corresponding feature parameters;

[0014] Based on the aforementioned feature parameters, the time consumption data for each of the aforementioned steps are filtered to obtain the corresponding target step time consumption data;

[0015] The quality index data of the time consumption data of each target step is obtained, and the time consumption data of each target step is filtered according to the quality index data to obtain the time consumption data of the corresponding abnormal steps.

[0016] Optionally, the step of filtering the time consumption data of each target step based on the quality index data to obtain the time consumption data of the corresponding abnormal steps includes:

[0017] The quality index data are normalized to obtain the corresponding target quality index data.

[0018] Based on preset quality weights, the target quality index data are weighted and calculated to obtain the corresponding target quality evaluation score.

[0019] If the target quality assessment score is less than or equal to a preset assessment score threshold, the target process time data corresponding to the target quality assessment score will be identified as abnormal process time data.

[0020] Optionally, the step of analyzing the time consumption data of each abnormal step based on the preset digital employee intelligent auxiliary task allocation engine tool to obtain a corresponding list of potential problems includes:

[0021] The feedback content set of time consumption data of each abnormal link is obtained, and the preset digital employee intelligent auxiliary task allocation engine tool is called to perform sentiment analysis on each feedback content in the feedback content set to obtain multiple sentiment evaluation values.

[0022] When the emotional assessment value corresponding to the feedback content is less than the preset negative threshold, the feedback content is determined as the target feedback content.

[0023] Extract the first key sentence of each target feedback content based on a preset dictionary;

[0024] Each of the first key sentences is used to generate a corresponding list of potential problems.

[0025] Optionally, the step of filtering out anomalous problems from each problem record in the potential problem list based on a rule-based classification algorithm to obtain a corresponding set of anomalous problems includes:

[0026] Based on the rule-based classification algorithm, the impact of each problem record in the potential problem list is evaluated according to the preset evaluation rules, and multiple impact scores are obtained.

[0027] When the impact score is greater than the preset impact level threshold, the impact score and the corresponding problem record are identified as high-risk problems.

[0028] Each of the high-risk issues is labeled to obtain the corresponding target high-risk issues, and the corresponding set of abnormal issues is generated using each of the target high-risk issues.

[0029] Optionally, the step of performing a risk assessment on the set of abnormal issues to obtain the corresponding work order monitoring results includes:

[0030] The work order conversion time and processing quality data associated with the abnormal problem set are obtained in real time, and the corresponding first target key is generated using each of the processing quality data.

[0031] Each of the first target keys is used to retrieve a preset list of work order rating key-value pairs to obtain multiple work order ratings;

[0032] Each work order score and corresponding work order conversion time are used to generate a corresponding second target key;

[0033] Each of the second target keys is used to retrieve a preset list of quality assessment key-value pairs to obtain multiple quality assessment results.

[0034] When the quality assessment result shows abnormal fluctuations, the support vector machine algorithm is used to predict the risk of the work order conversion time and processing quality data associated with the quality assessment result, and obtain the corresponding risk warning level.

[0035] The corresponding work order monitoring results are generated using the quality assessment results and risk warning levels described above.

[0036] The second aspect of this invention provides a digital employee intelligent assisted task allocation engine work order monitoring system, comprising:

[0037] The data acquisition module is used to obtain the time consumption data of each process corresponding to each work order based on a preset work order status database.

[0038] The anomaly analysis module is used to perform anomaly analysis on the time consumption data of each step corresponding to each work order, and obtain the time consumption data of the corresponding abnormal steps.

[0039] The potential problem analysis module is used to perform potential problem analysis on the time consumption data of each abnormal link based on the preset digital employee intelligent auxiliary task allocation engine tool, and obtain the corresponding potential problem list.

[0040] The filtering module is used to filter out abnormal problems from each problem record in the potential problem list based on a rule-based classification algorithm, so as to obtain the corresponding abnormal problem set.

[0041] The risk assessment module is used to assess the risks of the abnormal problem set and obtain the corresponding work order monitoring results.

[0042] A third aspect of the present invention provides an electronic device, including a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor causes the processor to perform the steps of the digital employee intelligent auxiliary task allocation engine work order monitoring method as described in any of the preceding claims.

[0043] The fourth aspect of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program, when executed, implements the digital employee intelligent auxiliary task allocation engine work order monitoring method as described in any of the preceding claims.

[0044] The fifth aspect of the present invention provides a computer program product, the computer program product comprising a computer program stored on a non-transitory computer-readable storage medium, the computer program comprising program instructions, wherein, when the program instructions are executed by a computer, the computer performs the digital employee intelligent auxiliary task allocation engine work order monitoring method as described in any of the preceding claims.

