Work order allocation method, work order allocation device, and electronic device

Abstract

The application relates to the field of task scheduling allocation, and provides a work order allocation method, a work order allocation device and electronic equipment. The method comprises the following steps: determining a first target processing object corresponding to a target work order from a work order mapping table based on the type of the target work order; inputting the target work order into a target matching model in the case that the first target processing object is not determined, so as to obtain a second target processing object output by the target matching model; and determining a third target processing object in a target processing team in the case that the second target processing object is not determined. The work order allocation method provided in the application can combine the actual situation of a target work order to quickly and accurately allocate the target work order by using a multi-level allocation mechanism, and can improve the allocation accuracy and response processing efficiency of the target work order while ensuring that the target work order is quickly allocated.

Description

Technical Field

[0001] This application relates to the field of task scheduling and allocation technology, specifically to a work order allocation method, work order allocation device, and electronic equipment. Background Technology

[0002] Currently, when operators are carrying out their own businesses, front-line staff often encounter many issues when handling various tasks, such as feedback and processing of on-site problems and requests, and scheduling and dispatching of system construction work orders. This requires a suitable work order task scheduling system to complete the scheduling and dispatch of work order tasks.

[0003] Current work order scheduling systems typically employ dispatching mechanisms such as designated dispatchers, centralized platform processing, or dispatching according to fixed rules. Manual dispatching has limitations, heavily relying on the experience of dispatchers and struggling to achieve reasonable allocation when dealing with large or complex orders. Dispatching according to fixed rules lacks flexibility, failing to promptly and flexibly call upon suitable personnel for immediate processing. Under these modes, work order allocation is inflexible, inefficient, and inaccurate, hindering rapid response and processing. Summary of the Invention

[0004] This application provides a work order allocation method, work order allocation device, and electronic device to solve the technical problem of low efficiency and accuracy in work order allocation.

[0005] In a first aspect, embodiments of this application provide a work order allocation method, including:

[0006] Based on the type of the target work order, the first target processing object corresponding to the target work order is determined from the work order mapping table, wherein the work order mapping table is a preset mapping table between work order types and candidate objects;

[0007] If the first target processing object is not determined, the target work order is input into the target matching model to obtain the second target processing object output by the target matching model. The target matching model is trained using historical work orders as samples and the processing objects corresponding to the historical work orders as labels.

[0008] If the second target processing object is not identified, a third target processing object is identified within the target processing team, which is associated with the target work order.

[0009] In one embodiment, inputting the target work order into a target matching model to obtain a second target processing object determined by the target matching model includes:

[0010] The target work order is input into the feature extraction layer of the target matching model to obtain the target word vector matrix output by the feature extraction layer;

[0011] The target word vector matrix is ​​input into the classification layer of the target matching model to obtain the classification label information output by the classification layer;

[0012] The classification label information is input into the matching layer of the target matching model to obtain the second target processing object output by the matching layer.

[0013] In one embodiment, inputting the classification label information into the matching layer of the target matching model to obtain the second target processing object output by the matching layer includes:

[0014] Based on the classification label information and the personnel profiles of the candidate objects, the matching coefficient between the target work order and the candidate objects is determined, wherein the personnel profiles are determined based on the attribute information of the processing objects;

[0015] Based on the matching coefficient, the number of target work orders, and the number of work orders being processed by the candidate objects, the matching degree between the target work orders and the candidate objects is determined.

[0016] Based on the matching degree, the second target processing object is determined from the candidate objects.

[0017] In one embodiment, before determining the third target processing object within the target processing team, the method further includes: determining the target processing team associated with the target work order; determining the third target processing object within the target processing team includes:

[0018] The target work order is input into the target matching model to obtain the third target processing object determined by the target matching model from the target processing team.

[0019] In one embodiment, determining the target processing team associated with the target work order includes:

[0020] Based on the organizational structure affiliation information of the candidate objects and the type of historical work orders corresponding to the processing objects, the candidate objects are divided into multiple candidate teams.

[0021] Based on the type of the target work order, the target processing team is determined from the plurality of candidate teams.

[0022] In one embodiment, after identifying the target processing team, the method further includes:

[0023] If the third target processing object is not identified, the status of the target work order will be changed to an unprocessed work order;

[0024] Receive request information from the fourth target processing object regarding the unprocessed work order;

[0025] The target work order is sent to the fourth target processing object.

