Power customer service timeliness rate evaluation method, storage medium, equipment and program product

Abstract

The invention discloses a power customer service timeliness rate evaluation method, a storage medium, equipment and a program product. The method comprises the following steps: acquiring a work order newly initiated by a power customer; based on a pre-obtained mode cluster of a work order process, matching a mode cluster to which a work order newly initiated by the power customer belongs; according to the compliance time corresponding to the mode cluster to which the work order newly initiated by the power customer belongs, evaluating whether the total execution time of the work order process approval of the work order newly initiated by the power customer and the execution time of each approval link are overtime; wherein the mode cluster of the work order process is obtained by performing mode mining on the approval process of the archived work order by using an ordered Apriori algorithm and a decision tree model and performing process optimization according to the timeout degree of the overtime archived work order, and the compliance time comprises the total execution time of the work order process approval and the execution time of each approval link. The work order process approval efficiency can be improved, and the requirements of customers can be responded in time.

Description

Technical Field

[0001] The present invention relates to an evaluation method, a storage medium, a device and a program product for the timeliness rate of power customer service, belonging to the technical field of power customer service. Background Art

[0002] As the power grid belongs to the service industry of "public infrastructure management", how to improve service efficiency is the key concern of the power grid.

[0003] At present, the State Grid has opened online and offline customer service channels. Customers can submit their own requirements at any time through the online or offline customer service channels to form corresponding work orders. However, in the approval process of work orders, there is a problem of redundant approval processes, and the approval efficiency of the work order process is low, resulting in the failure to respond to customers' requirements in a timely manner. At the same time, there is currently a lack of follow-up on the work order approval process. Summary of the Invention

[0004] The purpose of the present invention is to overcome the deficiencies in the prior art and provide an evaluation method, a storage medium, a device and a program product for the timeliness rate of power customer service, which can improve the approval efficiency of the work order process and respond to customers' requirements in a timely manner. To achieve the above purpose, the present invention is implemented by the following technical solutions:

[0005] In a first aspect, the present invention provides an evaluation method for the timeliness rate of power customer service, including:

[0006] Obtain a work order newly initiated by a power customer;

[0007] Based on the pre-obtained pattern clusters of the work order process, match the pattern cluster to which the work order newly initiated by the power customer belongs;

[0008] Use the compliance time corresponding to the pattern cluster to which the work order newly initiated by the power customer belongs to evaluate whether the total execution time of the work order process approval and the execution time of each approval link of the work order newly initiated by the power customer exceed the time limit;

[0009] Among them, the pattern clusters of the work order process are obtained by performing pattern mining on the approval process of archived work orders using the ordered Apriori algorithm and the decision tree model and optimizing the process according to the degree of timeout of the archived work orders that exceed the time limit. The compliance time includes the total execution time of the work order process approval and the execution time of each approval link.

[0010] In combination with the first aspect, optionally, obtaining the pattern clusters of the work order process by performing pattern mining on the approval process of archived work orders using the ordered Apriori algorithm and the decision tree model and optimizing the process according to the degree of timeout of the archived work orders that exceed the time limit includes:

[0011] Use the ordered Apriori algorithm to mine the frequent sequences in the archived work order approval process, and obtain a preliminary set of work order process patterns;

[0012] Based on the approval process, use a decision tree model to classify the preliminary set of work order process patterns to obtain a preliminary pattern cluster of the work order process;

[0013] According to the degree of timeout of the archived work orders, optimize the process of the pattern cluster to obtain a pattern cluster of the work order process.

[0014] Combined with the first aspect, optionally, the use of the ordered Apriori algorithm to mine the frequent sequences in the archived work order approval process to obtain a preliminary set of work order process patterns includes:

[0015] Preprocess the approval process data of the archived work orders to obtain an approval link sequence;

[0016] Generate an initial candidate sequence set based on the approval link sequence , indicating the th link of the th work order;

[0017] Initialize the support threshold T;

[0018] Use the ordered Apriori algorithm to perform support iteration calculation on the initial candidate sequence set, and screen out the frequent item sets with support greater than the support threshold T. The support calculation is to calculate the frequency of the sequences of each link appearing in the candidate sequence set;

[0019] Merge the approval processes of the frequent item sets to obtain a preliminary set of work order process patterns.

