CN117745223A

CN117745223A - Method, system, electronic equipment and medium for generating power platform work order data

Info

Publication number: CN117745223A
Application number: CN202311770525.9A
Authority: CN
Inventors: 张子健; 周明; 张靖; 马永; 王丽; 管建超; 程航; 薛晓茹; 郭洋; 徐道磊; 路宇; 张永梅; 许畅; 范莹; 刘佳; 许冬; 王俊; 赵煜阳; 涂冰花
Original assignee: Information and Telecommunication Branch of State Grid Anhui Electric Power Co Ltd
Current assignee: Information and Telecommunication Branch of State Grid Anhui Electric Power Co Ltd
Priority date: 2023-12-21
Filing date: 2023-12-21
Publication date: 2024-03-22

Abstract

The invention relates to the technical field of power work order analysis and processing, in particular to a power platform work order data generation method, a system, electronic equipment and a medium. If the work order data is not completed in the set time, generating a standard question sentence, reminding a customer representative of inquiring through a visual interface, and after the monitoring work order data is completed, checking the work order data by the customer representative and issuing. According to the method and the system for obtaining the keywords, keywords are obtained through the call audio of the data mining client, and the keywords are recorded after the keywords are judged to be the required information of the work order data, so that the generated work order data has higher accuracy.

Description

Method, system, electronic equipment and medium for generating power platform work order data

Technical Field

The invention relates to the technical field of power worksheet analysis and processing, in particular to a power platform worksheet data generation method, a system, electronic equipment and a medium.

Background

The electric power grid bears the important functions of guaranteeing safety, economy, cleanliness and sustainable power supply, and the grid customer service center is a communication bridge erected between a power grid enterprise and power users and provides quick and high-quality service for power grid customers. The power grid customer service center mainly receives complaints, reports, fault reports, service consultation, suggestions, comments and expressing services of users. The electric power work order data provides the requirements of the clients and related information to the processing department, and the processing department solves the problems of the clients according to the work order data.

The current electric power work order data is mainly generated by a customer representative through telephone communication with the customer and recording related information. When a customer call is received, the time for collecting customer appeal information is the most original, so that accurate and comprehensive recording of customer requirements is a key for efficiently solving customer appeal. However, since the customer representative receives different business processes and has a large amount of business volume every day, and the customer representative has different self quality and service skills due to working experience of the customer representative and personal factors of the customer, the customer representative and the customer have understanding deviation in the communication process or main information is omitted due to negligence of the customer representative, so that accurate work order data cannot be generated.

Aiming at the generation of accurate worksheet data in the related technology, the problem of customer appeal can be solved satisfactorily, users are satisfied, and no effective solution is proposed at present.

Disclosure of Invention

The invention aims to provide a method, a system, electronic equipment and a medium for generating power platform worksheet data, which are used for solving the problem of inaccurate power platform worksheet data generation, obtaining keywords through the conversation audio of a big data mining client, and obtaining a processing department through the keywords; by monitoring the conversation time, a standard question sentence is generated to prompt a customer representative, so that the accuracy of work order data is improved, and the problem of customer appeal is ensured to be satisfactorily solved.

In order to achieve the above purpose, the embodiment of the present application provides the following technical solutions:

in one aspect, an embodiment of the present application provides a method for generating power platform worksheets data, where the method includes:

when the client is monitored to be accessed into a power grid customer service center, acquiring real-time call audio of a client representative end and the client end, wherein the real-time call audio is obtained through a single-channel voice separation and enhancement algorithm and recorded as first audio, the first audio is obtained through a real-time noise reduction algorithm based on a statistical model to obtain second audio, the second audio is sentence-processed through a wavelet transformation method to obtain third audio, and the third audio is converted into text content through a hidden Markov model algorithm;

Obtaining a word list from the text content by a jieba word segmentation method, removing the stop words of the word list, obtaining a word list and inputting the word list; extracting keywords from the word list through a TF-IDF algorithm; training a preset electric Word library by using a Word2vec model, calculating a relation value of a keyword and a fault characteristic Word through a cosine similarity algorithm to obtain a first association coefficient, and marking the keyword when the first association coefficient is larger than a first set threshold;

inquiring historical work order data according to a client number and marking the historical work order data as first data, acquiring a first keyword of the first data, calculating a relation value between the keyword and the first keyword through a cosine similarity algorithm to obtain a second association coefficient, and judging that the historical work order data is repeated and generating a key label when the second association coefficient is larger than a second set threshold value to acquire a processing department of the historical work order data; when the second association coefficient is smaller than a second set threshold, judging different faults, inquiring historical work order data according to the client attribute, recording the historical work order data as second data, acquiring a second keyword of the second data, calculating a relation value between the keyword and the second keyword through a cosine similarity algorithm to obtain a third association coefficient, and when the third association coefficient is larger than the second set threshold, acquiring a processing department of the historical work order data;

