A big data-based work order processing method

Abstract

The present application relates to the field of work order processing, and more particularly to a work order processing method based on big data, which collects the historical interaction records of the proposed end work order, analyzes the processing content features, calculates the processing priority representation value combined with the feedback features, and sorts the priority of each proposed end; The historical processing response features of the processing node are called, the processing response degree representation parameters are evaluated, the matching results are obtained by sorting the priority of the proposed end; When the proposed end submits a work order, the processing node is allocated according to the matching result; Analyze the processing efficiency representation parameters of the same type of work order of the processing node, and determine whether to switch the processing node; When switching is required, select the appropriate transfer processing node according to the remaining response time of the work order, and determine the final transfer processing node combined with the transfer frequency and the number of transfer types. The present application ensures the processing efficiency of the work order, reduces the processing delay, improves the accuracy and rationality of the work order processing, and realizes the intelligentization and automation of the work order processing.

Description

Technical Field

[0001] This invention relates to the field of work order processing, and more particularly to a work order processing method based on big data. Background Technology

[0002] In today's rapidly developing digital business environment, work order processing systems play a crucial role in various industries and are widely used in customer service, operation and maintenance management, government hotlines and many other fields, aiming to efficiently handle various task requests and problem feedback.

[0003] The rapid development of technologies such as artificial intelligence, big data, and cloud computing has provided solid technical support for the intelligent development of work order processing. Big data technology can collect, store, and analyze massive amounts of work order data, providing a basis for accurate decision-making. AI technology enables intelligent classification, prioritization, and automatic allocation of work orders. Deep learning allows the system to autonomously optimize work order processing workflows and predict business trends. These technologies enable intelligent and automated work order processing.

[0004] Chinese Patent Application Publication No. CN118798537A discloses a method for intelligently allocating work orders based on real-time big data calculation. The method includes the following steps: S1, acquiring all work order information entered into the work order system after a system failure, creating task work orders, transmitting the task work orders to the WOID algorithm engine, and calculating a list of engineers matching the task work orders; S2, automatically assigning the task work orders to the corresponding engineers for processing and generating a work order completion report. By calculating the engineer's pending tasks, the user's VIP value, and the work order's category, the method calculates a list of engineers that can match the work orders in real time. Intelligent allocation is then performed based on the number of pending work orders for each engineer, thereby improving both the efficiency and accuracy of work order allocation, while reducing manual intervention and achieving a uniform distribution of work orders to all engineers.

[0005] However, the following problems still exist in the existing technology.

[0006] Work order allocation relies on human experience, which can easily lead to situations such as "the capable doing more work" and "newcomers handling complex orders." This results in long processing cycles, wasted resources, and repeated rejections or reassignments of work orders at some processing nodes due to mismatched capabilities. This creates process redundancy, significantly reduces work order processing efficiency, and leads to a poor user experience. Summary of the Invention

[0007] To address these issues, this invention provides a big data-based work order processing method to overcome the problems in existing technologies, such as work order allocation relying on manual experience, long processing cycles, resource waste, and process redundancy caused by mismatched capabilities of some processing nodes, which greatly reduces work order processing efficiency and results in a poor user experience.

[0008] To achieve the above objectives, the present invention provides a work order processing method based on big data, comprising:

[0009] The historical interaction records of the requesting client based on the requested work order are collected to analyze the processing content characteristics of the requesting client within a predetermined time period;

[0010] Combining the processing content characteristics and the feedback characteristics of the requesting end, calculate the processing priority characterization value of the requesting end, and sort the processing priorities of each requesting end.

[0011] The historical processing response characteristics of several processing nodes on the processing end are invoked to evaluate the processing response degree characterization parameters of the processing nodes, and then matched with the sorting to obtain the matching results.

[0012] In response to a work order submitted by the requesting client, the processing node is allocated based on the matching results;

[0013] Determine the processing feedback characteristics of the processing node for the same type of work order corresponding to the assigned work order, analyze the processing efficiency characterization parameters of the processing node for the same type of work order, and determine whether to switch the processing node for the work order.

[0014] Obtain the remaining processing nodes at the current moment, select the transfer processing node based on the remaining response time of the work order, and identify the corresponding work order transfer frequency and number of transfer types based on the work order processing records of the transfer processing node to determine the transfer processing node corresponding to the work order.

[0015] The processing content features include the proportion of work orders processed for upgrades and the frequency of unaccepted smart push notifications; the historical processing response features include the number of reminders received and the processing response speed; and the processing feedback features include the average time until the remaining validity period expires and the rate at which processing time decreases.

[0016] Furthermore, the process of calculating the processing priority representation value of the requesting end includes:

[0017] The sum of the ratio of the number of upgraded work orders to the work order number ratio threshold and the ratio of the frequency of unadopted smart pushes to the unadopted frequency threshold is used as the first processing priority feature.

[0018] The ratio of the feedback feature threshold to the feedback feature of the requesting end is used as the second priority feature for processing.

[0019] The first processing priority feature and the second processing priority feature are weighted and summed to determine the processing priority characterization value.

[0020] Furthermore, the processing priority of each requesting end is sorted, including:

[0021] The requesting endpoints are sorted in descending order based on their processing priority values.