[0020] Combined with the first aspect, optionally, the process optimization of the pattern cluster according to the degree of timeout of the archived work orders to obtain a pattern cluster of the work order process includes:

[0021] Initialize the threshold of outliers and the threshold of highly abnormal values ;

[0022] Use the outlier identification method to identify outliers and highly abnormal values for the timeout of each approval link in the archived work orders with timeout, and eliminate the highly abnormal values to obtain the processing time of each approval link in the pattern cluster to which the archived work orders with timeout belong;

[0023] Initialize the threshold of the degree of timeout ;

[0024] Calculate the statistics of the pattern cluster to which the archived work orders with timeout belong , where the statistic is used to evaluate the timeout degree of the pattern cluster and is expressed by the following formula:

[0025] ;

[0026] For pattern clusters, an approval process optimization reminder for the approval links with outliers in the pattern cluster is output, and the statistic of the pattern cluster is recalculated after the process optimization ; if it meets after the process optimization, the process optimization is effective; if it still after the process optimization, the support threshold T is increased in steps of , the outlier threshold is increased in steps of , and the highly outlier threshold is increased in steps of , until ; until ;

[0027] In response to the statistics of all pattern clusters all meeting , the compliance time of the pattern cluster to which the timed-out archived work order belongs converges, and the pattern clusters of the work order process and the corresponding compliance time of each pattern cluster are output.

[0028] Combined with the first aspect, optionally, matching the pattern cluster to which the work order newly initiated by the power customer belongs based on the pre-obtained pattern clusters of the work order process includes:

[0029] Extracting approval process data from the work order newly initiated by the power customer, and respectively performing feature matching between the extracted approval process data and the approval processes of the pre-obtained pattern clusters of the work order process;

[0030] If there are features of a single pattern cluster in the extracted approval process data, output the pattern cluster as the pattern cluster to which the work order newly initiated by the power customer belongs;

[0031] If there are features of multiple pattern clusters in the extracted approval process data, output the pattern cluster with the longest same feature as the pattern cluster to which the work order newly initiated by the power customer belongs.

[0032] Combined with the first aspect, optionally, aggregating the work orders that have completed the work order process approval includes:

[0033] Based on the pre-obtained pattern clusters of the work order process, matching the pattern cluster to which the work order that has completed the work order process approval belongs;

[0034] Clustering the work orders that have completed the approval and are not processed and belong to the same pattern cluster to obtain multiple work order clusters;

[0035] Batch process the work orders that have completed approval but are not yet processed based on a preset configurable logical order.

[0036] In combination with the first aspect, optionally, it further includes outputting reminders, including:

[0037] Output a timeout reminder in response to the total execution time of the work order process approval for a newly initiated work order by a power customer exceeding the time limit;

[0038] Output a warning reminder in response to the execution time of each approval link in the work order process approval for a newly initiated work order by a power customer being less than the preset threshold from the compliance time of that approval link.

[0039] In a second aspect, the present invention provides a computer-readable storage medium, on which a computer program / instructions are stored. When the computer program / instructions are executed by a processor, the steps of the method for evaluating the timeliness rate of power customer services described in the first aspect are implemented.

[0040] In a third aspect, the present invention provides a computer device, including:

[0041] A memory for storing computer programs / instructions;

[0042] A processor for executing the computer programs / instructions to implement the steps of the method for evaluating the timeliness rate of power customer services described in the first aspect.

[0043] In a fourth aspect, the present invention provides a computer program product, including computer programs / instructions. When the computer programs / instructions are executed by a processor, the steps of the method for evaluating the timeliness rate of power customer services described in the first aspect are implemented.

[0044] Compared with the prior art, the beneficial effects achieved by the method, storage medium, device, and program product for evaluating the timeliness rate of power customer services provided by the embodiments of the present invention include:

[0045] The present invention obtains the work orders newly initiated by power customers; based on the pre-obtained pattern clusters of the work order process, it matches the pattern cluster to which the work orders newly initiated by power customers belong; through the pre-classification of the pattern clusters, the present invention can match the work orders newly initiated by power customers to the pattern clusters, reducing the time for manual classification and subjective judgment errors, and improving the work order processing efficiency;

[0046] The present invention uses the compliance time corresponding to the pattern cluster to which the work orders newly initiated by power customers belong to evaluate whether the total execution time and the execution time of each approval link in the work order process approval for the work orders newly initiated by power customers exceed the time limit; the present invention independently monitors the execution time of each approval link, enabling real-time monitoring of the work order approval progress;