[0045] As can be seen from the above technical solutions, the present invention has the following advantages:

[0046] This invention obtains the time consumption data for each step of a work order based on a pre-set work order status database. It then performs anomaly analysis on this data to identify anomalous step time consumption data. Using a pre-set digital employee intelligent auxiliary task allocation engine tool, it analyzes the potential problems in these anomalous step time consumption data to obtain a list of potential problems. Based on a rule-based classification algorithm, it filters out anomalous problems from each record in the potential problem list to obtain a set of anomalous problems. Finally, it conducts a risk assessment on this set of anomalous problems to obtain the corresponding work order monitoring results. This invention overcomes the technical problem that traditional work order monitoring methods mostly focus on the total time consumption of a work order, failing to accurately capture subtle changes in the status of each step, easily overlooking potential risks, and reducing the accuracy of work order monitoring. Compared with traditional work order monitoring methods, this invention obtains the time consumption data of each step in a pre-set work order status database, performs anomaly analysis on the time consumption data of each step in each work order to obtain the time consumption data of the corresponding abnormal steps, and combines the digital employee intelligent assisted task allocation engine tool and rule classification algorithm to perform potential problem analysis and anomaly problem screening on the time consumption data of each abnormal step to obtain the corresponding abnormal problem set. Then, the abnormal problem set is subjected to risk assessment to obtain the corresponding work order monitoring results, which effectively improves the accuracy of work order monitoring. Attached Figure Description

[0047] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0048] Figure 1 This is a flowchart illustrating the steps of a digital employee intelligent auxiliary task allocation engine work order monitoring method provided in Embodiment 1 of the present invention.

[0049] Figure 2 This is a flowchart illustrating the steps of a digital employee intelligent auxiliary task allocation engine work order monitoring method provided in Embodiment 2 of the present invention.

[0050] Figure 3 This is a structural block diagram of a digital employee intelligent auxiliary task allocation engine work order monitoring system provided in Embodiment 3 of the present invention;

[0051] Figure 4 This is a structural block diagram of an electronic device provided in Embodiment 4 of the present invention. Detailed Implementation

[0052] This invention provides a digital employee intelligent auxiliary task allocation engine work order monitoring method and system, which solves the technical problem that traditional work order monitoring methods mostly focus on the total time of the work order, cannot accurately capture the subtle status changes of each stage of the work order, are prone to missing potential risks, and reduce the accuracy of work order monitoring.

[0053] To make the objectives, features, and advantages of this invention more apparent and understandable, the technical solutions of the embodiments of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the embodiments described below are only some embodiments of this invention, and not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0054] Please see Figure 1 , Figure 1 The flowchart illustrates the steps of a work order monitoring method for a digital employee intelligent auxiliary task allocation engine provided in Embodiment 1 of the present invention.

[0055] This invention provides a work order monitoring method for a digital employee intelligent auxiliary task allocation engine, comprising:

[0056] Step 101: Based on the preset work order status database, obtain the time consumption data of each process corresponding to each work order.

[0057] The time consumption data of each step refers to the time interval between two adjacent state transitions in the work order processing flow. It is a quantitative record of the time consumption of a single specific step in the work order processing process.

[0058] In this embodiment of the invention, based on a preset work order status database, the status transition timestamp corresponding to the work order number in each work order is obtained, and the process time data corresponding to each work order is calculated according to the status transition node corresponding to each status transition timestamp. For example, the work order status database stores 10,000 work order records. Each record contains a work order ID, a status transition node (such as submit, review, processing, and completion), and a corresponding status transition timestamp. The database extracts the status transition timestamps for each work order at different stages through a query (e.g., SELECT (i.e., Structured Query Language) Work Order ID, Status, Timestamp FROM Work Order Status Table WHERE Status IN ('Submit', 'Review', 'Processing', 'Completed') ORDER BY (i.e., query sorting keyword) Work Order ID, Timestamp). Then, for each stage's time consumption, the difference in status transition timestamps between adjacent status transition nodes is calculated to obtain the corresponding stage's time consumption data (e.g., a work order takes 2.5 hours from submission to review and 5.8 hours from review to processing; the algorithm uses timestamp subtraction, i.e., time consumption = next status timestamp - current status timestamp), and the result is stored in a temporary table to obtain the time consumption data for each stage (i.e., stage time consumption data)).

[0059] It should be noted that the state transition timestamp refers to the specific time point recorded when a work order transitions from one state to another during the processing flow. It is the basic information used to calculate the time consumption of each step and analyze the efficiency of state transitions.

[0060] Step 102: Perform anomaly analysis on the time consumption data of each work order to obtain the time consumption data of the corresponding abnormal process.

[0061] Abnormal process time data refers to time interval data that exceeds a preset threshold.

[0062] In this embodiment of the invention, a grouping and aggregation algorithm is used to classify the time consumption data of each stage, generating statistical results of the time consumption for each stage. A kernel density estimation algorithm is used to extract features from the statistical results of the time consumption for each stage, obtaining corresponding feature parameters. Based on the feature parameters, the time consumption data of each stage is filtered to obtain the corresponding target stage time consumption data. Quality index data of the time consumption data of each target stage is obtained, and the time consumption data of each target stage is filtered according to the quality index data to obtain the corresponding abnormal stage time consumption data.

[0063] Step 103: Based on the preset digital employee intelligent auxiliary task allocation engine tool, perform potential problem analysis on the time consumption data of each abnormal link to obtain the corresponding potential problem list.