When the conversation time between the customer representative and the customer is monitored to be longer than the set time, generating standard questions on a visual interface according to a customer service center telephone service rule method by using information which is not recorded in the work order data according to the set sequence; when monitoring that the conversation between the customer representative and the customer is finished, generating a prompt sentence on a visual interface by using data which are not generated in the work order; and after the monitoring work order data is generated, generating review prompt information on a visual interface.

Optionally, the method for generating the power platform worksheet data further includes:

the method comprises the steps of obtaining average processing time of a customer representative historical work order, marking the time of the customer representative finishing the work order as work order finishing time, generating confirmation prompt information on a visual interface when the finishing time of the monitored work order is smaller than the first time, and calculating the ratio of the work order finishing time to the first time to obtain a first efficiency coefficient; obtaining a first historical standard deviation rate of a customer representative, and multiplying a first correction efficiency coefficient by the first standard deviation rate to obtain a first accuracy coefficient of work order data; obtaining a correction coefficient according to the working experience duration of the customer representative, and multiplying the first accurate coefficient by the correction coefficient to obtain a first correction accurate coefficient; and when the monitored first correction accuracy coefficient is smaller than the set safety value, generating confirmation information on the visual interface.

Optionally, the method further comprises:

acquiring average processing time of the customer representative in a set period to be recorded as second time, and calculating the ratio of the work order completion time to the second time to obtain a second efficiency coefficient; obtaining a second historical standard single rate of the customer representative in a set period, and multiplying a second efficiency coefficient by the second standard single rate to obtain a second accurate coefficient of work order data; and obtaining a correction coefficient according to the working experience time length of the customer representative, multiplying the second correction coefficient by the correction coefficient to obtain a second correction coefficient, and generating an early warning label for the type of service processed by the customer representative when the second correction coefficient is smaller than the first correction coefficient and the ratio of the second correction coefficient to the first correction coefficient is smaller than a set early warning value, wherein the system adds the customer representative into a customer representative queue of the service without the early warning label.

the emergency degree of the work order is obtained through a comprehensive fuzzy evaluation method, and when the judgment score is larger than the set score, the work order is added with an emergency label; when the judgment score is smaller than the set score, adding a general label to the work order; the set score is determined by adopting a Delphi method; the comprehensive fuzzy evaluation method is to use the keywords as the judgment factors of the emergency degree, determine the evaluation scores of the keywords through a fuzzy statistical method, determine the weights of the keywords through an analytic hierarchy process, and obtain the judgment scores of the keywords through a linear weighting method; when the label is judged to be an urgent label, generating an urgent confirmation prompt on a visual interface; when the keyword triggers the sensitive word, the keyword is immediately judged as an urgent label.

In a second aspect, based on the same inventive concept, an embodiment of the present application provides an electric power platform work order data generating system, the system including:

the text content acquisition module is used for acquiring real-time call audio of a client representative end and a client end when the client is monitored to be accessed to a power grid customer service center, wherein the real-time call audio is obtained through a single-channel voice separation and enhancement algorithm and is recorded as first audio, the first audio is obtained through a real-time noise reduction algorithm based on a statistical model to obtain second audio, the second audio is sentence-processed through a wavelet transformation method to obtain third audio, and the third audio is converted into text content through a hidden Markov model algorithm;

the keyword acquisition module is used for obtaining a word list from the text content through a jieba word segmentation method, removing the stop words of the word list, obtaining a word list and inputting the word list; extracting keywords from the word list through a TF-IDF algorithm; training a preset electric Word library by using a Word2vec model, calculating a relation value of a keyword and a fault characteristic Word through a cosine similarity algorithm to obtain a first association coefficient, and marking the keyword when the first association coefficient is larger than a first set threshold;