[0047] The mode clusters of the work order process of the present invention are obtained by using the ordered Apriori algorithm and the decision tree model to perform pattern mining on the approval process of archived work orders and optimizing the process according to the overtime degree of the archived work orders that exceed the time limit; the present invention can extract frequent sequences from the data of archived work orders by using the ordered Apriori algorithm, and can effectively identify high-frequency approval processes; the present invention can classify based on the work order attribute features by using the decision tree model to form accurate mode clusters; the present invention combines the optimization according to the overtime degree, can specifically improve inefficient links, and solve the problem of redundant approval processes;

[0048] The compliance time of the present invention includes the total execution time of the work order process approval and the execution time of each approval link; the present invention directly associates the compliance time with each approval link of the mode cluster to ensure that the time standard matches the actual process; it can reflect the actual needs of different process modes, avoid a one-size-fits-all time standard, and improve the accuracy of evaluation;

[0049] When the total execution time of the work order process approval for a newly initiated work order by an electricity customer exceeds the time limit, the present invention outputs an overtime reminder; when the execution time of each approval link for a newly initiated work order by an electricity customer is less than the preset threshold from the compliance time of that approval link, the present invention outputs a warning reminder; the present invention outputs a reminder when it times out and outputs a warning when each approval link is approaching timeout, which can avoid affecting the customer service quality due to delays; the present invention can follow up the work order approval process, can respond to the needs of customers in a timely manner, can improve service efficiency and customer satisfaction; the present invention can give an early warning by setting a preset threshold and can also avoid frequent false alarms;

[0050] The present invention can improve the efficiency of work order process approval, shorten the customer waiting time, respond to the needs of customers in a timely manner, and enhance service reliability and satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 is a schematic flowchart of a method for evaluating the timeliness rate of an electricity customer service in Embodiment 1 of the present invention;

[0052] Figure 2 is a schematic flowchart of the aggregation process in a method for evaluating the timeliness rate of an electricity customer service in Embodiment 1 of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0053] The present invention will be further described below with reference to the accompanying drawings. The following embodiments are only used to more clearly illustrate the technical solutions of the present invention and should not be used to limit the protection scope of the present invention.

[0054] Embodiment 1:

[0055] As Figure 1As shown in the figure, this embodiment provides an evaluation method for the timeliness rate of power customer service, including:

[0056] Obtain the work orders newly initiated by power customers;

[0057] Based on the pre-obtained pattern clusters of the work order process, match the pattern cluster to which the work order newly initiated by the power customer belongs;

[0058] Use the compliance time corresponding to the pattern cluster to which the work order newly initiated by the power customer belongs to evaluate whether the total execution time of the work order process approval and the execution time of each approval link of the work order newly initiated by the power customer exceed the time limit.

[0059] In this embodiment, the ordered Apriori algorithm and the decision tree model are used to perform pattern mining on the approval process of the archived work orders, and process optimization is performed according to the degree of overtime of the archived work orders with overtime to obtain the pattern clusters of the work order process. The compliance time includes the total execution time of the work order process approval and the execution time of each approval link.

[0060] Each work order has its own approval process. This embodiment only targets the archived work orders because the unarchived work orders do not have a complete approval process.

[0061] For work orders of the same type, using the idea of the decision tree model, combining the algorithm kernel of the Apriori association rule and the given support threshold T, for work orders of the same business type, find the ordered frequent sequences, and organize the set of ordered frequent sequences to form a preliminary pattern cluster. Then, according to the degree of overtime of the archived work orders with overtime, perform process optimization on the pattern cluster to which they belong, obtain the pattern clusters of the work order process, and update the compliance time.

[0062] The specific steps to obtain the pattern clusters of the work order process are as follows:

[0063] Step 1: Use the ordered Apriori algorithm to mine the frequent sequences in the approval process of the archived work orders to obtain a preliminary set of work order process patterns.

[0064] Denote the th link of the th work order as .

[0065] Since this embodiment only targets the archived work orders, for any archived work order, the name of the last link (the nth link) is file archiving, that is, , and {file archiving} constitutes a frequent 1-item set.

[0066] Further, determine whether the last two steps of the work order form a frequent 2-itemset. If it forms a frequent 2-itemset, continue to find the frequent 3-itemset; otherwise, terminate... If there is no frequent n-itemset, terminate.

[0067] Step 1.1: Preprocess the approval process data of the archived work orders to obtain the approval step sequence.

