CN111325475A - Emergency repair work order evaluation factor analysis method based on total log-likelihood algorithm - Google Patents
Emergency repair work order evaluation factor analysis method based on total log-likelihood algorithm Download PDFInfo
- Publication number
- CN111325475A CN111325475A CN202010142596.4A CN202010142596A CN111325475A CN 111325475 A CN111325475 A CN 111325475A CN 202010142596 A CN202010142596 A CN 202010142596A CN 111325475 A CN111325475 A CN 111325475A
- Authority
- CN
- China
- Prior art keywords
- repair
- evaluation
- user
- sample
- work order
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 230000008439 repair process Effects 0.000 title claims abstract description 72
- 238000011156 evaluation Methods 0.000 title claims abstract description 47
- 238000000034 method Methods 0.000 title claims abstract description 28
- 238000000556 factor analysis Methods 0.000 title claims abstract description 9
- 238000000611 regression analysis Methods 0.000 claims abstract description 6
- 238000004458 analytical method Methods 0.000 claims description 10
- 238000013473 artificial intelligence Methods 0.000 claims description 6
- 238000011160 research Methods 0.000 claims description 6
- 238000006467 substitution reaction Methods 0.000 claims description 6
- 238000007476 Maximum Likelihood Methods 0.000 claims description 4
- 238000012937 correction Methods 0.000 claims description 3
- 239000011159 matrix material Substances 0.000 claims description 3
- 230000017105 transposition Effects 0.000 claims description 3
- 238000010801 machine learning Methods 0.000 abstract description 4
- 238000004519 manufacturing process Methods 0.000 abstract description 2
- 230000006870 function Effects 0.000 description 8
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000012886 linear function Methods 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000007619 statistical method Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
- G06Q10/063—Operations research, analysis or management
- G06Q10/0637—Strategic management or analysis, e.g. setting a goal or target of an organisation; Planning actions based on goals; Analysis or evaluation of effectiveness of goals
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/18—Complex mathematical operations for evaluating statistical data, e.g. average values, frequency distributions, probability functions, regression analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/20—Administration of product repair or maintenance
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/02—Marketing; Price estimation or determination; Fundraising
- G06Q30/0282—Rating or review of business operators or products
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/06—Energy or water supply
Landscapes
- Engineering & Computer Science (AREA)
- Business, Economics & Management (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Human Resources & Organizations (AREA)
- Strategic Management (AREA)
- Economics (AREA)
- Theoretical Computer Science (AREA)
- Entrepreneurship & Innovation (AREA)
- Marketing (AREA)
- General Business, Economics & Management (AREA)
- Development Economics (AREA)
- Tourism & Hospitality (AREA)
- Operations Research (AREA)
- Data Mining & Analysis (AREA)
- Game Theory and Decision Science (AREA)
- Mathematical Physics (AREA)
- Educational Administration (AREA)
- Accounting & Taxation (AREA)
- Finance (AREA)
- Pure & Applied Mathematics (AREA)
- Mathematical Optimization (AREA)
- Quality & Reliability (AREA)
- Mathematical Analysis (AREA)
- Health & Medical Sciences (AREA)
- Computational Mathematics (AREA)
- Water Supply & Treatment (AREA)
- General Health & Medical Sciences (AREA)
- Evolutionary Biology (AREA)
- Life Sciences & Earth Sciences (AREA)
- Public Health (AREA)
- Probability & Statistics with Applications (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Algebra (AREA)
- Bioinformatics & Computational Biology (AREA)
- Databases & Information Systems (AREA)
- Software Systems (AREA)
- General Engineering & Computer Science (AREA)
- Primary Health Care (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
Provided is a rush-repair work order evaluation factor analysis method based on a total log likelihood algorithm. The method relates to the field of power supply, in particular to a rush-repair work order evaluation factor analysis method based on a total log likelihood algorithm. The method for analyzing the evaluation factors of the emergency repair work orders based on the total log likelihood algorithm can analyze and predict user evaluation according to intermediate data in the repair process. The invention adopts the maximum log-likelihood algorithm, can carry out regression analysis on the data of inventory and increment service production repair report through machine learning to obtain the real relationship between the user evaluation of the repair work order and each link data, can calculate the evaluation and predict the evaluation, is convenient to improve the service quality, and further improves the user satisfaction.
Description
Technical Field
The invention relates to the field of power supply, in particular to a rush-repair work order evaluation factor analysis method based on a total log likelihood algorithm.
Background
The power supply service command center mainly relates to a more important closed flow loop about telephone repair, and the flow links are as follows: the method comprises the steps of user telephone repair, failure study and judgment of a virtual Ai commander, path planning analysis, automatic dispatching to an optimal emergency repair team, machine or manual acquisition of user evaluation on current emergency repair or service. The service is used as a service department, the final purpose is to improve the satisfaction degree of users, and the evaluation of the satisfaction degree is over 90 percent.
Although the flow can realize intelligent order receiving, analysis and work order distribution through artificial intelligent voice recognition, program customization development and algorithm design, and a large amount of multidimensional process data and final user evaluation in each link are stored; but the linear relation between various process data in each link and the final user evaluation cannot be known exactly.