[0026] In one embodiment, if the actual processing object is determined to be one of the first target processing object, the second target processing object, the third processing object, and the fourth processing object, the method further includes:

[0027] Send the target work order to the actual processing object;

[0028] If the target work order is completed, obtain the processing data of the target work order;

[0029] Based on the processed data, update the personnel profile of the actual processed object;

[0030] If the amount of data processed is greater than or equal to the target value, the target matching model is updated using the target work order as a sample and the actual processing object after updating the personnel profile as a label.

[0031] Secondly, embodiments of this application provide a work order allocation device, comprising:

[0032] The first processing module is used to determine the first target processing object corresponding to the target work order from the work order mapping table based on the type and quantity of the target work order. The work order mapping table is a preset mapping table between the type of work order and the candidate object.

[0033] The second processing module is used to input the target work order into the target matching model when the first target processing object is not determined, and obtain the second target processing object output by the target matching model. The target matching model is trained using historical work orders as samples and the processing objects corresponding to the historical work orders as labels.

[0034] The third processing module is used to determine a third target processing object from the target processing team if the second target processing object is not determined. The target processing team is associated with the target work order.

[0035] Thirdly, embodiments of this application provide an electronic device, including a processor and a memory storing a computer program, wherein the processor executes the program to implement the steps of the work order allocation method described in the first or second aspect.

[0036] Fourthly, embodiments of this application provide a computer program product, including a computer program that, when executed by a processor, implements the steps of the work order allocation method described in the first or second aspect.

[0037] The work order allocation method, work order allocation device, and electronic device provided in this application embodiment, through a multi-level allocation mechanism, can quickly and accurately allocate target work orders according to the actual situation of the target work order using different allocation mechanisms. While ensuring that the target work order is quickly allocated, it improves the allocation accuracy and response processing efficiency of the target work order, thereby improving operational efficiency and user satisfaction. Attached Figure Description

[0038] To more clearly illustrate the technical solutions in this application 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 some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0039] Figure 1 This is one of the flowcharts illustrating the work order allocation method provided in the embodiments of this application;

[0040] Figure 2 This is a second flowchart illustrating the work order allocation method provided in the embodiments of this application;

[0041] Figure 3 This is a schematic diagram of the work order allocation device provided in the embodiments of this application;

[0042] Figure 4 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0043] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings of the embodiments. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0044] Figure 1 A flowchart illustrating the work order assignment method. (Refer to...) Figure 1 This application provides a work order allocation method. The execution subject of the work order allocation method can be a controller. Of course, in some embodiments, the execution subject can also be a server, etc. The work order allocation method of this application embodiment is described below with the server as the execution subject.

[0045] The work order allocation method in this application embodiment may include steps 110, 120 and 130.

[0046] Step 110: Based on the type of the target work order to be assigned, determine the first target processing object corresponding to the target work order from the work order mapping table.

[0047] Understandably, a target work order is a work order to be assigned. After receiving a target work order, the type of the target work order can be determined first.

[0048] For example, for operators, the work orders to be assigned can be system construction work orders in the course of their own business operations or business work orders for front-line staff to handle various tasks.

[0049] The target work order type can be automatically labeled based on historical processing experience, or manually labeled by staff during the work order creation process.

[0050] For example, when a telecom operator receives a customer's request to cancel an account, it can directly assign a work order through the processing module corresponding to account cancellation. Before assigning the target work order, the type of the target work order can be automatically marked as "account cancellation".

[0051] In some embodiments, the type of the target work order can be automatically obtained, and the first target processing object for processing the target work order can be determined from the candidate objects in the work order mapping table.

[0052] It is understandable that the work order mapping table is a preset mapping table between work order types and candidate objects. The work order mapping table can be manually determined based on the processing status of historical work orders, and has high applicability and practicality.

[0053] The work order mapping table includes candidate objects corresponding to various types of work orders. You can match the candidate objects with the type of the target work order from the candidate objects.

[0054] It is understood that there may be multiple target candidates or only one target candidate, or in some embodiments, the number of target candidates is zero, that is, no target candidate matching the type of the target work order can be determined from the candidates.

[0055] In some embodiments, after identifying target candidates, the processing capacity of the target candidates can be judged based on the number of work orders processed by the target candidates and the expected processing time.

[0056] When multiple target work orders need to be assigned at any given time, their distribution can be optimized based on their quantity and type, combined with a work order mapping table. In this case, the total processing time for all target work orders can be considered, and multiple target work orders can be assigned based on minimizing the total completion time. For example, a single primary target processing object can be assigned multiple target work orders.