[0064] It should be noted that the digital employee intelligent assisted task allocation engine tool refers to a natural language processing tool.

[0065] A potential problem list refers to a summary list of potential hidden problems identified by analyzing user feedback based on abnormal process time data, which includes negative comments or descriptions of unresolved issues.

[0066] In this embodiment of the invention, a set of feedback content data on the time consumption of each abnormal step is obtained. A preset natural language processing tool is used to perform sentiment analysis on each feedback content in the set, resulting in multiple sentiment evaluation values. When the sentiment evaluation value corresponding to a feedback content is less than a preset negative threshold, the feedback content is identified as the target feedback content. The first key sentence of each target feedback content is extracted based on a preset dictionary. A corresponding list of potential problems is generated using each first key sentence.

[0067] Step 104: Based on the rule-based classification algorithm, perform abnormal problem screening on each problem record in the potential problem list to obtain the corresponding abnormal problem set.

[0068] An abnormal problem set refers to a summary collection of high-risk abnormal problems and their classification labels obtained after classifying, filtering and marking a list of potential hidden problems.

[0069] In this embodiment of the invention, based on a rule-based classification algorithm, the impact of each problem record in the potential problem list is evaluated according to preset evaluation rules, resulting in multiple impact scores. When the impact score is greater than a preset impact level threshold, the impact score and the corresponding problem record are identified as high-risk problems. Each high-risk problem is labeled to obtain the corresponding target high-risk problem, and a corresponding set of abnormal problems is generated using each target high-risk problem.

[0070] Step 105: Conduct a risk assessment on the set of abnormal issues and obtain the corresponding work order monitoring results.

[0071] In this embodiment of the invention, the work order conversion time and processing quality data associated with the abnormal problem set are acquired in real time, and a corresponding first target key is generated using each processing quality data. A preset work order rating key-value pair list is retrieved using each first target key to obtain multiple work order ratings. A corresponding second target key is generated using each work order rating and its corresponding work order conversion time. A preset quality assessment key-value pair list is retrieved using each second target key to obtain multiple quality assessment results. When the quality assessment result shows abnormal fluctuation, a support vector machine algorithm is used to perform risk prediction on the work order conversion time and processing quality data associated with the quality assessment result to obtain the corresponding risk warning level. A corresponding work order monitoring result is generated using each quality assessment result and each risk warning level.

[0072] In this embodiment of the invention, time consumption data for each step of a work order is obtained based on a preset work order status database. Anomaly analysis is performed on the time consumption data for each step of the work order to obtain corresponding abnormal step time consumption data. A potential problem analysis is then performed on the abnormal step time consumption data using a preset digital employee intelligent auxiliary task allocation engine tool to obtain a corresponding potential problem list. Based on a rule-based classification algorithm, abnormal problems are filtered from each problem record in the potential problem list to obtain a corresponding abnormal problem set. A risk assessment is then performed on the abnormal problem set to obtain the corresponding work order monitoring result. This overcomes the technical problem that traditional work order monitoring methods mostly focus on the total time consumption of the work order, failing to accurately capture subtle status changes in each step of the work order, easily overlooking potential risks, and reducing the accuracy of work order monitoring. Compared with traditional work order monitoring methods, this invention obtains the time consumption data of each step in a pre-set work order status database, performs anomaly analysis on the time consumption data of each step in each work order to obtain the time consumption data of the corresponding abnormal steps, and combines the digital employee intelligent assisted task allocation engine tool and rule classification algorithm to perform potential problem analysis and anomaly problem screening on the time consumption data of each abnormal step to obtain the corresponding abnormal problem set. Then, the abnormal problem set is subjected to risk assessment to obtain the corresponding work order monitoring results, which effectively improves the accuracy of work order monitoring.

[0073] Please see Figure 2 , Figure 2 The flowchart illustrates the steps of a digital employee intelligent auxiliary task allocation engine work order monitoring method provided in Embodiment 2 of the present invention.

[0074] This invention provides a work order monitoring method for a digital employee intelligent auxiliary task allocation engine, comprising:

[0075] Step 201: Based on the preset work order status database, obtain the time consumption data of each process corresponding to each work order.

[0076] In this embodiment of the invention, based on a preset work order status database, the status transition timestamp corresponding to the work order number in each work order is obtained, and the process time data corresponding to each work order is calculated according to the status transition node corresponding to each status transition timestamp.

[0077] Step 202: Use a grouping and aggregation algorithm to classify the time consumption data of each stage and generate statistical results of the time consumption of each stage.

[0078] In this embodiment of the invention, a grouping aggregation algorithm is used to aggregate the time consumption data of each stage (such as submission to review) and calculate the average time, maximum time, minimum time and standard deviation of each stage (for example, the average time of submission to review is 3.2 hours and the standard deviation is 1.5 hours). The average time, maximum time, minimum time, standard deviation and stage time consumption data of each stage are used as the stage time consumption statistics results.

[0079] Step 203: Use the kernel density estimation algorithm to extract features from the time consumption statistics of each stage to obtain the corresponding feature parameters.