[0022] Furthermore, the process of evaluating the processing response characteristics of the processing node includes:

[0023] The ratio of the number of times a person is urged to the threshold number of times a person is urged is used as the first processing response feature;

[0024] The ratio of the processing response speed to the processing response speed threshold is used as the second processing response feature;

[0025] The sum of the first processing response feature and the second processing response feature is used as the parameter representing the degree of processing response.

[0026] Further, the process of matching with the sorted order to obtain the matching result includes:

[0027] The processing nodes are sorted in ascending order based on the processing response level characterization parameters of each node.

[0028] The request nodes for descending sorting are assigned one-to-one with the processing nodes for ascending sorting to obtain the matching results.

[0029] Furthermore, the process of analyzing the processing efficiency characterization parameters of the processing node for the same type of work order includes:

[0030] The ratio of the average remaining validity period to the average remaining validity period threshold is used as the first processing efficiency feature.

[0031] The ratio of the rate of reduction in processing time to the reduction rate threshold is used as the second processing efficiency feature;

[0032] The sum of the first processing efficiency feature and the second processing efficiency feature is used as the processing efficiency characterization parameter.

[0033] Further, determining whether to switch the processing node for the work order includes:

[0034] If the processing efficiency parameter of a processing node for the same type of work order is less than the threshold of the processing efficiency parameter, then it is determined that the processing node should be switched for the work order.

[0035] Furthermore, the process of selecting a transfer processing node based on the remaining response time of the work order includes:

[0036] Pre-set the correspondence between the available working states of the processing nodes and the remaining response time intervals;

[0037] Determine the remaining response time interval to which the work order belongs, in order to determine the corresponding available working status;

[0038] Select the processing node that satisfies the available working state from the remaining processing nodes as the transfer processing node;

[0039] The remaining response time interval is preset, and each working state can be matched with a one-to-one remaining response time interval.

[0040] Further, determining the transfer processing node corresponding to the work order includes:

[0041] If the work order transfer frequency of any transfer processing node is greater than the work order transfer frequency threshold and the number of transfer types is greater than the transfer type number threshold, then the transfer processing node is determined as the transfer processing node corresponding to the work order.

[0042] Furthermore, the selectable working states include task processing state and basic availability state.

[0043] Compared with existing technologies, this invention collects historical interaction records from the requesting end based on the submitted work order to analyze the processing content characteristics of the requesting end within a predetermined time; combines the feedback characteristics of the requesting end to calculate the processing priority characterization value of the requesting end, and sorts the processing priorities of each requesting end; calls the historical processing response characteristics of several processing nodes of the processing end to evaluate the processing response degree characterization parameters of the processing nodes, matches them with the ranking, and obtains the matching results; responds to the work order submitted by the requesting end, and allocates processing nodes based on the matching results; determines the processing feedback characteristics of the processing nodes for the same type of work order, analyzes the processing efficiency characterization parameters of the processing nodes for the same type of work order, and determines whether to switch processing nodes for the work order; obtains the remaining processing nodes at the current time, selects the transfer processing node based on the remaining response time of the work order, and identifies the corresponding work order transfer frequency and the number of transfer types based on the work order processing records of the transfer processing node to determine the transfer processing node corresponding to the work order. This invention ensures efficient work order processing, reduces processing delays, improves the accuracy and rationality of work order processing, optimizes user experience, increases satisfaction, and ultimately achieves intelligent and automated work order processing.

[0044] In particular, this invention considers the requester's historical interactions with the work order process. Typically, work orders with high urgency and severity, or those based on explicit user requests or strong feedback, are escalated. The proportion of escalated work orders reflects the complexity or urgency of the requesting work order. Furthermore, the processing end intelligently pushes solutions based on the work order content. The acceptance rate of these intelligent pushes by the requesting end analyzes the suitability of the automatically pushed solutions for the work order. By incorporating feedback characteristics, the invention directly links to the requesting end's service experience, ensuring that requesting ends with poor historical experiences receive higher priority processing, helping to quickly resolve their legacy issues and improve their service impression. Simultaneously, by combining objective data on processing content characteristics, the invention avoids subjective arbitrariness in priority determination, allowing requesting ends to perceive fairness in the processing and thus improving overall satisfaction. By combining these two types of features to calculate priority values, the sorting results are made more closely aligned with actual needs, ensuring the rationality of work order processing and the timeliness of processing priority adjustments. After sorting in this way, higher-priority requesters can obtain processing resources first, improving the accuracy of resource allocation. This invention, while ensuring work order processing efficiency and reducing processing delays, improves the accuracy and rationality of work order processing, optimizes user experience, increases satisfaction, and ultimately achieves intelligent and automated work order processing.