The processing department acquisition module is used for inquiring historical work order data according to the client numbers and marking the historical work order data as first data, acquiring first keywords of the first data, calculating relation values of the keywords and the first keywords through a cosine similarity algorithm to obtain second association coefficients, judging repeated faults and generating key labels when the second association coefficients are larger than a second set threshold value, and acquiring the processing department of the historical work order data; when the second association coefficient is smaller than a second set threshold, judging different faults, inquiring historical work order data according to the client attribute, recording the historical work order data as second data, acquiring a second keyword of the second data, calculating a relation value between the keyword and the second keyword through a cosine similarity algorithm to obtain a third association coefficient, and when the third association coefficient is larger than the second set threshold, acquiring a processing department of the historical work order data;

the work order confirmation generation module is used for generating standard questions on the visual interface according to a customer service center telephone service rule method according to a set sequence by using information which is not input in the work order data when the conversation time between a customer representative and a customer is monitored to be longer than the set time; when monitoring that the conversation between the customer representative and the customer is finished, generating a prompt sentence on a visual interface by using data which are not generated in the work order; and after the monitoring work order data is generated, generating review prompt information on a visual interface.

Optionally, the power platform worksheet data generating system further includes:

the first acquisition module is used for acquiring the average processing time of the historical work order of the customer representative, recording the time of the work order completion of the customer representative as the work order completion time, generating confirmation prompt information on the visual interface when the monitored work order completion time is smaller than the first time, and calculating the ratio of the work order completion time to the first time to obtain a first efficiency coefficient; obtaining a first historical standard deviation rate of a customer representative, and multiplying a first correction efficiency coefficient by the first standard deviation rate to obtain a first accuracy coefficient of work order data; obtaining a correction coefficient according to the working experience duration of the customer representative, and multiplying the first accurate coefficient by the correction coefficient to obtain a first correction accurate coefficient; and when the monitored first correction accuracy coefficient is smaller than the set safety value, generating confirmation information on the visual interface.

Optionally, the system further comprises:

the second acquisition module is used for acquiring the average processing time of the customer representative in the set period as second time, and calculating the ratio of the work order completion time to the second time to obtain a second efficiency coefficient; obtaining a second historical standard single rate of the customer representative in a set period, and multiplying a second efficiency coefficient by the second standard single rate to obtain a second accurate coefficient of work order data; and obtaining a correction coefficient according to the working experience time length of the customer representative, multiplying the second correction coefficient by the correction coefficient to obtain a second correction coefficient, and generating an early warning label for the type of service processed by the customer representative when the second correction coefficient is smaller than the first correction coefficient and the ratio of the second correction coefficient to the first correction coefficient is smaller than a set early warning value, wherein the system adds the customer representative into a customer representative queue of the service without the early warning label.

Optionally, the system further comprises:

the emergency degree judging module is used for obtaining a judging score from the emergency degree of the work order through a comprehensive fuzzy evaluation method, and adding an emergency label to the work order when the judging score is larger than a set score; when the judgment score is smaller than the set score, adding a general label to the work order; the set score is determined by adopting a Delphi method; the comprehensive fuzzy evaluation method is to use the keywords as the judgment factors of the emergency degree, determine the evaluation scores of the keywords through a fuzzy statistical method, determine the weights of the keywords through an analytic hierarchy process, and obtain the judgment scores of the keywords through a linear weighting method; when the label is judged to be an urgent label, generating an urgent confirmation prompt on a visual interface; when the keyword triggers the sensitive word, the keyword is immediately judged as an urgent label.

In a third aspect, based on the same inventive concept, an embodiment of the present application provides an electronic device for generating power platform worksheet data, the electronic device including a memory and a processor.

The memory is used for storing a computer program; the processor is used for realizing the steps of the power platform work order data generation method when executing the computer program.

In a fourth aspect, based on the same inventive concept, embodiments of the present application provide a medium on which a computer program is stored, the computer program implementing the steps of the above-described method for generating power platform worksheet data when executed by a processor.

The beneficial effects of the invention are as follows:

1. obtaining keywords through the conversation audio of the big data mining client, and obtaining a processing department through the keywords; by monitoring the conversation time, a standard question sentence is generated to prompt a customer representative, so that the accuracy of work order data is improved, and the problem of customer appeal is ensured to be satisfactorily solved.

2. And by generating the work order label attribute, the processing efficiency of the customer representative on the work order is improved.

Drawings

In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings that are needed in the embodiments will be briefly described below, it being understood that the following drawings only illustrate some embodiments of the present invention and therefore should not be considered as limiting the scope, and other related drawings may be obtained according to these drawings without inventive effort for a person skilled in the art.

FIG. 1 is a schematic flow chart of a method for generating power platform worksheet data according to an embodiment of the present invention;

Fig. 2 is a schematic structural diagram of an electronic device for generating power platform worksheet data according to an embodiment of the present invention.