[0080] In this embodiment of the invention, a kernel density estimation algorithm is used to extract the quantiles (e.g., 50% quantile, 90% quantile, etc.) of the time consumption statistics of each stage. The 90% quantile is taken as the historical baseline value, and each historical baseline value is used as the corresponding feature parameter. The feature parameter includes multiple historical baseline values.

[0081] Step 204: Filter the time consumption data of each step based on the feature parameters to obtain the corresponding target step time consumption data.

[0082] In this embodiment of the invention, the time consumption data of each step is filtered based on feature parameters to obtain the corresponding target step time consumption data. For example, when the time consumption data of a step is greater than the corresponding historical benchmark value, the step time consumption data is determined as the target step time consumption data.

[0083] Step 205: Obtain the quality indicator data of the time consumption data of each target step, and filter the time consumption data of each target step according to the quality indicator data to obtain the time consumption data of the corresponding abnormal steps.

[0084] Furthermore, step 205 includes the following sub-steps:

[0085] S11. Normalize the data of each quality indicator to obtain the corresponding target quality indicator data.

[0086] Quality metrics data refer to the processing time, first response time, resolution rate, and customer satisfaction score for each step in a work order.

[0087] Target quality indicator data refers to the normalized quality indicator data.

[0088] In this embodiment of the invention, quality indicator data corresponding to the time consumption data of each target step is obtained from the work order database (for example, work order records within the corresponding time period are extracted from the work order database using the timestamps of the target step time consumption data, and then quality indicator data is extracted from the work order processing records). The quality indicator data is then normalized to obtain the corresponding target quality indicator data.

[0089] S12. Based on the preset quality weights, perform weighted calculations on the data of each target quality indicator to obtain the corresponding target quality assessment score.

[0090] Quality weights refer to the percentage of processing time, first response time, resolution rate, and customer satisfaction rating. For example, quality weights could be set as follows: processing time 20%, first response time 30%, resolution rate 30%, and customer satisfaction 20%.

[0091] In this embodiment of the invention, based on preset quality weights, the data of each target quality indicator are weighted and calculated to obtain the corresponding target quality assessment score. For example, when the processing time (2.5 hours), first response time (0.5 hours), resolution rate (90%), and customer satisfaction score (4.2 points out of 5) are all within the specified range, the target quality assessment score is calculated as follows: (2.5 / 5)*20% + (0.5 / 1)*30% + 90%*30% + (4.2 / 5)*20% = 0.658 (i.e., 65.8 points).

[0092] S13. When the target quality assessment score is less than or equal to the preset assessment score threshold, the target process time data corresponding to the target quality assessment score is determined as abnormal process time data.

[0093] The evaluation score threshold refers to the critical value at which the time consumption data of the target process is deemed abnormal. The value is set at 65 points.

[0094] In this embodiment of the invention, when the target quality assessment score is less than or equal to 65 points, the target process time data corresponding to the target quality assessment score is determined as abnormal process time data.

[0095] Step 206: Based on the preset digital employee intelligent auxiliary task allocation engine tool, perform potential problem analysis on the time consumption data of each abnormal link to obtain the corresponding potential problem list.

[0096] Furthermore, step 206 includes the following sub-steps:

[0097] S21. Obtain the feedback content set of time consumption data for each abnormal step, call the preset digital employee intelligent auxiliary task allocation engine tool to perform sentiment analysis on each feedback content in each feedback content set, and obtain multiple sentiment evaluation values.

[0098] In this embodiment of the invention, a set of feedback content data on the time consumption of each abnormal step is obtained. A preset digital employee intelligent auxiliary task allocation engine tool (a sentiment analysis algorithm in natural language processing technology) is invoked to perform sentiment analysis on each feedback content in the set, resulting in multiple sentiment evaluation values. For example, a natural language processing algorithm is invoked to perform sentiment analysis on each feedback content, setting the sentiment score range to -1 to 1, resulting in multiple sentiment evaluation values.

[0099] S22. When the sentiment assessment value corresponding to the feedback content is less than the preset negative threshold, the feedback content is determined as the target feedback content.

[0100] The negative threshold refers to the critical value at which feedback is negative. Its value is -0.5.

[0101] In this embodiment of the invention, when the sentiment assessment value corresponding to the feedback content is less than -0.5, the feedback content is determined as the target feedback content. For example, if a piece of feedback content is "The problem has been delayed for too long and has not been effectively resolved," and its sentiment analysis score is -0.7, the system will classify it as the target feedback content.

[0102] S23. Extract the first key sentence of each target feedback content based on the preset dictionary.

[0103] In this embodiment of the invention, the first key sentence of each target feedback content is extracted based on a preset dictionary, wherein the first key sentence contains descriptions such as "unresolved", "delayed", and "unsatisfactory".

[0104] It is worth mentioning that a keyword matching algorithm can be used to identify whether each target feedback content contains keywords such as "unresolved," "delayed," and "unsatisfactory." The frequency of these keywords in the target feedback content can be counted. For example, if "unresolved" appears 3 times, accounting for 15% of the total number of words in the feedback, exceeding the 10% warning line, the target feedback content is identified as the first key sentence.