[0045] In particular, this invention considers both the satisfaction of requesting clients with their work order processing needs and the processing response of processing nodes. This includes quantifying the proactive response capability and processing efficiency of processing nodes through the number of reminders and processing response speed, respectively. These two characteristics, used to evaluate the degree of processing response, objectively reflect the comprehensive service capability of processing nodes. Furthermore, by sorting several processing nodes in ascending order and matching them one-to-one with requesting clients in descending order, it ensures that high-priority requesting clients are matched with processing nodes that have stronger response capabilities and higher efficiency. This allows high-quality resources to prioritize serving more critical needs, avoiding inefficient processing caused by resource mismatch and achieving precise matching of processing resources. Moreover, high-priority requesting clients typically have higher requirements for processing timeliness. Among the ascending-ordered processing nodes, those at the top are reminded less and have faster processing response speeds, enabling them to respond to high-priority work orders more quickly. Additionally, matching relatively low-priority requesting clients with processing nodes with slightly weaker response capabilities allows processing to be completed within a reasonable range, avoiding waste of high-quality processing node resources and improving the overall efficiency and resource utilization of work order processing. In summary, the "one-to-one" allocation rule adopted in this invention ensures that all requesting clients with different processing priorities can obtain the corresponding level of processing resources. This guarantees that high-priority requests are processed first, while also preventing low-priority requests from being ignored, making the entire processing process more stable and fair.

[0046] In particular, based on the rational allocation of requesting terminals and processing nodes, this invention provides a refined analysis of the compatibility between work orders requested by the requesting terminals and processing nodes. By tracking and analyzing the processing efficiency of similar work orders, work orders can be transferred from inefficient processing nodes to more suitable ones, achieving rational and precise dynamic resource adjustment. Therefore, this invention quantifies the processing node's ability to complete similar work orders within their validity period by using the average remaining validity period until expiration; while the rate of reduction in processing time reflects the trend of improved processing efficiency of processing nodes for similar work orders. By calculating processing efficiency parameters using these two characteristics, the actual processing capacity of processing nodes for the current work order can be dynamically evaluated, avoiding the limitations of judging solely based on historical static data. Furthermore, continuing to process work orders using the currently assigned processing nodes may lead to processing timeouts or inefficient processing of that type of work order, potentially reducing customer satisfaction. Switching processing nodes promptly mitigates these risks, preventing delays and overdue orders due to insufficient capacity of the current node, ensuring proper processing within the validity period, and reducing disputes or losses caused by overdue processing. Moreover, this invention, through an objective data-driven node switching mechanism, ensures that work order processing consistently moves towards efficiency and reliability, reducing service fluctuations caused by bottlenecks or insufficient capacity of individual processing nodes. For the customer, dynamic optimization of the work order processing leads to more stable processing results, thereby enhancing customer trust and satisfaction.

[0047] In particular, regarding the switching of processing nodes, a suitable processing node is selected to process the work order based on the working status of other processing nodes and the transfer status of the work order to be replaced. By pre-setting the correspondence between the remaining response time range and the available working status, the subjective arbitrariness of manual intervention is reduced, making the process more standardized and controllable. At the same time, based on the urgency of the remaining response time of the work order, other processing nodes in the corresponding working status are accurately matched. Furthermore, working statuses such as "task processing status" and "basic availability status" are distinguished to adapt to the processing needs of different work orders. Within the framework of the standard, the system possesses the flexibility to handle diverse scenarios, avoiding resource waste and allocating processing nodes in a targeted manner. It provides a safety net for work order processing from a time perspective, reducing the risk of timeouts due to mismatches between processing node response capabilities and time requirements, thus ensuring the timeliness of work order processing. Furthermore, the final selection of processing nodes is based on their work order transfer frequency and the number of transfer types. This ensures that the selected processing nodes are suitable. A high work order transfer frequency indicates that the processing node has extensive experience in handling similar work orders, reducing the error rate. A large number of transfer types reflects the processing node's comprehensive processing capabilities, enabling it to handle potentially complex work order situations. This dual selection avoids blindly allocating idle nodes and ensures efficient use of idle resources, improving the adaptability and utilization rate of processing node resources. In addition, processing nodes selected through the above mechanism can process work orders more efficiently and accurately within the specified time, avoiding processing delays or quality issues caused by unsuitable processing nodes, improving work order processing efficiency and client satisfaction with the work order processing results. Attached Figure Description

[0048] Figure 1 This is a schematic diagram illustrating the steps of a work order processing method based on big data, as described in an embodiment of the invention.

[0049] Figure 2 This is a logic diagram for determining whether to switch processing nodes for a work order, as shown in the embodiments of the invention.

[0050] Figure 3 A logic decision diagram for selecting a transfer processing node in an embodiment of the invention;

[0051] Figure 4 A logic decision diagram for determining the transfer processing node corresponding to a work order in an embodiment of the invention. Detailed Implementation

[0052] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0053] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0054] Please see Figure 1 The diagram illustrates the steps of a big data-based work order processing method according to an embodiment of the present invention. The big data-based work order processing method according to an embodiment of the present invention includes:

[0055] Step S1: Collect the historical interaction records of the requesting terminal based on the requested work order, so as to analyze the processing content characteristics of the requesting terminal within a predetermined time.

[0056] Step S2: Combining the processing content features and the feedback features of the requesting end, calculate the processing priority characterization value of the requesting end, and sort the processing priorities of each requesting end.

[0057] Step S3: Call the historical processing response characteristics of several processing nodes on the processing end, evaluate the processing response degree characterization parameters of the processing nodes, match them with the sorting, and obtain the matching results;

[0058] Step S4: In response to the work order submitted by the requesting end, allocate processing nodes based on the matching results;

[0059] Step S5: Determine the processing feedback characteristics of the processing node for the same type of work order corresponding to the assigned work order, and analyze the processing efficiency characterization parameters of the processing node for the same type of work order to determine whether to switch the processing node for the work order.