[0057] If the processing capacity of the target candidate object can meet the processing requirements of the target work order, the target candidate object can be determined as the first target processing object.

[0058] In this embodiment, determining the first target processing object based on the work order mapping table can greatly shorten the allocation time for target work orders with clear types, improve the allocation accuracy of target work orders, and greatly improve the processing efficiency of target work orders.

[0059] It should be noted that if the type of the target work order cannot be determined, the target candidate cannot be determined based on the work order mapping table, and therefore the first target processing object cannot be determined.

[0060] In other embodiments, when a target candidate is identified, the target candidate may be unable to process the target work order in a timely manner due to reasons such as having a large number of work orders currently being processed, and thus may not be identified as the first target processing object.

[0061] Step 120: If the first target processing object has not been determined, input the target work order into the target matching model to obtain the second target processing object output by the target matching model.

[0062] It is understandable that the target matching model is trained using historical work orders as samples and the processing objects corresponding to the historical work orders as labels.

[0063] The target matching model is a neural network model, such as a convolutional neural network model. Based on the input target work order, the target matching model can identify target candidates that match the type of the target work order from a pool of available options.

[0064] In some embodiments, the target matching model can be pre-trained based on a large number of accumulated historical work orders.

[0065] Based on information from previous historical work orders, the data from those work orders is extracted, cleaned, and stored in the database.

[0066] It is understandable that the basic unit of Chinese semantics is the word, so in order to vectorize the description text of historical work orders, the work order data can be processed by word segmentation.

[0067] In some embodiments, the descriptive text of historical work orders can be segmented using the word2vec Chinese word segmentation technology based on the DL algorithm, and then the pre-stored thesaurus and the relevant content of historical work orders can be converted into corresponding fixed density word vectors according to the defined data dictionary.

[0068] After extracting features from the generated dense word vector information, convolutional neural networks can be used to extract features from the dense word vectors of the text, and then a word vector matrix can be established.

[0069] In some embodiments, the cleaned historical work order data can be divided into training work orders and verification work orders. The training work orders are trained based on the word vector matrix to generate a first-level convolutional neural network layer, i.e., the classification layer, and training data, resulting in a second classification label for each type of work order. Based on the first classification label corresponding to the verification work order, the word vector matrix corresponding to the verification work order is input into the corresponding second-level convolutional neural network layer, i.e., the matching layer, to obtain the optimal weight parameter matrix of the matching network layer, ultimately forming a target matching model.

[0070] In this embodiment, the distance between the word vector matrices corresponding to each pair of training work orders can be calculated based on the K-means algorithm; according to the distance between the word vector matrices of each work order, the K-means algorithm is a clustering algorithm based on Euclidean learning.

[0071] The K-means algorithm can be used to cluster the training work orders corresponding to each word vector matrix. The word vector matrix corresponding to the clustered training work orders is then used as the training data for the first-level convolutional neural network layer to obtain the second-level classification label of the corresponding training work order. At the same time, the label of the second-level classification is used as the label for the second-level convolutional neural network layer.

[0072] Based on the first category label corresponding to the verification work order, the word vector matrix corresponding to the verification work order is input into the corresponding second-level convolutional neural network layer to obtain the optimal weight parameter matrix of the second-level convolutional neural network layer. After a large number of iterations of training based on the training work orders and the processing objects corresponding to the verification work orders, the model with the highest matching accuracy and its weight parameters are saved, and this model is used as the target matching model.

[0073] If the processing capacity of the target candidate object meets the processing requirements of the target work order, the target candidate object can be identified as the second target processing object, which can process the target work order in a timely manner.

[0074] Understandably, if the target matching model cannot determine the target candidate from the alternative objects, it also cannot determine the target candidate as the second target processing object.

[0075] In some embodiments, because there are many work orders being processed by the target candidate object, the target candidate object cannot process the target work orders in a timely manner. That is, the processing capacity of the target candidate object cannot meet the processing requirements of the target work orders, so the target candidate object output by the target matching model cannot be determined as the second target processing object.

[0076] It should be noted that the target matching model is trained based on a large number of historical work orders as samples. The target matching model has higher matching efficiency and higher matching accuracy.

[0077] Step 130: If a second target has not been identified, identify a third target within the target handling team.