[0068] Step 1.2: Generate an initial candidate sequence set based on the approval step sequence 。

[0069] Step 1.3: Initialize the support threshold T.

[0070] Step 1.4: Use the ordered Apriori algorithm to perform iterative support calculation on the initial candidate sequence set, and filter out the frequent itemsets with support greater than the support threshold T.

[0071] The support calculation is to calculate the frequency of the sequence of each step appearing in the candidate sequence set.

[0072] The iterative support calculation is specifically expressed as follows:

[0073] Step 1.4.1: Calculate the support: ,

[0074] Among them, is the statistical frequency, is the (n - 1)-th step (the second last step) of the th work order, is the n-th step (the last step) of the th work order, is the sequence set of all steps of the th work order (candidate sequence set),

[0075] When m > T, then the 2-itemset S is a frequent itemset of the initial candidate sequence set W, and it is also called the frequent 2-itemset of W.

[0076] It should be noted that in an ordered itemset, a frequent p-itemset is called a subset of a frequent q-itemset if the sequence of the p-itemset is a subsequence of the q-itemset, that is, p can be found in q.

[0077] Step 1.4.2: Based on the frequent 2-itemset, continue to find the frequent 3-itemset.

[0078] If { , } forms a frequent 2-itemset, then further determine whether { , , } Whether it constitutes a frequent 3-itemset.

[0079] Calculate the support: ,

[0080] Among them, is the second iteration calculation.

[0081] When m > T, then the 3-itemset S is a frequent itemset of the initial candidate sequence set W, and it is also called the frequent 3-itemset of W.

[0082] Step 1.4.3: For any work order, if it does not have a frequent n-itemset, then skip this work order when searching for the n + 1-itemset. If there is a frequent n-itemset, then continue to search for whether there is an n + 1-itemset. When there is no frequent n-itemset, stop the search.

[0083] It should be noted that the complexity of the traditional Apriori association rule is , and in this embodiment, because it is ordered, so when there is no frequent n-itemset, then there must be no frequent n + 1-itemset. Therefore, the complexity of the algorithm is .

[0084] If the frequent n-itemset of W is a subset of the frequent n + 1-itemset, then this frequent n-itemset is regarded as a pattern cluster of the frequent n + 1-itemset. Find the set Q composed of all frequent itemsets (that is, the preliminary work order process pattern set). Each element in Q is a maximum frequent itemset of W, and each such maximum frequent itemset is a pattern cluster. All ordered subsets of the frequent itemset that are greater than 1 are called the characteristics of this pattern cluster.

[0085] Step 1.5: Merge the approval processes of the frequent itemsets to obtain the preliminary work order process pattern set.

[0086] Specifically, after stopping the search, perform pattern cluster merging.

[0087] Step 2: Based on the approval process, use the decision tree model to classify the preliminary work order process pattern set to obtain the pattern clusters of the preliminary work order process.

[0088] Step 2.1: Extract the approval process data of the archived work orders.

[0089] Step 2.2: According to the approval process, use the decision tree model to classify the frequent itemsets in the preliminary work order process pattern set to obtain the pattern clusters of the preliminary work order process.

[0090] Step 3: Optimize the process of the pattern clusters according to the degree of overtime of the archived work orders with overtime, and obtain the pattern clusters of the work order process.

[0091] Step 3.1: Initialize the threshold of outliers and the threshold of highly outliers .

[0092] Step 3.2: Use the outlier identification method to identify outliers and highly outliers in the overtime of the archived work orders with overtime, eliminate the highly outliers, and obtain the processing time of each approval link in the pattern cluster to which the archived work orders with overtime belong.

[0093] Step 3.3: Initialize the threshold of the degree of overtime .

[0094] Step 3.4: Calculate the statistic of the pattern cluster to which the archived work orders with overtime belong .

[0095] Statistic is used to evaluate the degree of overtime of the pattern cluster, and is expressed by the following formula:

[0096] ,

[0097] Relatively speaking, the more the overtime ratio of the work order and the more the overtime ratio of the approval link, the more serious the overtime degree of the pattern cluster. Among them, is essentially a penalty function.

[0098] For the pattern cluster, output the process optimization reminder of the approval link with outliers in this pattern cluster, and recalculate the statistic of this pattern cluster after process optimization ; if it meets after process optimization, the process optimization is effective and the compliance time of this pattern cluster does not need to be adjusted.