[0060] Step S6: Obtain the remaining processing nodes at the current time, select the transfer processing node based on the remaining response time of the work order, and identify the corresponding work order transfer frequency and number of transfer types based on the work order processing records of the transfer processing node to determine the transfer processing node corresponding to the work order.

[0061] The processing content features include the proportion of work orders processed for upgrades and the frequency of unaccepted smart push notifications; the historical processing response features include the number of reminders received and the processing response speed; and the processing feedback features include the average time until the remaining validity period expires and the rate at which processing time decreases.

[0062] Specifically, the processing node plays a role in the entire chain from "initiation" to "resolution" of a work order, and is the processing recipient assigned by the work order. It has the authority and ability to process work orders and complete the response, processing or closed-loop operation of the work order within the prescribed process. This will not be elaborated further.

[0063] Specifically, in order to ensure that the behavioral characteristics of the requesting end during the work order processing interaction, such as communication style and satisfaction tendency, are stable and representative, the predetermined time is set to 3 months.

[0064] Specifically, the time interval from when a processing node “receives a work order” to when it “responds to the work order for the first time” is used as the processing response speed.

[0065] Specifically, the historical interaction records include processing content characteristics, feedback characteristics from the requesting end, historical processing response characteristics, processing feedback characteristics, work order processing records, etc., wherein the work order processing records include work order transfer frequency and the number of transfer types.

[0066] Understandably, given the client's desire for immediate response, they often expect quick feedback or a preliminary solution after submitting a ticket. If the processing node remains unresponsive for an extended period, it can easily lead to dissatisfaction and even escalation of the problem. For example, in customer service scenarios, if a user submits a ticket and a human agent doesn't connect promptly, intelligent push notifications of common problem solutions can provide initial support and reduce user anxiety. Furthermore, setting up intelligent solution push notifications is also necessary for system resource optimization. Temporary resource constraints may prevent processing nodes from responding promptly. Pushing matching solutions intelligently can both offload some simpler tickets (such as standardized issues) to reduce the pressure on subsequent processing nodes and provide preliminary support to complex tickets while they await a response, improving overall process efficiency. Therefore, the adoption rate of intelligent push notifications by the client should be considered during implementation.

[0067] In this context, if the requesting client receives the smart push and does not submit any new supplementary information or initiate a manual intervention request within a predetermined time period, it is considered as acceptance of the smart push; otherwise, it is considered as non-acceptance of the smart push. The predetermined time period is set to 5 minutes.

[0068] Specifically, after a work order is processed, the system automatically triggers a satisfaction evaluation process. While pushing the processing result to the requesting client, it simultaneously sends an evaluation request for processing satisfaction. Then, the evaluation data is quantified, and the star rating can be converted into a corresponding numerical value (e.g., 5 stars corresponds to 100 points, 4 stars corresponds to 80 points, and so on). The positive or negative keywords mentioned in the unstructured feedback are combined for weighted calibration, and finally, a satisfaction value for the requesting client regarding the processing of this work order is generated. The satisfaction value is used as the feedback feature.

[0069] The evaluation request includes structured rating options (such as star rating) and unstructured feedback entry points. The requester can submit evaluations of the processing process (such as response speed, solution effectiveness, service attitude, etc.) by clicking on the rating or entering text.

[0070] Specifically, natural language processing (NLP) technology can be used to identify urging keywords / phrases in the content information sent by the requesting party. Combined with deep learning models, such as intent recognition models, the potential intent of the content information can be analyzed to determine whether it implicitly urges the processing progress, and then the number of times it has been urged can be determined.

[0071] Specifically, the process of calculating the processing priority representation value of the requesting end includes:

[0072] The sum of the ratio of the number of upgraded work orders to the work order number ratio threshold and the ratio of the frequency of unadopted smart pushes to the unadopted frequency threshold is used as the first processing priority feature.

[0073] The ratio of the feedback feature threshold to the feedback feature of the requesting end is used as the second priority feature for processing.

[0074] The first processing priority feature and the second processing priority feature are weighted and summed to determine the processing priority characterization value.

[0075] Specifically, the work order in which the requesting end submits a request for processing upgrade is referred to as the work order for upgrade processing. Accordingly, natural language processing (NLP) technology and deep learning models can be used to identify whether the requesting end has expressed a request for processing upgrade, which will not be elaborated further here.

[0076] Specifically, the core objective of work order processing is to efficiently resolve the actual problems of the requesting party, especially urgent and complex needs. The first priority feature directly addresses the "problem-solving requirement itself," while the second priority feature addresses the "need for improvement of past service experience." Giving the first priority feature higher weight aligns with the logic of "ensuring core issues are addressed promptly first, then improving long-term satisfaction through experience optimization." This maximizes processing efficiency with limited resources, avoiding neglecting the urgency of actual problems due to excessive focus on user feedback. The proportion of work orders requiring escalation processing directly correlates with the complexity or urgency of the requesting party's work orders, while the frequency of unaccepted intelligent push notifications reflects the requesting party's adaptability to the system's automatic processing. Both of these features are based on objective data from historical interactions, directly reflecting the urgency of the current work order's actual processing needs. They are the core basis for determining priority. Giving higher weight to the first priority feature, calculated based on processing content characteristics—namely, the proportion of work orders requiring escalation processing and the frequency of unaccepted intelligent push notifications—ensures that these objective and critical needs dominate priority determination, avoiding misjudgments of urgent needs due to subjective factors (such as fluctuations in feedback characteristics).