Detailed Description

For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention are clearly and completely described, and it is obvious that the described embodiments are some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

Meanwhile, in the description of the present invention, the terms "first", "second", and the like are used only to distinguish the description, and are not to be construed as indicating or implying relative importance.

Before the explanation, the application scenario of the present invention needs to be described, the power platform work order data can provide effective information for the processing department and assist the processing department to solve the problem, at present, the customer representative generates the work order by communicating with the customer, the problem that the work order data is inaccurate due to the difference of self quality and service skills caused by the working experience of the customer representative, and the problem of secondary communication confirmation possibly exists, and the negative influence is brought to the experience of the customer and the timeliness of the work order dispatch. The system automatically acquires the processing department according to the keywords so as to improve the dispatching efficiency of the work orders. In order to facilitate understanding of the inventive concept, the inventor has exemplified a specific case, for example, a elderly person living alone in a 72-year old city of a mountain, and when the elderly person calls a power grid customer service center due to a power outage occurring in the residence time, the relevant information cannot be perfectly expressed, the talk time will be too long, and the customer representative is required to guide the elderly person step by step with great patience and care to provide the relevant information, otherwise important information will be omitted, so that the problem of the elderly person cannot be solved timely. In another case, for example, a customer representative of a new job in a customer service center of the power grid in Yunnan province sometimes has errors in information recording or unclear communication contents when communicating with the customer due to lack of working experience, and inquires the customer again, and sometimes the customer resorts to not verifying related information or not merging the work orders, and sends the work orders, thereby causing repetitive work and manpower waste to the processing department. Both cases are problems existing in the actual process of generating the worksheet data, and the technical problem solved by the technical scheme in the aspect is generated based on the problems.

Example 1

As shown in fig. 1, the present embodiment provides a method for generating power platform worksheet data, the method including:

firstly, when a client is monitored to be accessed into a power grid customer service center, acquiring real-time call audio of a client representative end and the client end, wherein the real-time call audio is obtained through a single-channel voice separation and enhancement algorithm and recorded as first audio, the first audio is obtained through a real-time noise reduction algorithm based on a statistical model to obtain second audio, the second audio is sentence-processed through a wavelet transformation method to obtain third audio, and the third audio is converted into text content through a hidden Markov model algorithm; the single-channel voice separation algorithm is to separate the voice frequency of the client representative end from the voice frequency of the client representative end in the real-time call voice frequency of the client end, only the voice frequency of the client end is reserved, and the voice data of the client representative can be obtained by adopting a model-based single-channel voice separation method, the voice of the client representative is trained to obtain voice model parameters of the client representative, and then the voice data of the client representative is participated in the voice frequency mixing of the client representative end and the client end for separation based on the voice model parameters of the client representative. The voice enhancement algorithm can improve voice quality, eliminate background music, and improve voice intelligibility, for example, a customer representative can better hear the customer's appeal content through voice enhancement, and can communicate effectively. The communication audio of the client often generates noise because of the surrounding environment of the client, the voice quality is reduced due to the existence of background noise, the performance of a voice processing system is reduced due to the pollution of the environmental noise, the recognition of the audio is difficult, and the converted text content is wrong, so that the noise of the audio of the client is reduced through a real-time noise reduction algorithm based on a statistical model, clearer audio can be obtained, and the accuracy of the text content is improved. The real-time noise reduction algorithm based on the statistical model belongs to the most common algorithm for real-time audio noise reduction, and is to estimate the noise and the voice components corresponding to each frequency point in an audio frequency spectrum by using a statistical method, and then remove the noise by using the statistical method. The call audio of the client is key information of customer appeal content, and in order to ensure timeliness of acquiring work order data, a wavelet transform-based method is adopted to divide the real-time call audio of the client. The wavelet transformation belongs to a time-frequency analysis method and has a multi-resolution analysis characteristic, can display local characteristics of time and frequency of an audio signal, can divide the audio signal into different sub-bands, and the wavelet coefficients can reflect energy distribution of the audio signal along a time axis in each sub-band. Since the energy distribution of the audio signal in adjacent subbands is similar, the wavelet coefficients of the adjacent subbands are highly correlated, but the noise and mute signals do not have this property, the speech signal is split according to the cross-correlation of the wavelet coefficients of the adjacent subbands, and the audio is split. For example, when the audio expression content of a certain customer is "i is a resident in the south aster district, power failure occurs in my home, the building has a power failure in the upper and lower floors", and when the first stop occurs in the monitored customer audio, i.e. the audio at the stop is a noise or mute signal, the phrase obtained by the wavelet transform-based method is "i is a resident in the south aster district", i is a segment of audio, and the hidden markov model algorithm is converted into text content; when the second pause occurs in the customer audio, namely the audio at the pause is a noise or mute signal, the phrase obtained by the wavelet transformation-based method is 'My home blackout', and the 'My home blackout' is used as a section of audio and is converted into text content by a hidden Markov model algorithm;