[0105] S24. Generate a list of potential problems using each first key sentence.

[0106] In this embodiment of the invention, a corresponding list of potential problems is generated using each first key sentence, wherein the list of potential problems includes multiple first key sentences.

[0107] In another embodiment, the first key sentences can be sorted from high to low based on the proportion of keywords such as "unresolved," "delayed," and "unsatisfactory" in the first key sentences, thereby generating a corresponding list of potential problems.

[0108] It is worth mentioning that cluster analysis can be performed on each first key sentence and each target feedback content to generate a list of potential problems, thereby avoiding the omission of the first key sentence in the target feedback content.

[0109] It is worth mentioning that if a potential problem is among the most frequent issues, relevant feedback content is extracted from the real-time feedback information database to verify the accuracy of the potential problem list, resulting in a verified potential problem list. If the accuracy of the potential problem list is low, a keyword matching algorithm is used to re-extract the first key sentence of each target feedback content, and execution proceeds to step S24.

[0110] Step 207: Based on the rule-based classification algorithm, perform abnormal problem screening on each problem record in the potential problem list to obtain the corresponding abnormal problem set.

[0111] Furthermore, step 207 includes the following sub-steps:

[0112] S31. Based on the rule-based classification algorithm, the impact of each problem record in the potential problem list is evaluated according to the preset evaluation rules, and multiple impact scores are obtained.

[0113] It should be noted that the evaluation rules are as follows: if the frequency of the problem is greater than or equal to 5 times per week and the number of departments affected exceeds 3, the severity is rated as "high" with a quantitative value of 8 (out of 10); if the frequency is between 2 and 4 times and the number of departments affected is 1 to 2, it is rated as "medium" with a value of 5; the rest are rated as "low" with a value of 2. At the same time, the scope of impact is calculated by multiplying the number of departments by a weight of 0.5.

[0114] In this embodiment of the invention, based on a rule-based classification algorithm, the severity and scope of impact of each problem record in the potential problem list are analyzed according to preset evaluation rules to obtain multiple impact scores. For example, if a problem occurs 6 times and affects 4 departments, its severity value is 8, and its impact score is 4 multiplied by 0.5, which equals 2.0.

[0115] S32. When the impact score is greater than the preset impact level threshold, the impact score and the corresponding problem record are identified as high-risk problems.

[0116] The impact threshold is the critical value used to classify an issue as a high-risk issue. Its value is 6.

[0117] In this embodiment of the invention, it is determined whether each impact score is greater than a preset impact level threshold. When the impact score is greater than 6, the impact score and the corresponding problem record are identified as high-risk problems.

[0118] S33. Label each high-risk problem to obtain the corresponding target high-risk problem, and use each target high-risk problem to generate the corresponding set of abnormal problems.

[0119] In this embodiment of the invention, anomaly classification labels (such as "high-risk system failure") are generated based on each high-risk issue. These labels are then used to annotate the corresponding high-risk issues, resulting in target high-risk issues. Finally, a set of corresponding anomalous issues is generated using each target high-risk issue. For example, assuming that 200 out of 1000 issues have an impact score exceeding 6, the system generates anomaly classification labels and records the impact score and corresponding issue records, including the issue ID, severity value of 8, and impact score of 2.0, thus obtaining the corresponding high-risk issues. These anomaly classification labels are then used to annotate the corresponding high-risk issues, resulting in target high-risk issues. Finally, a set of corresponding anomalous issues is generated using each target high-risk issue.

[0120] It is worth mentioning that a preset information processing module can be used to verify the anomaly classification label of each anomaly in the anomaly problem set. If the verification finds that the anomaly classification label does not match the anomaly features, the anomaly problem is re-analyzed in depth to obtain the verified anomaly problem set.

[0121] Step 208: Conduct a risk assessment on the abnormal problem set and obtain the corresponding work order monitoring results.

[0122] Furthermore, step 208 includes the following sub-steps:

[0123] S41. Obtain the work order conversion time and processing quality data associated with the abnormal problem set in real time, and generate the corresponding first target key using each processing quality data.

[0124] Work order transition time refers to the time interval during which a work order transitions from one state to the next in the processing flow.

[0125] Processing quality data refers to multi-dimensional quality indicator information extracted from work order processing records to evaluate the effectiveness of work order processing. It is the core data for measuring the quality of work order processing.

[0126] In this embodiment of the invention, the conversion time and processing quality data of the work orders associated with the abnormal problem set are obtained in real time through the Kafka streaming platform (i.e., Kafka streaming platform), and the corresponding first target key is generated using each processing quality data.

[0127] S42. Retrieve the preset work order rating key-value pair list using each first target key to obtain multiple work order ratings.

[0128] In this embodiment of the invention, each first target key is input into a preset work order scoring key-value pair list to obtain multiple work order scores.

[0129] S43. Generate the corresponding second target key by using the work order score and the corresponding work order conversion time respectively.

[0130] In this embodiment of the invention, a second target key is generated using each work order score and the corresponding work order conversion time, wherein the second target key includes the work order score and the corresponding work order conversion time.