[0078] If the target matching model cannot output a target candidate or the output target candidate cannot handle the target work order, that is, if a second target processing object is not determined, it is difficult to determine a target candidate that can handle the target work order, or a small number of target candidates can be determined, and thus a third target processing object cannot be determined.

[0079] In this case, the target work order can be a complex work order with a high processing difficulty; or, the target work order can be a rare type of work order, and the candidate object has not processed this type of target work order before, so the candidate object cannot process the target work order and therefore cannot obtain a third target processing object.

[0080] For target work orders under the above circumstances, a target processing team can be established associated with the target work order. One or more alternative targets in the target processing team can process the target work orders of the above types in a timely manner.

[0081] In other words, alternative targets that specifically handle the above-mentioned types of target work orders within the target processing team can be identified as third target processing targets to achieve timely processing of various types of target work orders.

[0082] In this embodiment, by assigning target work orders to target processing teams associated with the target work orders and identifying third target processing objects, target work orders for which processing objects are difficult to identify can be effectively allocated, enabling timely processing of target work orders and improving the processing efficiency of target work orders.

[0083] The work order allocation method of this application embodiment can allocate different types of target work orders. By prioritizing the allocation of target work orders according to the work order mapping table, the method can quickly and accurately allocate target work orders, thereby ensuring that target work orders can be processed in a timely manner.

[0084] If the allocation fails, the target work order can be allocated again through the target matching model, which can accurately determine the second target processing object and process the target work order in a timely manner.

[0085] For complex or rare types of target work orders, a third target processing object is determined by associating with the target processing team to process the target work order in a timely manner. This enables full coverage of response and processing for all types of work orders, improves the responsiveness and processing capabilities for target work orders, and shortens the processing time.

[0086] According to the work order allocation method of this application embodiment, through a multi-level allocation mechanism, the target work order can be quickly and accurately allocated using different allocation mechanisms based on the actual situation of the target work order. This ensures that the target work order is allocated quickly while improving the allocation accuracy and response processing efficiency, thereby enhancing operational efficiency and user satisfaction.

[0087] In some embodiments, the step of inputting the target work order into the target matching model to obtain the second target processing object determined by the target matching model may include inputting the target work order into the feature extraction layer of the target matching model to obtain the target word vector matrix output by the feature extraction layer.

[0088] It is understandable that in the above steps, after the target work order is input into the target matching model, the feature extraction layer of the target matching model can extract features from the descriptive text of the target work order.

[0089] For example, the feature extraction layer can use word2vec Chinese word segmentation technology based on DL algorithm to segment the description text of the target work order, and then convert the description text related to the type of the target work order into corresponding fixed density word vectors according to the defined data dictionary. Finally, feature extraction is performed on the generated dense word vectors to obtain the target word vector matrix.

[0090] After determining the word vector matrix of the target work order, the target word vector matrix is ​​input into the classification layer of the target matching model to obtain the classification label information output by the classification layer.

[0091] The classification label information is input into the matching layer of the target matching model to obtain the second target processing object output by the matching layer.

[0092] In this embodiment, the target matching model can automatically identify the type of target work order, and then quickly match a suitable second target processing object according to the type of target work order, thereby improving the allocation efficiency and accuracy of target work orders.

[0093] In some embodiments, the classification label information is input into the matching layer of the target matching model to obtain the second target processing object output by the matching layer, including: determining the matching coefficient between the target work order and the candidate object based on the classification label information and the personnel profile of the candidate object.

[0094] It is understandable that personnel profiles are determined based on the attribute information of candidate candidates, which includes at least personality information, behavioral information, regional information, skill information, and experience information.

[0095] In other words, we can analyze data such as personality, gender, region, age, work order processing statistics, performance rating, working hours, habit level, and professional experience of each candidate, and then use deep learning personality analysis technology, behavioral analysis technology, and big data analysis technology to form a profile of each candidate.

[0096] Before calculating the matching coefficient, the mean value of each attribute information can be calculated by using the truncated mean calculation method. Then, the cumulative weight coefficient of each candidate object can be calculated to establish an array of candidate object attribute weight coefficients. This allows for the calculation of the summary data of the weight coefficients of various candidate object attribute coefficients and the summary data of the weight coefficients of various candidate object work order attribute coefficients, thus obtaining the matching coefficient. Finally, an array of candidate objects and matching coefficients can be constructed.