Then, obtaining a word list from the text content through a jieba word segmentation method, removing the stop words of the word list, obtaining a word list and inputting the word list; extracting keywords from the word list through a TF-IDF algorithm; training a preset electric Word library by using a Word2vec model, calculating a relation value of a keyword and a fault characteristic Word through a cosine similarity algorithm to obtain a first association coefficient, and marking the keyword when the first association coefficient is larger than a first set threshold; the jieba word segmentation technique uses a Chinese word stock to determine the association probability between Chinese characters and the composition word groups with large probability between Chinese characters, thereby forming word segmentation results, and the Chinese word stock in the embodiment is a percentage word stock. Removing stop words means removing words such as 'mock', 'yes' and the like in a word list, prepositions, adverbs, connecting words and the like, and a word list obtained by removing the stop words is used as a text content abstract to be recorded into accepted contents in a work order, and the text content provides information reference for customer representatives. And extracting keywords from the word list after the stop words are removed through a TF-IDF algorithm, wherein an electric power industry corpus used for calculating the IDF value is from a modern Chinese corpus of the national language commission. Training a preset electric Word library by using a Word2vec model, and mapping each vocabulary of the preset electric Word library into a dense vector by using Word2vec so as to convert discrete words into continuous vector space representation. The fault characteristic words are contained in a preset electric word stock and are words for specially describing fault characteristics. The method comprises the steps of calculating a relation value of a keyword and a fault characteristic word by adopting a cosine similarity algorithm, wherein the cosine similarity algorithm is to assume that a vector of the keyword is A, an absolute value is |A| and a vector of the fault characteristic word is B, an absolute value is |B| and a cosine similarity calculation formula is cosθ=A.times.B/(|A|B|), the value range of the cosine similarity calculation formula is-1 to 1, when directions of the two vectors are the same, the cosine similarity is the largest, the value is 1, and when directions of the two vectors are completely different, the cosine similarity is the smallest, and the value is-1. For example, when the text content "power failure occurs in my home" and the word list is "i/m/suddenly/power failure/m" by jieba word segmentation technology, the stop word is removed to obtain "i/m/power failure", and the key word is extracted by TF-IDF algorithm, because "i" and "m" belong to common words, the key word "power failure" is obtained. The fault feature words also comprise 'power failure', the key words 'power failure' and the fault feature words 'power failure' are calculated, the obtained cosine similarity is 1, the 'power failure' is recorded into the work order acceptance content and marked by special colors or fonts, and the key points of the content which can be more clearly obtained when the customer representative reviews the work order data are beneficial to the customer;

Secondly, inquiring historical work order data according to a client number and marking the historical work order data as first data, acquiring a first keyword of the first data, calculating a relation value between the keyword and the first keyword through a cosine similarity algorithm to obtain a second association coefficient, and judging that the historical work order data is repeated and generating a key label when the second association coefficient is larger than a second set threshold value, so as to acquire a processing department of the historical work order data; and when the second association coefficient is smaller than a second set threshold, judging different faults, inquiring historical work order data according to the client attribute, recording the historical work order data as second data, acquiring a second keyword of the second data, calculating a relation value between the keyword and the second keyword through a cosine similarity algorithm to obtain a third association coefficient, and when the third association coefficient is larger than the second set threshold, acquiring a processing department of the historical work order data. The customer representative in the work order data manually matches the processing department in the work order according to the requirements of the customer, the effectiveness is slow, the processing department is selected to be unmatched with the actual situation due to the fact that the customer representative has insufficient experience or inaccurate judgment, the work order is subjected to the situation of returning the work order, the timeliness of processing the work order is seriously influenced, the processing department of the same type of faults in the historical work order of the customer is obtained through the customer number generated by the work order data, and the processing department is used as the processing department of the work order data, so that the accuracy of the selection of the processing department in the work order data can be ensured, and the timeliness of processing the work order is ensured. For example, when a system detects that the fault content of the customer's requirement belongs to repeated faults, a processing department in the historical work order data is automatically acquired as a unit for processing the customer's fault, and when the customer is monitored to have an important label, the processing department will strengthen fault investigation, avoid the fault to occur again, so as to effectively improve the customer satisfaction. However, another customer resides in the new urban area, the content of the customer's appeal belongs to the first occurrence, for example, the keyword of the customer's appeal is "power failure", then the system will query the historical work order according to the customer's attribute, and the processing department of the same type of fault is obtained through the cosine similarity algorithm, and the processing department is used as the processing department for processing the customer. Of course, the above-described query scope is queried according to customer attributes, which include industrial electricity customers, agricultural electricity customers, commercial electricity customers, and residential electricity customers, because the fault keywords of different customer attributes may be the same, but there are differences in the processing departments;