[0099] Specifically, perform process optimization according to the process optimization reminder. The process optimization includes: merging approval links for pattern clusters with severe overtime, introducing automated approval nodes, or deleting redundant links.

[0100] If it still after process optimization, increase the support threshold T in step , increase the threshold of outliers in step , and adjust and increase the threshold of highly outliers in step , and adjust and increase the threshold of highly outliers in step , until .

[0101] ​​It should be noted that the staff need to regularly pay attention to the support threshold T, the threshold of outliers and the threshold of highly abnormal values , so as to avoid the large value affecting the timeliness rate of power customer service.

[0102] Step 3.5: In response to the statistics of each mode cluster all meet , then the compliance time of the mode cluster to which the timed-out archived work order belongs converges, and the mode cluster of the work order process and the corresponding compliance time of each mode cluster are output.

[0103] It should be noted that the staff need to regularly pay attention to the updated compliance time to avoid the large compliance time affecting the timeliness rate of power customer service.

[0104] This embodiment can extract frequent sequences from the data of archived work orders by using the ordered Apriori algorithm, and can effectively identify high-frequency approval processes. This embodiment can classify based on the work order attribute characteristics by using the decision tree model to form accurate mode clusters. This embodiment is optimized in combination with the degree of timeout, can specifically improve inefficient links, and solve the problem of redundant approval processes.

[0105] This embodiment directly associates the compliance time with each approval link of the mode cluster to ensure that the time standard matches the actual process; it can reflect the actual needs of different process modes, avoid a one-size-fits-all time standard, and improve the accuracy of evaluation.

[0106] As Figure 1 shown, this embodiment obtains the work orders newly initiated by power customers; based on the pre-obtained mode clusters of the work order process, it obtains the mode cluster to which the work orders newly initiated by power customers belong. Through the pre-classification of the mode clusters, this embodiment can match the work orders newly initiated by power customers to the mode clusters, reduce the time of manual classification and subjective judgment errors, and improve the work order processing efficiency.

[0107] As Figure 1 shown, this embodiment matches the mode cluster to which the work orders newly initiated by power customers belong based on the pre-obtained mode clusters of the work order process, including:

[0108] Extract the approval process data from the work orders newly initiated by power customers, and respectively perform feature matching between the extracted approval process data and the approval processes of the pre-obtained mode clusters of the work order process;

[0109] If there are features of a single mode cluster in the extracted approval process data, output this mode cluster as the mode cluster to which the work orders newly initiated by power customers belong;

[0110] If there are features of multiple pattern clusters in the extracted approval process data, the pattern cluster to which the longest same feature belongs is output as the pattern cluster to which the work order newly initiated by the power customer belongs.

[0111] For example: If there are 5 approval links in the approval process data extracted from the approval chain of the work order newly initiated by the power customer, and if no pattern cluster with the features of these 5 approval links simultaneously is found in the pattern cluster, find the pattern cluster with the features of the last 4 approval links until a pattern cluster with the features of several approval links is found in the pattern cluster. Output the pattern cluster to which the longest same feature belongs as the pattern cluster to which the work order newly initiated by the power customer belongs. Use the compliance time corresponding to this pattern cluster to evaluate whether the total execution time of the work order process approval for the work order newly initiated by the power customer and the execution time of each approval link exceed the time limit.

[0112] It should be noted that the working hours of the State Grid are defined as 8:30 to 12:00 and 13:00 to 17:00 on weekdays.

[0113] This embodiment further includes output reminders:

[0114] In response to the total execution time of the work order process approval for the work order newly initiated by the power customer exceeding the time limit, an overtime reminder is output;

[0115] In response to the execution time of each approval link of the work order process approval for the work order newly initiated by the power customer being less than the preset threshold from the compliance time of this approval link, a warning reminder is output.

[0116] Furthermore, in response to the overtime reminder, process optimization is performed. The process optimization includes: merging approval links for the pattern cluster with serious overtime, introducing automated approval nodes, or deleting redundant links. In response to the warning reminder, process reminder is performed, such as sending a text message to remind the approver.

[0117] This embodiment can monitor the progress of work order approval in real time, actively trigger a warning when approaching overtime, and avoid affecting the quality of customer service due to delays. The warning mechanism set in this embodiment can follow up the work order approval process, can respond to customer needs in a timely manner, and can improve service efficiency and customer satisfaction. By setting a preset threshold, this embodiment can give an early warning and avoid frequent false alarms.