[0131] S44. Retrieve the preset quality assessment key-value pair list using each second target key to obtain multiple quality assessment results.

[0132] In this embodiment of the invention, a preset list of quality assessment key-value pairs is retrieved using each second target key to obtain multiple quality assessment results. For example, if the work order conversion time in the second target key is 35 minutes and the work order score is 60, a normal quality assessment result is generated when the work order conversion time is less than or equal to a preset delay threshold (40 minutes) and the work order score is greater than or equal to a preset quality threshold (65). When the work order conversion time is greater than the delay threshold or the work order score is less than the quality threshold, an abnormally fluctuating quality assessment result is generated.

[0133] It should be noted that the delay threshold can be taken as the average of the conversion times of the most recent 1000 work orders plus twice the standard deviation of the work order conversion times (for example, if the average of the conversion times of 1000 work orders is 30 minutes and the standard deviation is 5 minutes, then the delay threshold is 40 minutes). The quality threshold can be taken as the average of the ratings of 1000 work orders plus twice the standard deviation of the work order ratings (for example, if the average of the ratings of 1000 work orders is 85 and the standard deviation is 10, then the delay threshold is 65).

[0134] S45. When the quality assessment result shows abnormal fluctuation, the support vector machine algorithm is used to predict the risk of the work order conversion time and processing quality data associated with the quality assessment result, and obtain the corresponding risk warning level.

[0135] In this embodiment of the invention, when the quality assessment result is abnormally fluctuating, the support vector machine algorithm is used to predict the risk of the work order conversion time and processing quality data associated with the quality assessment result, and to determine the corresponding risk warning level.

[0136] S46. Generate corresponding work order monitoring results using the results of each quality assessment and each risk warning level.

[0137] In this embodiment of the invention, each quality assessment result and each risk warning level are used as the corresponding work order monitoring results.

[0138] In this embodiment of the invention, time consumption data for each step of a work order is obtained based on a preset work order status database. Anomaly analysis is performed on the time consumption data for each step of the work order to obtain corresponding abnormal step time consumption data. A potential problem analysis is then performed on the abnormal step time consumption data using a preset digital employee intelligent auxiliary task allocation engine tool to obtain a corresponding potential problem list. Based on a rule-based classification algorithm, abnormal problems are filtered from each problem record in the potential problem list to obtain a corresponding abnormal problem set. A risk assessment is then performed on the abnormal problem set to obtain the corresponding work order monitoring result. This overcomes the technical problem that traditional work order monitoring methods mostly focus on the total time consumption of the work order, failing to accurately capture subtle status changes in each step of the work order, easily overlooking potential risks, and reducing the accuracy of work order monitoring. Compared with traditional work order monitoring methods, this invention obtains the time consumption data of each step in a pre-set work order status database, performs anomaly analysis on the time consumption data of each step in each work order to obtain the time consumption data of the corresponding abnormal steps, and combines the digital employee intelligent assisted task allocation engine tool and rule classification algorithm to perform potential problem analysis and anomaly problem screening on the time consumption data of each abnormal step to obtain the corresponding abnormal problem set. Then, the abnormal problem set is subjected to risk assessment to obtain the corresponding work order monitoring results, which effectively improves the accuracy of work order monitoring.

[0139] Please see Figure 3 , Figure 3 This is a structural block diagram of a digital employee intelligent auxiliary task allocation engine work order monitoring system provided in Embodiment 3 of the present invention.

[0140] This invention provides a digital employee intelligent assisted task allocation engine work order monitoring system, comprising:

[0141] The data acquisition module 301 is used to acquire the time consumption data of each process corresponding to each work order based on a preset work order status database.

[0142] Anomaly analysis module 302 is used to perform anomaly analysis on the time consumption data of each work order and obtain the time consumption data of the corresponding abnormal process.

[0143] The potential problem analysis module 303 is used to perform potential problem analysis on the time consumption data of each abnormal link based on the preset digital employee intelligent auxiliary task allocation engine tool, and obtain the corresponding potential problem list.

[0144] The filtering module 304 is used to filter out abnormal problems from each problem record in the potential problem list based on a rule-based classification algorithm, and obtain the corresponding abnormal problem set.

[0145] The risk assessment module 305 is used to perform risk assessment on the set of abnormal issues and obtain the corresponding work order monitoring results.

[0146] Furthermore, the anomaly analysis module 302 includes:

[0147] The stage classification submodule is used to classify the time consumption data of each stage using a grouping and aggregation algorithm, and generate statistical results of the time consumption of each stage.

[0148] The feature extraction submodule is used to extract features from the time consumption statistics of each stage using the kernel density estimation algorithm to obtain the corresponding feature parameters.

[0149] The first filtering submodule is used to filter the time consumption data of each step based on the feature parameters to obtain the time consumption data of the corresponding target step.

[0150] The second filtering submodule is used to obtain the quality indicator data of the time consumption data of each target step, and to filter the time consumption data of each target step according to the quality indicator data to obtain the time consumption data of the corresponding abnormal steps.