Secondly, when the conversation time between the customer representative and the customer is monitored to be longer than the set time, generating standard questions on a visual interface according to a customer service center telephone service rule method by using information which is not recorded in the work order data according to the set sequence; when monitoring that the conversation between the customer representative and the customer is finished, generating a prompt sentence on a visual interface by using data which are not generated in the work order; and after the monitoring work order data is generated, generating review prompt information on a visual interface. According to different types of service statistics, for example, about 95% of customer representatives in the consultation service can complete the generation of consultation work order data in 1 minute, and about 95% of customer representatives in the fault repair service can complete the generation of fault repair work order data in 2 minutes. When the system detects that the conversation time is greater than 2 minutes, the system indicates that the communication between the current customer representative and the customer is abnormal, such as that the customer representative has reduced work attention due to physical reasons or has insufficient experience due to just entering, or the customer has tension, poor language expression capability, high age, and the like, and the conversation time between the customer representative and the customer is too long, so that the customer representative is required to continuously inquire the customer, or the communication between the customer representative and the customer is finally caused to exceed the set time due to the fact that the communication understanding between the two parties is not smooth. At this time, the system will produce standard question sentences for the information which is not recorded in the work order data according to the customer service center telephone service standard specification term on the visual interface of the customer representative terminal, where the customer service center telephone service standard specification term is a specification rule formulated by the power grid customer service center, for example, the standard question sentence is "please ask you for no electricity in one household, no electricity in the surrounding, or no electricity in the whole cell or region? ", avoid customer representatives asking or missing inquiry information multiple times, such as just asking" ask you for a user to have no electricity, or all around to have no electricity "without asking" or the whole cell or sector to have no electricity? The fault determination may not be accurate because the occurrence of "the entire cell or the patch is dead" may be a planned outage, and the customer does not notice the relevant outage notification or the planned outage notification at the grid end is not sent out in time. After the call is ended, some data in the work order data need to be manually input, for example, the same customer or the same incoming call reflects the same fault, the customer representative does not perform work order association combination and issues the work order, the work order is repeated and forms return occurs, and a prompt sentence is generated on a visual interface of a customer representative terminal aiming at the information, for example, "please confirm, whether the work order is associated? ". When the monitoring work order data are all generated, a customer representative is reminded to check the work order, for example, forced check countdown time is set, and class option errors and the like in the work order data generation process are avoided.

Secondly, the power platform work order data generation method further comprises the following steps:

additionally, the average processing time of the customer representative historical work order is obtained and recorded as first time, the time of the customer representative finishing the work order is recorded as work order finishing time, when the finishing time of the monitored work order is smaller than the first time, confirmation prompt information is generated on a visual interface, and the ratio of the work order finishing time to the first time is calculated to obtain a first efficiency coefficient; obtaining a first historical standard deviation rate of a customer representative, and multiplying a first correction efficiency coefficient by the first standard deviation rate to obtain a first accuracy coefficient of work order data; obtaining a correction coefficient according to the working experience duration of the customer representative, and multiplying the first accurate coefficient by the correction coefficient to obtain a first correction accurate coefficient; and when the monitored first correction accuracy coefficient is smaller than the set safety value, generating confirmation information on the visual interface. The average processing time of the historical work order of the customer representative is used as a key factor for judging the working state of the customer representative on the same day, and when the time for the customer representative to process the work order is detected to be far smaller than the historical average processing time, the customer representative does not sufficiently communicate and know about customer requirements, does not accurately record relevant information or has the condition of unexamined verification. The standard bill rate is the ratio of the difference value of subtracting the withdrawal number from the total number of the historical work bill by the customer representative, and the withdrawal number is the fact that the customer representative does not collect the effective information of the customer in the work bill description, so that the processing department cannot process according to the content generated by the customer representative, or the customer representative can predict power failure (such as arrearage power failure, planned power failure, and the like), issues the power failure after the customer representative is not verified, and similar faults are not associated. Or because the working experience time of different customer representatives is different, for example, a customer representative with long working experience time completes a work order in a short time, but the accuracy of the work order is relatively high, and the correction coefficient is matched according to the working experience time of the customer representative. The correction coefficient is based on the corresponding relation between the accuracy of the work order and the working experience time of a large number of customer representatives according to the power grid statistics along with the change of the working experience time. When the first correction accuracy coefficient detected by the system is smaller than the set safety value, the fact that the recording of the work order data is possibly unsafe is indicated, namely the risk of returning the work order is extremely high, and at the moment, a customer representative is required to verify whether the recording of the work order data is imperfect or not, and the condition that the recording is omitted or information is not verified is indicated.