[0077] Feedback characteristics from the requesting client primarily reflect their subjective evaluation of past services and are an important reference for optimizing the service experience. However, they have certain limitations. For example, feedback characteristics may be affected by a single, accidental event or have no direct correlation with the urgency of the current work order. Assigning too high a weight to these characteristics could lead to requests with "low satisfaction but non-urgent current needs" crowding out resources for requests with "high urgency but historically high satisfaction," resulting in resource misallocation. Therefore, assigning a lower weight to the second processing priority characteristic calculated based on the client's feedback characteristics allows for prioritizing the processing of objectively urgent needs while still considering user experience.

[0078] Therefore, in the weighted summation, the weight of the first processing priority feature is set to 0.6, and the weight of the second processing priority feature is set to 0.4;

[0079] In this embodiment, the purpose of setting the work order quantity ratio threshold, the non-adoption frequency threshold, and the feedback characteristic threshold is to characterize the overall complexity or urgency of the work orders submitted by the requesting party. By obtaining the historical interaction records of several work orders submitted by the requesting party, and calling the historical data of the work order quantity ratio for escalation processing, the historical data of the non-adoption frequency of intelligent push, and the historical data of the feedback characteristics of the requesting party, the average work order quantity ratio, the average non-adoption frequency, and the average feedback characteristic are calculated and used as the baseline values ​​under normal circumstances. Based on the purpose of setting the above three thresholds, the... The threshold for the proportion of work orders is determined as the product of the average proportion of work orders and the proportion deviation coefficient. The threshold for the frequency of non-adoption is determined as the product of the average frequency of non-adoption and the frequency deviation coefficient. The threshold for the feedback feature is determined as the product of the average feedback feature and the feedback deviation coefficient. The proportion deviation coefficient is selected within the interval [1.1, 1.15], preferably 1.1 in practice. The frequency deviation coefficient is selected within the interval [1.15, 1.2], preferably 1.15 in practice. The feedback deviation coefficient is selected within the interval [0.9, 0.95], preferably 0.9 in practice.

[0080] Specifically, the processing priority of each requesting client is sorted, including:

[0081] The requesting endpoints are sorted in descending order based on their processing priority values.

[0082] Specifically, this invention considers the historical interactions of the requester regarding the processing of work orders. Generally, for work orders with high urgency and severity, or based on explicit user requests or strong feedback, the work order will be escalated. The proportion of work orders escalated reflects the complexity or urgency of the work orders submitted by the requester. Furthermore, the processing end will intelligently push solutions based on the content of the work order in advance. By analyzing the acceptance rate of the intelligent push by the requester, the adaptability of the automatically pushed solutions to the work order can be determined. If the frequency of non-adoption is high, it may indicate that the intelligently pushed solutions do not match the actual needs, requiring more precise manual intervention or adjustment of the push strategy, which will affect the processing priority of the requester.

[0083] By incorporating feedback characteristics into the evaluation, directly linking them to the service experience of the requesting client, this invention ensures that clients with historically poor experiences receive higher priority processing, facilitating the rapid resolution of their legacy issues and improving their service impression. Simultaneously, by combining objective data on processing content characteristics, it avoids subjective arbitrariness in priority determination, allowing requesting clients to perceive fairness in the processing and thus improving overall satisfaction. Combining these two types of characteristics to calculate a processing priority value makes the ranking results more aligned with actual needs, ensuring the rationality of work order processing and the timeliness of priority adjustments. After ranking in this way, higher-priority requesting clients receive processing resources first. For example, requesting clients with a high proportion of escalation work orders and low feedback characteristics may have issues that require more urgent resolution or more personalized service; prioritizing these clients for resource allocation can promptly improve their experience, and early ranking ensures that such needs receive priority responses, improving the accuracy of resource allocation. This invention, while ensuring work order processing efficiency and reducing processing delays, improves the accuracy and rationality of work order processing, optimizes user experience, increases satisfaction, and ultimately achieves intelligent and automated work order processing.

[0084] Specifically, the process of evaluating the processing response characteristics of the processing node includes:

[0085] The ratio of the number of times a person is urged to the threshold number of times a person is urged is used as the first processing response feature;

[0086] The ratio of the processing response speed to the processing response speed threshold is used as the second processing response feature;

[0087] The sum of the first processing response feature and the second processing response feature is used as the parameter representing the degree of processing response.

[0088] In this embodiment, the purpose of setting the threshold for the number of reminders and the threshold for the processing response speed is to characterize the situation where the processing node's processing response is relatively slow and the service efficiency is low. By obtaining the historical interaction records of several work orders submitted by the requesting end, calling the historical data of the number of reminders and the historical data of the processing response speed, the average number of reminders and the average processing response speed are calculated and used as the benchmark values ​​under normal circumstances. Based on the purpose of setting the above two thresholds, the threshold for the number of reminders is determined as the product of the average number of reminders and the number deviation coefficient, and the threshold for the processing response speed is determined as the product of the average processing response speed and the response deviation coefficient. The number deviation coefficient is selected in the interval [1.2, 1.25], preferably 1.2 in the implementation, and the response deviation coefficient is selected in the interval [1.15, 1.2], preferably 1.15 in the implementation.