[0118] As Figure 2 shown, this embodiment also performs aggregation processing on the work orders that have completed the work order process approval, including:

[0119] Based on the pattern cluster of the work order process obtained in advance, match the pattern cluster to which the work order that has completed the work order process approval belongs;

[0120] Cluster the work orders that have completed approval and are not processed and belong to the same pattern cluster to obtain multiple work order clusters;

[0121] Batch process the work orders that have completed approval but are not yet processed based on a preset configurable logical order.

[0122] For example, the preset configurable logical order is the priority of each category.

[0123] This embodiment can improve the approval efficiency of the work order process, shorten the customer waiting time, respond to customer needs in a timely manner, and enhance service reliability and satisfaction.

[0124] Embodiment 2:

[0125] Taking two types of archived work orders with work order types of low-voltage residential new installation, capacity increase, and transfer of ownership as examples, this embodiment executes the pattern cluster of the work order process obtained by pattern mining of the approval process of the archived work orders using the ordered Apriori algorithm and decision tree model in Embodiment 1 and optimizing the process according to the timeout degree of the pattern.

[0126] In a certain province, there are a total of 804 work orders with the work order type of low-voltage residential new installation and capacity increase from August 2022 to August 2023.

[0127] There are 468 archived work orders with the approval process of "door-to-door service ===> spatial topology maintenance ===> file archiving". There are 240 archived work orders with the approval process of "business hall acceptance ===> door-to-door service ===> spatial topology maintenance ===> file archiving". There are 96 archived work orders with the approval process of "door-to-door service ===> file archiving".

[0128] The binomial sets are: = {spatial topology maintenance, file archiving}, = {door-to-door service, file archiving}. Among them The number of occurrences = 468 + 240 = 708 times, The support degree of = 708 / 804 = 0.88 > the specified support degree threshold of 0.05. So is a frequent 2-item set of low-voltage residential new installation and capacity increase work orders. Among them The number of occurrences of is 96 times, The support degree of = 96 / 804 ≈ 0.12 > the specified support degree threshold of 0.1, so is also a frequent 2-item set of low-voltage residential new installation and capacity increase work orders.

[0129] The trinomial set has = {door-to-door service, spatial topology maintenance, file archiving}, The support degree of = 708 / 804 = 0.88 > the specified threshold of 0.05, so is a frequent 3-item set of low-voltage residential new installation and capacity increase work orders.

[0130] Four-item set ={Business hall acceptance, on-site service, spatial topology maintenance, file archiving} is the frequent 4-item set of work orders for new installations and capacity increases of low-voltage residential customers.

[0131] There are two work order process mode clusters for the archived work orders of new installations and capacity increases of low-voltage residential customers, namely {Business hall acceptance, on-site service, spatial topology maintenance, file archiving} and {On-site service, file archiving}. Therefore, the work orders for new installations and capacity increases of low-voltage residential customers are divided into two categories according to the transfer mode. One category follows the approval process of {Business hall acceptance, on-site service, spatial topology maintenance, file archiving}, and the other category follows the approval process of {On-site service, file archiving}.

[0132] In a certain province, from August 2022 to August 2023, there were a total of 3,409 work orders of the transfer type.

[0133] The approval process and quantity of the archived work orders of the transfer type are shown in Table 1.

[0134] Table 1 Approval process and quantity of the archived work orders of the transfer type

[0135]

[0136] The two-item sets of the archived work orders of the transfer type are shown in Table 2.

[0137] Table 2 Two-item sets of the archived work orders of the transfer type

[0138]

[0139] Serial numbers A1, A2, and A3 are the frequent 2-item sets of the transfer work orders.

[0140] The three-item sets of the archived work orders of the transfer type are shown in Table 3.

[0141] Table 3 Three-item sets of the archived work orders of the transfer type

[0142]

[0143] Serial number B1 is the frequent 3-item set of the transfer work orders.

[0144] The four-item sets of the archived work orders of the transfer type are shown in Table 4.

[0145] Table 4 Four-item sets of the archived work orders of the transfer type

[0146]

[0147] Serial number C1 is a frequent 4-itemset of transfer work orders, and serial number C2 is not a frequent 4-itemset of transfer work orders.

[0148] The five-itemsets of the archived work orders with the work order type of transfer are shown in Table 5.

[0149] Table 5 Five-itemsets of the archived work orders with the work order type of transfer

[0150]

[0151] Serial numbers D1, D2, D3, and D4 are frequent 5-itemsets of transfer work orders.