acquiring average processing time of the customer representative in a set period to be recorded as second time, and calculating the ratio of the work order completion time to the second time to obtain a second efficiency coefficient; obtaining a second historical standard single rate of the customer representative in a set period, and multiplying a second efficiency coefficient by the second standard single rate to obtain a second accurate coefficient of work order data; and obtaining a correction coefficient according to the working experience time length of the customer representative, multiplying the second correction coefficient by the correction coefficient to obtain a second correction coefficient, and generating an early warning label for the type of service processed by the customer representative when the second correction coefficient is smaller than the first correction coefficient and the ratio of the second correction coefficient to the first correction coefficient is smaller than a set early warning value, wherein the system adds the customer representative into a customer representative queue of the service without the early warning label. The recent relevant data of the customer representative is a reliable index of the accuracy of generating the relational worksheet data, and because the customer representative has abnormal working states in recent or a period of time due to personal factors and the like, the conditions of low working efficiency, worksheet return and the like of the customer representative can be effectively avoided, so that a great deal of labor cost is lost and negative effects are brought to the customer by acquiring the recent average processing time and the quasi-worksheet rate of the customer representative and comparing the historical data. For example, when the first correction accuracy coefficient is 0.95, the second correction accuracy coefficient is 0.84, the ratio of the second correction accuracy coefficient to the first correction accuracy coefficient is 0.88, assuming that the set early warning value is 0.90, it is satisfied that the second correction accuracy coefficient 0.84 is smaller than the first correction accuracy coefficient 0.95 and the ratio of the second correction accuracy coefficient to the first correction accuracy coefficient 0.88 is smaller than the set early warning value 0.90, the early warning label is added to the type of the business processed by the customer representative, which indicates that the customer representative is currently processing such business with abnormal work, and the customer representative can be retrained or learn to enhance the business so as to improve the work efficiency.

the emergency degree of the work order is obtained through a comprehensive fuzzy evaluation method, and when the judgment score is larger than the set score, the work order is added with an emergency label; when the judgment score is smaller than the set score, adding a general label to the work order; the set score is determined by adopting a Delphi method; the comprehensive fuzzy evaluation method is to use the keywords as the judgment factors of the emergency degree, determine the evaluation scores of the keywords through a fuzzy statistical method, determine the weights of the keywords through an analytic hierarchy process, and obtain the judgment scores of the keywords through a linear weighting method; when the label is judged to be an urgent label, generating an urgent confirmation prompt on a visual interface; when the keyword triggers the sensitive word, the keyword is immediately judged as an urgent label. When the label is judged to be urgent, an urgent confirmation prompt is generated on a visual interface to remind a customer representative of judging whether the label is urgent or not, and if the label is urgent, the generation and the dispatch of the work order are quickened; when the keyword triggers the sensitive word, the keyword is immediately judged as an urgent label. The comprehensive fuzzy evaluation method is a method for processing fuzzy information by converting information of uncertainty or fuzziness into an evaluation result of certainty. The delta film is an expert decision and prediction method which can be used for technical evaluation, risk evaluation and the like. Determining the evaluation scores of the keywords by a fuzzy statistical method; the analytic hierarchy process is a subjective weight design to set the weight of keywords; the linear weighting method calculates a composite score by multiplying the score of a keyword by a weight set. And calculating the sensitivity of the keywords and the sensitive words through a cosine similarity algorithm, and judging that the keywords belong to the sensitive words when the sensitivity is larger than a set value, wherein the sensitive words are contained in a preset electric word stock and comprise casualties, explosions, fires and the like.

Example 2

The embodiment provides a power platform work order data generation system, which comprises:

Wherein, the power platform worksheet data generation system further comprises:

Secondly, the power platform work order data generation system further comprises:

It should be noted that, regarding the system in the above embodiment, the specific manner in which the respective modules perform the operations has been described in detail in the embodiment regarding the method, and will not be described in detail herein.