[0089] Specifically, the process of matching with the sorted order to obtain the matching result includes:

[0090] The processing nodes are sorted in ascending order based on the processing response level characterization parameters of each node.

[0091] The request nodes for descending sorting are assigned one-to-one with the processing nodes for ascending sorting to obtain the matching results.

[0092] Specifically, this invention considers both the satisfaction of the requesting end with the work order processing needs and the processing response of the processing nodes. This includes quantifying the proactive response capability and processing efficiency of the processing nodes by the number of reminders and the processing response speed. For example, fewer reminders indicate that the processing node is proactive, while a fast processing response speed indicates that the processing node has high response efficiency. The processing response level characterization parameters evaluated by these two features can objectively reflect the comprehensive service capability of the processing nodes.

[0093] Furthermore, by sorting several processing nodes in ascending order and matching them one-to-one with requesting ends in descending order, we can ensure that high-priority requesting ends, such as those with more urgent, complex, or low-feedback characteristics, are matched with processing nodes that have stronger response capabilities and higher efficiency. This allows high-quality resources to prioritize serving more critical needs, avoiding inefficient processing caused by resource mismatch and achieving precise matching of processing resources. In addition, high-priority requesting ends usually have higher requirements for processing timeliness. Among the processing nodes in ascending order, the earlier processing nodes are urged less and have a faster processing response speed, enabling them to respond to high-priority work orders more quickly. Furthermore, matching relatively low-priority requesting ends with processing nodes with slightly weaker response capabilities can also complete the processing within a reasonable range, avoiding the waste of high-quality processing node resources and improving the overall efficiency of work order processing and resource utilization. In summary, the "one-to-one" allocation rule adopted in this invention ensures that all requesting clients with different processing priorities can obtain the corresponding level of processing resources. This guarantees that high-priority requests are processed first, while also preventing low-priority requests from being ignored, making the entire processing process more stable and fair.

[0094] Specifically, the process of analyzing the processing efficiency characteristics of the processing node for the same type of work order includes:

[0095] The ratio of the average remaining validity period to the average remaining validity period threshold is used as the first processing efficiency feature.

[0096] The ratio of the rate of reduction in processing time to the reduction rate threshold is used as the second processing efficiency feature;

[0097] The sum of the first processing efficiency feature and the second processing efficiency feature is used as the processing efficiency characterization parameter.

[0098] In this embodiment, the purpose of setting the threshold for the average remaining validity period and the threshold for the shortening rate is to characterize the situation where the initial allocation of processing nodes and work orders has a low degree of matching. By obtaining the historical interaction records of several work orders submitted by the requesting end, the historical data of the average remaining validity period and the historical data of the shortening rate of the processing time of several similar work orders corresponding to the initial allocation of processing nodes are called to calculate the average value of the average remaining validity period and the average value of the shortening rate, and these are used as the benchmark quantities under normal circumstances. Based on the purpose of setting the above two thresholds, the threshold for the average remaining validity period is determined as the product of the average value of the average remaining validity period and the remaining deviation coefficient, and the threshold for the shortening rate is determined as the product of the average shortening rate and the shortening deviation coefficient. The remaining deviation coefficient is selected in the interval [1.15, 1.2], preferably 1.15 in the implementation, and the shortening deviation coefficient is selected in the interval [1.1, 1.15], preferably 1.1 in the implementation.

[0099] Specifically, please refer to Figure 2 As shown, this is a logic diagram for determining whether to switch processing nodes for a work order according to an embodiment of the present invention. Determining whether to switch processing nodes for the work order includes:

[0100] If the processing efficiency parameter of a processing node for the same type of work order is less than the threshold of the processing efficiency parameter, then it is determined to switch the processing node for the work order.

[0101] If the processing efficiency parameter of a processing node for the same type of work order is greater than or equal to the threshold of the processing efficiency parameter, it is determined that there is no need to switch processing nodes for the work order.

[0102] The threshold value for the processing efficiency characterization parameter is predetermined. The processing efficiency characterization parameter is determined by calculating the average remaining validity period until the end of the validity period is equal to the threshold value for the average remaining validity period until the end of the validity period, and by calculating the shortening rate of the processing time as equal to the threshold value for the shortening rate.