[0151] Furthermore, the second screening submodule includes:

[0152] The normalization unit is used to normalize the data of each quality indicator to obtain the corresponding target quality indicator data.

[0153] The weighting unit is used to perform weighted calculations on the data of each target quality indicator based on preset quality weights to obtain the corresponding target quality assessment score.

[0154] The filtering unit is used to identify the time consumption data of the target process corresponding to the target quality assessment score as abnormal process time consumption data when the target quality assessment score is less than or equal to a preset assessment score threshold.

[0155] Furthermore, the potential problem analysis module 303 includes:

[0156] The sentiment analysis submodule is used to obtain the feedback content set of time consumption data of each abnormal link, and call the preset digital employee intelligent auxiliary task allocation engine tool to perform sentiment analysis on each feedback content in each feedback content set to obtain multiple sentiment evaluation values.

[0157] The first analysis submodule is used to determine the feedback content as the target feedback content when the sentiment assessment value corresponding to the feedback content is less than the preset negative threshold.

[0158] Extract the first key sentence of each target feedback content based on a pre-set dictionary;

[0159] Each first key sentence is used to generate a corresponding list of potential problems.

[0160] Furthermore, the filtering module 304 includes:

[0161] The impact assessment submodule is used to assess the impact of each problem record in the potential problem list based on a rule-based classification algorithm and according to preset assessment rules, and obtain multiple impact scores.

[0162] The second analysis submodule is used to identify the impact score and the corresponding problem record as high-risk problems when the impact score is greater than the preset impact level threshold.

[0163] Each high-risk problem is labeled to obtain the corresponding target high-risk problem, and the corresponding set of abnormal problems is generated using each target high-risk problem.

[0164] Furthermore, risk assessment module 305 includes:

[0165] The third analysis submodule is used to obtain the work order conversion time and processing quality data associated with the abnormal problem set in real time, and generate the corresponding first target key using each processing quality data.

[0166] The fourth analysis submodule is used to retrieve the preset list of work order rating key-value pairs using each first target key to obtain multiple work order ratings;

[0167] The corresponding second target key is generated using the work order score and the corresponding work order conversion time;

[0168] Each second target key is used to retrieve a pre-defined list of quality assessment key-value pairs to obtain multiple quality assessment results.

[0169] The fifth analysis submodule is used to predict the risk of work order conversion time and processing quality data associated with the quality assessment results when the quality assessment results show abnormal fluctuations, and obtain the corresponding risk warning level by using the support vector machine algorithm.

[0170] The corresponding work order monitoring results are generated using the results of each quality assessment and each risk warning level.

[0171] Please see Figure 4 , Figure 4 This is a structural block diagram of an electronic device provided in Embodiment 4 of the present invention.

[0172] An electronic device according to an embodiment of the present invention includes: a memory 401 and a processor 402. The memory 401 stores a computer program. When the computer program is executed by the processor 402, the processor 402 executes the digital employee intelligent auxiliary task allocation engine work order monitoring method as described in any of the above embodiments.

[0173] Memory 401 may be an electronic memory such as flash memory, EEPROM (Electrically Erasable Programmable Read-Only Memory), EPROM, hard disk, or ROM. Memory 401 has storage space 403 for program code 413 for performing any of the method steps described above. For example, storage space 403 for program code may include individual program codes 413 for implementing the various steps in the methods described above. This program code may be read from or written to one or more computer program products. These computer program products include program code carriers such as hard disks, CDs, memory cards, or floppy disks. The program code may be compressed, for example, in a suitable form. When run by a computing processing device, this code causes the computing processing device to perform the various steps in the methods described above. This program code may be read from or written to one or more computer program products. These computer program products include program code carriers such as hard disks, CDs, memory cards, or floppy disks. The program code may be compressed, for example, in a suitable form. When this code is run by a computing device, it causes the computing device to execute the various steps in the digital employee intelligent assisted task allocation engine work order monitoring method described above.

[0174] Embodiment 5 of the present invention also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the digital employee intelligent auxiliary task allocation engine work order monitoring method as described in any of the above embodiments.

[0175] Embodiment 6 of the present invention also provides a computer program product, which includes a computer program stored on a non-transitory computer-readable storage medium. The computer program includes program instructions, wherein when the program instructions are executed by a computer, the computer executes the digital employee intelligent auxiliary task allocation engine work order monitoring method as described in any of the above embodiments.

[0176] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0177] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces, or indirect coupling or communication connection between apparatuses or units, and may be electrical, mechanical, or other forms.

[0178] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0179] Furthermore, the functional units in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit.

[0180] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0181] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A digital employee intelligence assisted task assignment engine ticket monitoring method, characterized in that, The method comprises the following steps: Based on the preset work order state database, the link time consumption data corresponding to each work order is obtained; Abnormal analysis is performed on the link time consumption data corresponding to each work order to obtain corresponding abnormal link time consumption data; Based on the preset digital employee intelligent auxiliary task allocation engine tool, potential problem analysis is performed on each abnormal link time consumption data to obtain a corresponding potential problem list; Based on a rule classification algorithm, abnormal problem screening is performed on each problem record in the potential problem list to obtain a corresponding abnormal problem set; Risk assessment is performed on the abnormal problem set to obtain a corresponding work order monitoring result.