Example 3

Corresponding to the above method embodiment, the embodiment of the present disclosure further provides an electronic device for generating power platform worksheet data, where the electronic device for generating power platform worksheet data described below and the method for generating power platform worksheet data described above may be referred to correspondingly.

As shown in fig. 2, the apparatus 800 may include: a processor 801, a memory 802. The device 800 may also include one or more of a multimedia component 803, an input/output (I/O) interface 804, and a communication component 805.

The processor 801 is configured to control overall operation of the apparatus 800 to perform all or part of the steps in a method for generating power platform worksheet data as described above. The memory 802 is used to store various types of data to support operation at the device 800, which may include, for example, instructions for any application or method operating on the device 800, as well as application-related data, such as contact data, messages, pictures, audio, video, and so forth. The memory 802 may be implemented by any type or combination of volatile or non-volatile memory devices, such as Static Random Access Memory (SRAM), electrically erasable programmable Read-only memory (EEPROM), erasable programmable Read-only memory (EPROM), programmable Read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk, or optical disk. The multimedia component 803 may include a screen and an audio component. Wherein the screen may be, for example, a touch screen, the audio component being for outputting and/or inputting audio signals. For example, the audio component may include a microphone for receiving external audio signals. The received audio signals may be further stored in the memory 802 or transmitted through the communication component 805. The audio assembly further comprises at least one speaker for outputting audio signals. The I/O interface 804 provides an interface between the processor 801 and other interface modules, which may be a keyboard, mouse, buttons, etc. These buttons may be virtual buttons or physical buttons. The communication component 805 is used for wired or wireless communication between the device 800 and other devices. Wireless communication, such as Wi-Fi, bluetooth, near Field Communication (NFC), 2G, 3G, or 4G, or a combination of one or more thereof, the corresponding communication component 805 may include: wi-Fi module, bluetooth module, NFC module.

In an exemplary embodiment, the device 800 may be implemented by one or more Application Specific Integrated Circuits (ASICs), digital Signal Processors (DSPs), digital Signal Processing Devices (DSPDs), programmable Logic Devices (PLDs), field Programmable Gate Arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components for performing one of the above-described power platform single data generation methods.

In another exemplary embodiment, there is also provided a computer readable storage medium including program instructions which, when executed by a processor, implement the steps of a power platform worksheet data generation method described above. For example, the computer readable storage medium may be the memory 802 including the program instructions described above, which may be executed by the processor 801 of the apparatus 800 to perform one of the power platform worksheet data generation methods described above.

Example 4

Corresponding to the above method embodiments, the present disclosure further provides a readable storage medium, where a readable storage medium described below and a power platform worksheet data generating method described above may be referred to correspondingly with each other.

A readable storage medium having stored thereon a computer program which when executed by a processor implements the steps of a power platform worksheet data generating method of the above-described method embodiment.

The readable storage medium may be a usb disk, a removable hard disk, a Read-only memory (ROM), a random access memory (RandomAccessMemory, RAM), a magnetic disk, or an optical disk, and the like.

The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. The power platform work order data generation method is characterized by comprising the following steps:

2. The power platform work order data generation method according to claim 1, further comprising:

3. The power platform work order data generation method according to claim 2, further comprising:

4. The power platform work order data generation method according to claim 1, further comprising:

obtaining a judgment score by the emergency degree of the work order through a comprehensive fuzzy evaluation method, and adding an emergency label into the work order when the judgment score is larger than a set score; when the judgment score is smaller than the set score, adding a general label to the work order; the set score is determined by adopting a Delphi method; the comprehensive fuzzy evaluation method is to use the keywords as the judgment factors of the emergency degree, determine the evaluation scores of the keywords through a fuzzy statistical method, determine the weights of the keywords through an analytic hierarchy process, and obtain the judgment scores of the keywords through a linear weighting method; when the label is judged to be an urgent label, generating an urgent confirmation prompt on a visual interface; when the keyword triggers the sensitive word, the keyword is immediately judged as an urgent label.

5. An electric power platform worksheet data generation system, the system comprising:

6. The power platform worksheet data generation system of claim 5, further comprising:

7. The power platform worksheet data generation system of claim 5, further comprising:

8. The power platform worksheet data generation system of claim 5, further comprising:

9. An electronic device, comprising:

a memory for storing a computer program;

a processor for carrying out the method steps of any one of claims 1-4 when executing a program stored on a memory.

10. A medium having stored thereon a computer program, which when executed by a processor, implements the method steps of any of claims 1-4.