[0103] Specifically, this invention, based on the rational allocation of requesting terminals and processing nodes, meticulously analyzes the compatibility between work orders requested by the requesting terminals and processing nodes. By tracking and analyzing the processing efficiency of similar work orders, work orders can be transferred from inefficient processing nodes to more suitable ones, achieving rational and precise dynamic resource adjustment. For example, for a certain type of work order, processing nodes with a high rate of reduction in processing time can complete the task faster. Switching work orders to such nodes can accelerate the overall processing process, while forcing inefficient processing nodes to improve or focus on work order types they are better at, thereby improving the overall processing efficiency and resource utilization of the system. Therefore, this invention quantifies the ability of processing nodes to complete similar work orders within the validity period by using the average remaining validity period deadline. For example, the longer the remaining validity period deadline, the more buffer space the processing node can leave before the deadline, avoiding timeouts. The rate of reduction in processing time reflects the trend of improved processing efficiency of processing nodes for similar work orders. For example, the higher the rate of reduction in processing time, the stronger the processing node's proficiency and optimization ability for this type of work order. By calculating processing efficiency parameters using these two features, the actual processing capacity of processing nodes for the current work order can be dynamically evaluated, avoiding the limitations of relying solely on historical static data. Furthermore, continuing to rely on the currently assigned processing node may lead to processing timeouts or inefficiencies in that node for that type of work order, potentially reducing client satisfaction. Switching processing nodes promptly mitigates these risks, preventing delays and overdue orders due to insufficient capacity of the current node, ensuring proper processing within the validity period, and reducing disputes or losses caused by timeouts. Moreover, this invention, through an objective data-driven node switching mechanism, ensures that work order processing consistently moves towards efficiency and reliability, reducing service fluctuations caused by bottlenecks or insufficient capacity of individual processing nodes. For the client, dynamic optimization of the work order processing leads to more stable results, thereby enhancing client trust and satisfaction.

[0104] Specifically, please refer to Figure 3 As shown, this is a logic decision diagram for selecting a transfer processing node in an embodiment of the present invention. The process of selecting a transfer processing node based on the remaining response time of the work order includes:

[0105] Pre-set the correspondence between the available working states of the processing nodes and the remaining response time intervals;

[0106] Determine the remaining response time interval to which the work order belongs, in order to determine the corresponding available working status;

[0107] Select the processing node that satisfies the available working state from the remaining processing nodes as the transfer processing node;

[0108] The remaining response time interval is preset, and each working state can be matched with a one-to-one remaining response time interval.

[0109] In this embodiment, the correspondence between the available working states and the remaining response time intervals is determined in the following manner:

[0110] The remaining response time is divided into two preset intervals, combined with the two categories of available working states, including task processing state and basic availability state.

[0111] If the remaining response time of the work order is within the first preset remaining response time range (30min, 1h), then the corresponding task processing status is as follows:

[0112] If the remaining response time of the work order is within the second preset remaining response time range (0, 30 min), then the corresponding basic availability status is achieved.

[0113] The response time of the processing node to the work order is set to 1 hour.

[0114] Specifically, please refer to Figure 4 As shown, this is a logic decision diagram for determining the transfer processing node corresponding to a work order in an embodiment of the present invention. Determining the transfer processing node corresponding to the work order includes:

[0115] If the work order transfer frequency of any transfer processing node is greater than the work order transfer frequency threshold and the number of transfer types is greater than the transfer type number threshold, then the transfer processing node is determined as the transfer processing node corresponding to the work order.

[0116] In this embodiment, the purpose of setting the work order transfer frequency threshold and the transfer type quantity threshold is to characterize the situation where any other processing node has high processing efficiency and rich processing experience for the work order to be transferred. By obtaining the historical interaction records of several work orders submitted by the requesting end, the historical data of the work order transfer frequency and the corresponding historical data of the transfer type quantity of the transfer processing node are called to solve the average work order transfer frequency and the average transfer type quantity, which are used as the benchmark values ​​under normal circumstances. Based on the purpose of setting the above two thresholds, the work order transfer frequency threshold is determined as the product of the average work order transfer frequency and the transfer deviation coefficient, and the transfer type quantity threshold is determined as the product of the average transfer type quantity and the type deviation coefficient. The transfer deviation coefficient is selected in the interval [1.2, 1.25], preferably 1.2 in the implementation, and the type deviation coefficient is selected in the interval [1.25, 1.3], preferably 1.25 in the implementation.

[0117] Specifically, the selectable working states include task processing state and basic availability state.

[0118] Specifically, the task processing status refers to the state in which the processing node is processing work orders at a relatively high frequency; the basic availability status refers to the state in which the processing node has no processing tasks or a low task load. For example, idle status: there are currently no assigned processing tasks, and it can immediately accept new work orders; standby status: although there are no specific processing tasks, it is in a state of being ready to respond at any time (it may be in the process of waiting for the requesting end to supplement information); light load status: it is processing a small number of work orders, but the processing capacity is sufficient and will not affect the processing progress of the currently processed work orders.

[0119] Each processing node is configured with a corresponding working status label. The processing node can choose to set the working status label according to its own processing status, thereby determining the working status of the processing node.

[0120] Specifically, regarding the switching of processing nodes, a suitable processing node is selected to process the work order based on the working status of other processing nodes and the transfer status of the work order to be replaced. By pre-setting the correspondence between the remaining response time range and the available working status, the subjective arbitrariness of manual intervention is reduced, making the process more standardized and controllable. At the same time, other processing nodes in the corresponding working status are accurately matched according to the urgency of the remaining response time of the work order. Furthermore, working statuses such as "task processing status" and "basic availability status" are distinguished to adapt to the processing needs of different work orders. For example, work orders with extremely short remaining response time are assigned to processing nodes in a state with higher processing efficiency and stronger focus, ensuring that they can be completed before the deadline for work order processing.