[0097] The sum of the weights of various attribute coefficients of the candidate objects is denoted as X(X). 11 ,X 12 ,X 21 ,X 22 …, X mn ), m and n are integers greater than 0, X mn This represents the nth attribute under candidate object m; the array information is shown below:

[0098]

[0099] Among them, K i A is used to represent the mean of a certain attribute. i These are used to represent the rule attribute coefficients of candidate objects determined based on the attribute information of the candidate objects. For example, A1, A2, and A3 can represent the rule attribute coefficients corresponding to satisfaction, complaint rate, and timeliness, respectively.

[0100] The summary data of work order attribute weight coefficients can be denoted as B(B) 11 B 12 B 21 B 22 …, B mn ), m and n are integers greater than 0, B mnThis represents the nth work order attribute feature value matched under candidate object m, and F(B) is the matching coefficient function. The array information is shown in the following formula:

[0101]

[0102] Among them, B i Used to represent attribute feature values ​​determined based on the target work order classification tag information, such as B1, B2, and B3, which can respectively represent the work order processing flow, work order touchpoint, and work order business area. i This is used to represent the personnel profile information of the candidate.

[0103] In this case, the matching degree between the target work order and the candidate work order is determined based on the matching coefficient, the number of target work orders, and the number of work orders being processed for the candidate work orders. The matching degree calculation function is shown in the following formula:

[0104]

[0105] F(X i ) and F(B i ) is the matching coefficient function, E n U is used to indicate the target number of work orders within the target time period. i This is used to represent the number of work orders being processed by the candidate object within the target time period. F(p,e) is a step function defined based on the attributes of the target work order, where p represents the number of work orders being processed by the candidate object within the target time period, and e represents the number of target work orders to be processed within the target time period. This is a constant used to balance the matching degree algorithm, so as to achieve automatic matching and dispatch of target work orders.

[0106] After determining the matching degree of the candidate objects, the target candidate object can be determined from the candidate objects based on the matching degree, and then the target candidate object can be determined as the second target processing object.

[0107] Understandably, a first threshold can be set to determine the target candidate from the pool of candidates. If the matching degree of the current candidate is greater than or equal to the first threshold, the current candidate can be used as the target candidate.

[0108] Once a target candidate is identified, if the target candidate accepts the target work order, then the target work order will be designated as the second target to be processed.

[0109] According to the work order allocation method in this application, by considering the classification tags of the work order and the personnel profile of the candidate, and then calculating the matching degree, the automatic and accurate matching between the target work order and the candidate is achieved, which effectively improves the efficiency and accuracy of work order allocation.

[0110] In some embodiments, before determining a third target processing object in the target processing team, the work order allocation method of this application embodiment further includes: determining the target processing team associated with the target work order.

[0111] It should be noted that when a rare target work order type is added, or when the target work order is more complex, or other situations cause the personnel profile of the target work order to match zero with the candidate, a predefined number of qualified candidate candidates can be selected from the candidate candidates as the target processing team.

[0112] In this case, the candidate can be divided into multiple candidate teams based on the organizational structure of the candidate and the type of historical work orders corresponding to the candidate.

[0113] In this embodiment, each alternative team can handle a specific type of target work order, and the alternative objects in the alternative teams have organizational structure correlation, which can realize effective linkage between multiple business departments to improve the processing capability of target work orders.

[0114] For example, candidates from multiple departments in a certain region who are handling a certain type of work order can be divided into a candidate team. Of course, other specific methods can also be used to divide the candidate team, such as dividing a certain number of candidates with rich work experience and high business skills into a target team. There are no restrictions on the way the candidate team is divided.

[0115] Based on the type of the target work order, a target processing team is selected from multiple candidate teams, and then a third target processing object is selected from the target processing team. The third target processing object can process the target work order in a timely manner.

[0116] In some embodiments, all candidate objects in the target processing team can be directly used as third target processing objects to process the target work order.

[0117] In other embodiments, the target work order can be input into the target matching model described above to obtain a third target processing object determined by the target matching model from the target processing team.

[0118] Understandably, the target matching model can determine the matching degree of each candidate by analyzing the personnel profiles of the candidates in the target processing team and the type of the target work order. It can further preset a second threshold to filter the candidates, thereby enabling the selection of an appropriate number of third target processing objects from the target processing team.

[0119] Understandably, the number of objects processed by the third target can be one or more.

[0120] According to the work order allocation method of this application embodiment, a suitable number of third target processing objects are further determined from the target processing team through a target matching model. This can minimize the number of third target processing objects while meeting processing requirements, thereby improving the overall processing efficiency of the candidate objects.