2. The digital employee intelligence assisted task assignment engine ticket monitoring method of claim 1, wherein, The step of performing abnormal analysis on the link time consumption data corresponding to each work order to obtain corresponding abnormal link time consumption data comprises the following steps: A grouping aggregation algorithm is used to perform stage classification processing on each link time consumption data to generate stage time consumption statistical results; A kernel density estimation algorithm is used to perform feature extraction on each stage time consumption statistical result to obtain corresponding feature parameters; Based on the feature parameters, each link time consumption data is screened to obtain corresponding target link time consumption data; The quality indicator data of each target link time consumption data is obtained, and each target link time consumption data is screened according to each quality indicator data to obtain corresponding abnormal link time consumption data.

3. The digital employee intelligence assisted task assignment engine ticket monitoring method of claim 2, wherein, The step of screening each target link time consumption data according to each quality indicator data to obtain corresponding abnormal link time consumption data comprises the following steps: Each quality indicator data is normalized to obtain corresponding target quality indicator data; Based on a preset quality weight, a weighted operation is performed on each target quality indicator data to obtain a target quality evaluation score; When the target quality evaluation score is less than or equal to a preset evaluation score threshold, the target link time consumption data corresponding to the target quality evaluation score is determined as abnormal link time consumption data.

4. The digital employee intelligent assistant task assignment engine ticket monitoring method of claim 1, wherein, The step of performing potential problem analysis on each abnormal link time consumption data based on the preset digital employee intelligent auxiliary task allocation engine tool to obtain a corresponding potential problem list comprises the following steps: A feedback content set of each abnormal link time consumption data is obtained, a preset digital employee intelligent auxiliary task allocation engine tool is called to perform sentiment analysis on each feedback content in each feedback content set to obtain a plurality of sentiment evaluation values; When the sentiment evaluation value corresponding to the feedback content is less than a preset negative threshold, the feedback content is determined as target feedback content; Based on a preset dictionary, a first key sentence of each target feedback content is extracted; Each first key sentence is used to generate a corresponding potential problem list.

5. The digital employee intelligence assisted task assignment engine ticket monitoring method of claim 1, wherein, The step of performing abnormal problem screening on each problem record in the potential problem list based on a rule classification algorithm to obtain a corresponding abnormal problem set comprises the following steps: Based on a rule classification algorithm, the influence score of each problem record in the potential problem list is obtained according to a preset evaluation rule; When the influence score is greater than a preset influence degree threshold, the influence score and the corresponding problem record are determined as a high-risk problem. The high-risk problems are marked to obtain corresponding target high-risk problems, and the target high-risk problems are used to generate a corresponding abnormal problem set.

6. The digital employee intelligence assisted task assignment engine ticket monitoring method of claim 1, wherein, The step of performing risk assessment on the abnormal problem set to obtain a corresponding work order monitoring result comprises: Real-time acquisition of work order conversion time and processing quality data associated with the abnormal problem set, and generation of a corresponding first target key using each processing quality data; Each of the first target keys is used to retrieve a preset work order score key-value pair list to obtain a plurality of work order scores; Each of the work order scores and the corresponding work order conversion time is used to generate a corresponding second target key; Each of the second target keys is used to retrieve a preset quality assessment key-value pair list to obtain a plurality of quality assessment results; When the quality assessment result is a fluctuation anomaly, a support vector machine algorithm is used to perform risk prediction on the work order conversion time and processing quality data associated with the quality assessment result to obtain a corresponding risk warning level; Each of the quality assessment results and the risk warning levels is used to generate a corresponding work order monitoring result.

7. A digital workforce intelligence assisted task assignment engine ticket monitoring system, characterized in that, Comprise: The acquisition module is configured to acquire link time consumption data corresponding to each work order based on a preset work order state database; The abnormal analysis module is configured to perform abnormal analysis on the link time consumption data corresponding to each work order to obtain corresponding abnormal link time consumption data; The potential problem analysis module is configured to perform potential problem analysis on the abnormal link time consumption data based on a preset digital employee intelligent auxiliary task allocation engine tool to obtain a corresponding potential problem list; The screening module is configured to perform abnormal problem screening on each problem record in the potential problem list based on a rule classification algorithm to obtain a corresponding abnormal problem set; The risk assessment module is configured to perform risk assessment on the abnormal problem set to obtain a corresponding work order monitoring result.

8. An electronic device, comprising: The computer program is executed to implement the digital employee intelligent auxiliary task allocation engine work order monitoring method according to any one of claims 1-6.

9. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed to implement the digital employee intelligent auxiliary task allocation engine work order monitoring method according to any one of claims 1-6.

10. A computer program product, characterised in that, The computer program product comprises a computer program stored on a non-transitory computer-readable storage medium, and the computer program comprises program instructions, wherein when the program instructions are executed by a computer, the computer executes the digital employee intelligent auxiliary task allocation engine work order monitoring method according to any one of claims 1-6.

Images