[0121] Within the framework of the standard, the system possesses the flexibility to handle diverse scenarios, avoiding resource waste and allocating processing nodes in a targeted manner. It provides a safety net for work order processing from a time perspective, reducing the risk of timeouts due to mismatches between processing node response capabilities and time requirements, thus ensuring the timeliness of work order processing. Furthermore, the final selection of processing nodes is based on their work order transfer frequency and the number of transfer types. This ensures that the selected processing nodes are suitable. A high work order transfer frequency indicates that the processing node has extensive experience in handling similar work orders, reducing the error rate. A large number of transfer types reflects the processing node's comprehensive processing capabilities, enabling it to handle potentially complex work order situations. This dual selection avoids blindly allocating idle nodes (such as idle nodes that are not proficient in handling this type of work order) and ensures efficient utilization of idle resources, improving the adaptability and utilization rate of processing node resources. In addition, processing nodes selected through the above mechanism can process work orders more efficiently and accurately within the specified time, avoiding processing delays or quality issues caused by unsuitable processing nodes, improving work order processing efficiency and client satisfaction with the work order processing results.

[0122] If the big data-based work order processing method of the present invention 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, 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. The computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in the various embodiments of the present invention. The aforementioned storage medium includes various media that can store program code, such as USB flash drives, mobile hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0123] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

Claims

1. A work order processing method based on big data, characterized in that, include: The historical interaction records of the requesting client based on the requested work order are collected to analyze the processing content characteristics of the requesting client within a predetermined time period; Combining the processing content characteristics and the feedback characteristics of the requesting end, calculate the processing priority characterization value of the requesting end, and sort the processing priorities of each requesting end. The historical processing response characteristics of several processing nodes on the processing end are invoked to evaluate the processing response degree characterization parameters of the processing nodes, and then matched with the sorting to obtain the matching results. In response to a work order submitted by the requesting client, the processing node is allocated based on the matching results; Determine the processing feedback characteristics of the processing node for the same type of work order corresponding to the assigned work order, analyze the processing efficiency characterization parameters of the processing node for the same type of work order, and determine whether to switch the processing node for the work order. Obtain the remaining processing nodes at the current moment, select the transfer processing node based on the remaining response time of the work order, and identify the corresponding work order transfer frequency and number of transfer types based on the work order processing records of the transfer processing node to determine the transfer processing node corresponding to the work order. The processing content features include the proportion of work orders processed for upgrades and the frequency of smart push notifications not being adopted; the historical processing response features include the number of times reminders were given and the processing response speed; and the processing feedback features include the average time until the remaining validity period expires and the rate at which processing time is shortened. The process of calculating the processing priority value of the requesting end includes: The sum of the ratio of the number of upgraded work orders to the work order number ratio threshold and the ratio of the frequency of unadopted smart pushes to the unadopted frequency threshold is used as the first processing priority feature. The ratio of the feedback feature threshold to the feedback feature of the requesting end is used as the second priority feature for processing. The first processing priority feature and the second processing priority feature are weighted and summed to determine the processing priority characterization value; The process of evaluating the processing response characteristics of the processing node includes: The ratio of the number of times a person is urged to the threshold number of times a person is urged is used as the first processing response feature; The ratio of the processing response speed to the processing response speed threshold is used as the second processing response feature; The sum of the first processing response feature and the second processing response feature is used as the parameter representing the degree of processing response; The process of analyzing the processing efficiency parameters of the processing node for the same type of work order includes: The ratio of the average remaining validity period to the average remaining validity period threshold is used as the first processing efficiency feature. The ratio of the rate of reduction in processing time to the reduction rate threshold is used as the second processing efficiency feature; The sum of the first processing efficiency feature and the second processing efficiency feature is used as the processing efficiency characterization parameter.

2. The work order processing method based on big data according to claim 1, characterized in that, The processing priority of each requesting client is sorted, including: The requesting endpoints are sorted in descending order based on their processing priority values.

3. The work order processing method based on big data according to claim 1, characterized in that, The process of matching with the sorted data and obtaining the matching result includes: The processing nodes are sorted in ascending order based on the processing response level characterization parameters of each node. The request nodes for descending sorting are assigned one-to-one with the processing nodes for ascending sorting to obtain the matching results.

4. The work order processing method based on big data according to claim 1, characterized in that, Determining whether to switch processing nodes for the work order includes: If the processing efficiency parameter of a processing node for the same type of work order is less than the threshold of the processing efficiency parameter, then it is determined that the processing node should be switched for the work order.

5. The work order processing method based on big data according to claim 1, characterized in that, The process of selecting a transfer processing node based on the remaining response time of the work order includes: Pre-set the correspondence between the available working states of the processing nodes and the remaining response time intervals; Determine the remaining response time interval to which the work order belongs, in order to determine the corresponding available working status; Select the processing node that satisfies the available working state from the remaining processing nodes as the transfer processing node; The remaining response time interval is preset, and each working state can be matched with a one-to-one remaining response time interval.

6. The work order processing method based on big data according to claim 5, characterized in that, Determining the transfer processing node corresponding to the work order includes: If the work order transfer frequency of any transfer processing node is greater than the work order transfer frequency threshold and the number of transfer types is greater than the transfer type number threshold, then the transfer processing node is determined as the transfer processing node corresponding to the work order.

7. The work order processing method based on big data according to claim 5, characterized in that, The selectable working states include task processing state and basic availability state.

Images