[0121] In some embodiments, after determining the target processing team, the work order allocation method of this application embodiment further includes: if no third target processing object is determined, changing the status of the target work order to an unprocessed work order.

[0122] Understandably, if a suitable third target cannot be identified, the target work order can be placed in the work order pool, and this target work order will be considered an unprocessed work order. Alternative targets can send requests to process unprocessed work orders in the work order pool.

[0123] If a request is received from a candidate object, that candidate object can be used as the fourth target object for processing.

[0124] In this case, the system receives the request information from the fourth target processing object for the unprocessed work order and sends the target work order to the fourth target processing object, which can then process the target work order in a timely manner.

[0125] According to the work order allocation method of this application embodiment, by switching the status of unallocated target work orders so that alternative objects can bid for and allocate them, the probability of target work orders being processed in a timely manner can be increased, thereby improving the allocation efficiency of target work orders.

[0126] In some embodiments, when an actual processing object is determined, the actual processing object is one of a first target processing object, a second target processing object, a third processing object, and a fourth processing object.

[0127] Understandably, when the actual processing object is the first target processing object, the target work order is allocated according to the work order mapping table; when the actual processing object is the second target processing object, the target work order is allocated through the target matching model; when the actual processing object is the third target processing object, the target work order is allocated by determining the third target processing object from the target processing team; and when the actual processing object is the fourth target processing object, the target work order is allocated based on the request information of the alternative objects.

[0128] like Figure 2 As shown, the work order allocation method in this application embodiment further includes steps 210, 220, 230 and 240.

[0129] Step 210: Send the target work order to the actual processing object.

[0130] In this step, the target work order is sent to the identified actual processing object, which can process the target work order in a timely manner.

[0131] Step 220: If the target work order is completed, obtain the processing data of the target work order.

[0132] The processing data for target work orders includes the processing time, satisfaction level, complaint details, and timeliness of the actual work order being processed. After a target work order is completed, the processing data can be automatically archived and stored.

[0133] Step 230: Update the personnel profile of the actual processing object based on the processed data.

[0134] It is understandable that, based on the processing data of the actual processing objects, the attribute coefficient array of the actual processing objects is updated to obtain the updated personnel profile of the actual processing objects.

[0135] Step 240: If the amount of data processed is greater than or equal to the target value, use the target work order as a sample and the actual processing object after updating the personnel profile as the label to update the target matching model.

[0136] It is understandable that after the number of completed target work orders reaches a certain number, the amount of processed data is greater than or equal to the target value. The target value is a pre-set value, such as a value between 30 and 100.

[0137] In some embodiments, after processing 50 target work orders, the 50 target work orders and their corresponding processing data can be used as samples for deep learning of the target matching model. Then, the actual processing objects after updating the personnel profiles can be used as labels to continuously improve the target matching model and obtain an updated target matching model.

[0138] According to the work order allocation method of this application embodiment, by using the target work orders that have been processed as samples to continue to improve the target matching model, the matching accuracy of the target matching model can be further improved, thereby improving the allocation efficiency and accuracy of the target work orders.

[0139] The work order allocation device provided in the embodiments of this application is described below. The work order allocation device described below can be referred to in correspondence with the work order allocation method described above.

[0140] Reference Figure 3 The work order allocation device in this application embodiment includes a first processing module 310, a second processing module 320 and a third processing module 330.

[0141] The first processing module 310 is used to determine the first target processing object corresponding to the target work order from the work order mapping table based on the type of the target work order. The work order mapping table is a preset mapping table between the types of work orders and the candidate objects.

[0142] The second processing module 320 is used to input the target work order into the target matching model when the first target processing object has not been determined, and to obtain the second target processing object output by the target matching model. The target matching model is trained using historical work orders as samples and the processing objects corresponding to the historical work orders as labels.

[0143] The third processing module 330 is used to determine a third target processing object from the target processing team when the second target processing object has not been determined. The target processing team is associated with the target work order.

[0144] In some embodiments, the second processing module 320 is further configured to input the target work order into the feature extraction layer of the target matching model to obtain the target word vector matrix output by the feature extraction layer; input the target word vector matrix into the classification layer of the target matching model to obtain the classification label information output by the classification layer; and input the classification label information into the matching layer of the target matching model to obtain the second target processing object output by the matching layer.