[0152] The six-itemsets of the archived work orders with the work order type of transfer are shown in Table 6.

[0153] Table 6 Six-itemsets of the archived work orders with the work order type of transfer

[0154]

[0155] Serial numbers E1, E2, and E3 are frequent 6-itemsets of transfer work orders, and serial number E4 is not a frequent 6-itemset of transfer work orders.

[0156] The frequent item sets of the archived work orders with the work order type of transfer are summarized as shown in Table 7.

[0157] Table 7 Frequent item sets of the archived work orders with the work order type of transfer

[0158]

[0159] The mode clusters of the transfer work order process are {{Contract Approval, Contract Approval, Contract Signing, File Archiving}, {On-site Business Acceptance, Contract Review, Contract Approval, Contract Signing, File Archiving}, {Business Hall Acceptance, Survey and Contract Drafting, Contract Review, Contract Approval, Contract Signing, File Archiving}, {On-site Business Acceptance, Contract Review, Contract Review, Contract Approval, Contract Signing, File Archiving}, {On-site Business Acceptance, Contract Review, Contract Approval, Contract Review, Contract Approval, Contract Signing, File Archiving}}.

[0160] Example 3:

[0161] This embodiment provides a computer-readable storage medium, on which computer programs / instructions are stored. When the computer programs / instructions are executed by a processor, the steps of the evaluation method for the power customer service timeliness rate described in Example 1 are implemented.

[0162] Example 4:

[0163] This embodiment provides a computer device, including:

[0164] A memory for storing computer programs / instructions;

[0165] A processor for executing the computer programs / instructions to implement the steps of the method for evaluating the timeliness rate of power customer service described in Embodiment 1.

[0166] Embodiment 5:

[0167] This embodiment provides a computer program product including computer programs / instructions, and when the computer programs / instructions are executed by a processor, the steps of the method for evaluating the timeliness rate of power customer service described in Embodiment 1 are implemented.

[0168] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a completely hardware embodiment, a completely software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memories, CD-ROMs, optical memories, etc.) containing computer-usable program codes.

[0169] The present invention is described with reference to the flowcharts and / or block diagrams of methods, devices (systems), and computer program products according to the embodiments of the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram can be implemented by computer program instructions, and the combination of the processes and / or blocks in the flowchart and / or block diagram can also be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing devices generate a device for implementing the functions specified in one process Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0170] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer-readable memory generate a manufactured product including an instruction device, and the instruction device implements the functions specified in one process Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0171] These computer program instructions can also be loaded onto a computer or other programmable data processing device, so that a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process. Thus, the instructions executed on the computer or other programmable device provide for implementing the functions specified in one process Figure 1One process or multiple processes and / or boxes Figure 1 Steps of functions specified in one box or multiple boxes.

[0172] The embodiments of the present invention have been described above in conjunction with the accompanying drawings. However, the present invention is not limited to the above specific embodiments. The above specific embodiments are merely illustrative rather than restrictive. Under the inspiration of the present invention, those of ordinary skill in the art can also make many forms without departing from the spirit of the present invention and the scope protected by the claims. All of these fall within the protection scope of the present invention.

Claims

1. A method for evaluating the timeliness of electric power customer service, characterized in that: include: Obtain new work orders initiated by power customers; Based on the pattern cluster of the work order process obtained in advance, matching the pattern cluster to which the work order newly initiated by the power customer belongs; By using the compliance time corresponding to the mode cluster to which the work order newly initiated by the power customer belongs, it is evaluated whether the total execution time of the work order process approval and the execution time of each approval link of the work order newly initiated by the power customer have exceeded; Among them, the pattern cluster of the work order process is obtained by using the ordered Apriori algorithm and the decision tree model to conduct pattern mining on the approval process of archived work orders and optimizing the process according to the timeout degree of the timed archived work orders. The compliance time includes the total execution time of the work order process approval and the execution time of each approval link.

2. The method for evaluating the timeliness of electric power customer service according to claim 1 is characterized in that: The ordered Apriori algorithm and decision tree model are used to mine the approval process of archived work orders and optimize the process according to the timeout degree of archived work orders to obtain the pattern cluster of the work order process, including: Use the ordered Apriori algorithm to mine frequent sequences in the approval process of archived work orders and obtain a preliminary set of work order process patterns; Based on the approval process, the preliminary work order process pattern set is classified using a decision tree model to obtain a preliminary work order process pattern cluster; According to the timeout degree of the archived work orders that have been timed out, the pattern cluster is optimized to obtain the pattern cluster of the work order process.