[0145] In some embodiments, the second processing module 320 is further configured to determine the matching coefficient between the target work order and the candidate object based on the classification label information and the personnel profile of the candidate object, wherein the personnel profile is determined based on the attribute information of the processing object; determine the matching degree between the target work order and the candidate object based on the matching coefficient, the number of target work orders and the number of candidate objects in the processing work orders; and determine the second target processing object from the candidate objects based on the matching degree.

[0146] In some embodiments, the work order allocation device of this application further includes a fourth processing module, which is used to determine the target processing team associated with the target work order; the third processing module is further used to input the target work order into the target matching model to obtain a third target processing object determined by the target matching model from the target processing team.

[0147] In some embodiments, the fourth processing module is further configured to divide the candidate objects into multiple candidate teams based on the organizational structure affiliation information of the candidate objects and the type of historical work orders corresponding to the processing objects; and to determine the target processing team from the multiple candidate teams based on the type of the target work order.

[0148] In some embodiments, the work order allocation device of this application further includes a fifth processing module, which is used to change the status of the target work order to an unprocessed work order when a third target processing object is not determined; receive request information from a fourth target processing object for the unprocessed work order; and send the target work order to the fourth target processing object.

[0149] In some embodiments, when an actual processing object is determined, the actual processing object is one of a first target processing object, a second target processing object, a third processing object, and a fourth processing object. The work order allocation device of this application embodiment further includes a sixth processing module, which is used to send the target work order to the actual processing object; when the target work order is completed, obtain the processing data of the target work order; update the personnel profile of the actual processing object based on the processing data; and when the amount of processing data is greater than or equal to the target value, update the target matching model using the target work order as a sample and the actual processing object after updating the personnel profile as a label.

[0150] Figure 4 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 4 As shown, the electronic device may include a processor 410, a communication interface 420, a memory 430, and a communication bus 440, wherein the processor 410, the communication interface 420, and the memory 430 communicate with each other through the communication bus 440. The processor 410 can call the computer program in the memory 430 to execute the steps of the work order allocation method, such as: determining the first target processing object corresponding to the target work order from a work order mapping table based on the type of the target work order to be allocated; the work order mapping table is a preset mapping table between work order types and candidate objects; if the first target processing object is not determined, inputting the target work order into a target matching model to obtain the second target processing object output by the target matching model; the target matching model is trained using historical work orders as samples and the processing objects corresponding to historical work orders as labels; if the second target processing object is not determined, determining a third target processing object from the target processing team, and the target processing team is associated with the target work order.

[0151] Furthermore, the logical instructions in the aforementioned memory 430 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a 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 described in the various embodiments of this application. 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.

[0152] On the other hand, this application also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can perform the steps of the work order allocation method provided in the above embodiments, such as: determining a first target processing object corresponding to the target work order from a work order mapping table based on the type of the target work order to be allocated, wherein the work order mapping table is a preset mapping table between work order types and candidate objects; if the first target processing object is not determined, inputting the target work order into a target matching model to obtain a second target processing object output by the target matching model, wherein the target matching model is trained using historical work orders as samples and the processing objects corresponding to historical work orders as labels; if the second target processing object is not determined, determining a third target processing object in the target processing team, wherein the target processing team is associated with the target work order.

[0153] On the other hand, embodiments of this application also provide a processor-readable storage medium storing a computer program for causing a processor to execute the steps of the methods provided in the above embodiments, such as: determining a first target processing object corresponding to the target work order from a work order mapping table based on the type of the target work order to be assigned, the work order mapping table being a preset mapping table between work order types and candidate objects; if the first target processing object is not determined, inputting the target work order into a target matching model to obtain a second target processing object output by the target matching model, the target matching model being trained using historical work orders as samples and the processing objects corresponding to historical work orders as labels; if the second target processing object is not determined, determining a third target processing object in a target processing team, the target processing team being associated with the target work order.

[0154] The processor-readable storage medium can be any available medium or data storage device that the processor can access, including but not limited to magnetic memory (e.g., floppy disk, hard disk, magnetic tape, magneto-optical disk (MO)), optical memory (e.g., CD, DVD, BD, HVD), and semiconductor memory (e.g., ROM, EPROM, EEPROM, non-volatile memory (NAND FLASH), solid-state drive (SSD)).

[0155] The device embodiments described above are merely illustrative. 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 modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0156] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0157] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application 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 this application.

Claims

1. A work order allocation method, characterized in that, include: Based on the type of the target work order to be assigned, the first target processing object corresponding to the target work order is determined from the work order mapping table. The work order mapping table is a preset mapping table between the types of work orders and the candidate objects. The work order mapping table includes candidate objects corresponding to various types of work orders; Determining the first target processing object corresponding to the target work order from the work order mapping table includes: matching the first target processing object that matches the type of the target work order from the candidate objects according to the type of the target work order; If the first target processing object is not determined, the target work order is input into the target matching model to obtain the second target processing object output by the target matching model. The target matching model is trained using historical work orders as samples and the processing objects corresponding to the historical work orders as labels. If the second target processing object is not identified, a third target processing object is identified within the target processing team, which is associated with the target work order; Before determining the third target processing object in the target processing team, the method further includes: determining the target processing team associated with the target work order; determining the third target processing object in the target processing team includes: inputting the target work order into the target matching model to obtain the third target processing object determined by the target matching model from the target processing team; The step of determining the target processing team associated with the target work order includes: dividing the candidate objects into multiple candidate teams based on the organizational structure affiliation information of the candidate objects and the types of historical work orders corresponding to the processing objects; and determining the target processing team from the multiple candidate teams based on the type of the target work order.

2. The work order allocation method according to claim 1, characterized in that, The step of inputting the target work order into the target matching model to obtain the second target processing object determined by the target matching model includes: The target work order is input into the feature extraction layer of the target matching model to obtain the target word vector matrix output by the feature extraction layer; The target word vector matrix is ​​input into the classification layer of the target matching model to obtain the classification label information output by the classification layer; The classification label information is input into the matching layer of the target matching model to obtain the second target processing object output by the matching layer.

3. The work order allocation method according to claim 2, characterized in that, The classification label information is input into the matching layer of the target matching model to obtain the second target processing object output by the matching layer, including: Based on the classification label information and the personnel profiles of the candidate objects, the matching coefficient between the target work order and the candidate objects is determined, wherein the personnel profiles are determined based on the attribute information of the processing objects; Based on the matching coefficient, the number of target work orders, and the number of work orders being processed by the candidate objects, the matching degree between the target work orders and the candidate objects is determined. Based on the matching degree, the second target processing object is determined from the candidate objects.

4. The work order allocation method according to claim 1, characterized in that, After identifying the target processing team, the method further includes: If the third target processing object is not identified, the status of the target work order will be changed to an unprocessed work order; Receive request information from the fourth target processing object regarding the unprocessed work order; The target work order is sent to the fourth target processing object.

5. The work order allocation method according to claim 4, characterized in that, If the actual processing object is determined to be one of the first target processing object, the second target processing object, the third target processing object, and the fourth target processing object, the method further includes: Send the target work order to the actual processing object; If the target work order is completed, obtain the processing data of the target work order; Based on the processed data, update the personnel profile of the actual processed object; If the amount of data processed is greater than or equal to the target value, the target matching model is updated using the target work order as a sample and the actual processing object after updating the personnel profile as a label.

6. A work order distribution device, characterized in that, include: The first processing module is used to determine the first target processing object corresponding to the target work order from the work order mapping table based on the type of the target work order. The work order mapping table is a preset mapping table between the types of work orders and candidate objects. The work order mapping table includes candidate objects corresponding to various types of work orders; Determining the first target processing object corresponding to the target work order from the work order mapping table includes: matching the first target processing object that matches the type of the target work order from the candidate objects according to the type of the target work order; The second processing module is used to input the target work order into the target matching model when the first target processing object is not determined, and obtain the second target processing object output by the target matching model. The target matching model is trained using historical work orders as samples and the processing objects corresponding to the historical work orders as labels. The third processing module is used to determine a third target processing object from the target processing team when the second target processing object has not been determined, wherein the target processing team is associated with the target work order; The fourth processing module is used to determine the target processing team associated with the target work order; the third processing module is also used to input the target work order into the target matching model to obtain the third target processing object determined by the target matching model from the target processing team. The fourth processing module is also used to divide the candidate objects into multiple candidate teams based on the organizational structure affiliation information of the candidate objects and the types of historical work orders corresponding to the processing objects; and to determine the target processing team from the multiple candidate teams based on the type of the target work order.

7. An electronic device comprising a processor and a memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the work order allocation method according to any one of claims 1 to 5.

8. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the work order allocation method according to any one of claims 1 to 5.

Images