3. The method for evaluating the timeliness of electric power customer service according to claim 2 is characterized in that: The ordered Apriori algorithm is used to mine frequent sequences in the approval process of archived work orders, and a preliminary set of work order process patterns is obtained, including: Pre-process the approval process data of archived work orders to obtain the approval link sequence; Generate an initial set of candidate sequences based on the approval process sequence , Indicates The first links; Initialize support threshold T; The ordered Apriori algorithm is used to iteratively calculate the support of the initial candidate sequence set, and the frequent item sets with support greater than the support threshold T are screened out. The support calculation is to calculate the frequency of occurrence of the sequence of each link in the candidate sequence set; The approval processes of frequent itemsets are merged to obtain a preliminary set of work order process patterns.

4. The method for evaluating the timeliness of electric power customer service according to claim 3 is characterized in that: According to the timeout degree of the archived work orders, the pattern cluster is optimized to obtain the pattern cluster of the work order process, including: Initialize the outlier threshold and the threshold for highly outliers ; The outlier recognition method is used to identify outliers and highly outliers for the timeout time of each approval link in the archived work orders that have been overdue, and highly outliers are removed to obtain the processing time of each approval link in the pattern cluster to which the archived work orders that have been overdue belong. Initialization timeout threshold ; Calculates statistics for the pattern cluster to which the timed archived tickets belong , where the statistic It is used to evaluate the timeout degree of the pattern cluster and is expressed by the following formula: ； for The pattern cluster is output, and the process optimization reminder of the approval link with abnormal values ​​in the pattern cluster is output. After the process optimization, the statistics of the pattern cluster are calculated again. ; If the process is optimized to meet , then the process optimization is effective; if the process is still , with step length Increase the support threshold T, with a step size Increase the outlier threshold , with step length Increase the threshold for highly outliers ,until ; Statistics of responses to all pattern clusters All satisfied , the compliance time of the pattern cluster to which the timed archived work order belongs converges, and the pattern cluster of the work order process and the compliance time corresponding to each pattern cluster are output.

5. The method for evaluating the timeliness of electric power customer service according to claim 1 is characterized in that: The pattern cluster based on the pre-obtained work order process is matched with the pattern cluster to which the work order newly initiated by the power customer belongs, including: Extract approval process data from a new work order initiated by a power customer, and perform feature matching on the extracted approval process data and the approval process of the pattern cluster of the work order process obtained in advance; If the extracted approval process data contains features of a single pattern cluster, the pattern cluster is output as the pattern cluster to which the work order newly initiated by the power customer belongs; If there are features of multiple pattern clusters in the extracted approval process data, the pattern cluster to which the longest identical feature belongs is output as the pattern cluster to which the work order newly initiated by the power customer belongs.

6. The method for evaluating the timeliness of electric power customer service according to claim 1 is characterized in that: Aggregate the work orders that have completed the work order process approval, including: Based on the pattern cluster of the work order process obtained in advance, the pattern cluster to which the work order that has completed the work order process approval belongs is matched; Cluster the approved but unprocessed work orders that belong to the same pattern cluster to obtain multiple work order clusters; Based on the preset configurable logical sequence, batch processing is performed on the approved but unprocessed work orders.

7. The method for evaluating the timeliness of electric power customer service according to claim 1 is characterized in that: Also includes output reminders, including: In response to the total execution time of the work order process approval for the work order newly initiated by the power customer exceeding the time limit, a timeout reminder is output; In response to a work order newly initiated by an electric power customer, a warning reminder is output when the execution time of each approval link in the work order process approval is less than a preset threshold from the compliance time of the approval link.

8. A computer-readable storage medium having a computer program / instruction stored thereon, characterized in that: When the computer program / instruction is executed by a processor, the steps of the method for evaluating the timeliness of electric power customer service described in any one of claims 1-7 are implemented.

9. A computer device, characterized in that: include: Memory, for storing computer programs / instructions; A processor is used to execute the computer program / instructions to implement the steps of the method for evaluating the timeliness of electric customer service according to any one of claims 1-7.

10. A computer program product comprising a computer program / instructions, characterized in that When the computer program / instruction is executed by a processor, the steps of the method for evaluating the timeliness of electric power customer service described in any one of claims 1-7 are implemented.

Images