The existing evaluation analysis adopts customized formula analysis, one formula is applied to all data for service and repair, a machine learning algorithm is not adopted, evaluation cannot be predicted, only an optimal allocation strategy can be calculated, and a client cannot be guaranteed to have good evaluation on the allocation strategy.
Disclosure of Invention
Aiming at the problems, the invention provides the emergency repair work order evaluation factor analysis method based on the total log likelihood algorithm, which can analyze and predict the user evaluation according to the intermediate data in the repair reporting process.
The technical scheme of the invention is as follows: the method comprises the following steps:
s1, selecting the last year as a sample period, and collecting intermediate data related to user evaluation of each link in the repair flow as a repair sample;
s2, performing regression analysis by adopting a maximum likelihood estimation method to obtain a linear relation between all intermediate data in each repair sample and final evaluation of the user;
and S3, analyzing and predicting user evaluation according to the linear relation.
In step S1, links in the repair reporting flow include user telephone repair reporting, artificial intelligence research and judgment and work order distribution;
the repair variables in the repair report of the user telephone comprise a user mobile phone, a user number, a power failure type, a power failure cell and a power failure address;
the repair variables in the artificial intelligence research and judgment comprise a thoroughly-copied result, a recall result, a dispatch analysis result and a dispatch analysis time;
repair variables in the work order distribution include team information, personnel information, the number of work orders to be handled, arrival time, historical repair quality, repair duration, repair area, repair force configuration, vehicle configuration and repair spare parts.
The step S2 includes the following steps:
s2.1, the relationship between the reported correction variable and the error in each sample is as follows: true user rating-predicted user rating + error
y(i)=θTx(i)+ε(i)
y(i)Represents the true user rating, θ, of the ith sampleTMatrix transposition, x, representing the true value of the ith sample and the prediction evaluation(i)Represents the repair variable, θ, for the ith sampleTx(i)Denotes the predicted value of the ith sample, ∈(i)Representing the error of the ith real value and the ith predicted value;
s2.2, probability Density function substitution
(x-. mu.) represents the error ε in step S2.1(i)Wherein μ and σ are mean and variance, respectively, and substituting them into the formula in step S2.1, the probability density function substitution process for the ith sample is as follows:
p(ε(i))
the deformation is as follows:
the total likelihood represents the likelihood multiplication of each sample, the following formula represents the m rows of samples, and the total likelihood of the i intermediate variables is:
the total likelihood formula is logarithmized as follows:
J (theta) is a total log likelihood algorithm, m rows and i columns of repair sample repair variables are sequentially substituted into the formula to obtain the minimum total log likelihood, hθNamely a linear relation model of the report variables and the final evaluation of the user.
The invention adopts the maximum log-likelihood algorithm, can carry out regression analysis on the data of inventory and increment service production repair report through machine learning to obtain the real relationship between the user evaluation of the repair work order and each link data, can calculate the evaluation and predict the evaluation, is convenient to improve the service quality, and further improves the user satisfaction.
Detailed Description
The invention comprises the following steps:
s1, selecting the last year as a sample period, and collecting intermediate data related to user evaluation of each link in the repair flow as a repair sample;
s2, performing regression analysis by adopting a maximum likelihood estimation method to obtain a linear relation between all intermediate data in each repair sample and final evaluation of the user;
and S3, analyzing and predicting user evaluation according to the linear relation.
In step S1, as shown in table 1, the links in the repair reporting process include user telephone repair reporting, artificial intelligence research and judgment, and work order distribution;
the repair variables in the repair report of the user telephone comprise a user mobile phone, a user number, a power failure type, a power failure cell and a power failure address;
the repair variables in the artificial intelligence research and judgment comprise a thoroughly-copied result, a recall result, a dispatch analysis result and a dispatch analysis time;
repair variables in the work order distribution include team information, personnel information, the number of work orders to be handled, arrival time, historical repair quality, repair duration, repair area, repair force configuration, vehicle configuration and repair spare parts.
TABLE 1
In step S2, the maximum likelihood estimation is a statistical method, which is used to solve the parameters of the related probability density function of the sample set, and it can be obtained from the mathematical proof that the total likelihood of all the repair sample sets is maximum, which indicates that the sample set is more in line with the positive distribution, and the fitting is the most sufficient, and the obtained repair variables and the final user evaluation relationship model are in the most optimal linear function relationship.
According to equivalent substitution: the total likelihood is maximum and is approximately equal to the multiplication of the likelihood of each sample and the maximum time is approximately equal to the multiplication of the probability of each sample and the maximum time is approximately equal to the optimal solution
The step S2 includes the following steps:
s2.1, the relationship between the reported correction variable and the error in each sample is as follows:
y(i)=θTx(i)+ε(i)
y(i)represents the true user rating, θ, of the ith sampleTMatrix transposition, x, representing the true value of the ith sample and the prediction evaluation(i)Represents the repair variable, θ, for the ith sampleTx(i)Denotes the predicted value of the ith sample, ∈(i)Representing the error of the ith real value and the ith predicted value;
s2.2, substituting the probability density function:
(x-. mu.) represents the error ε in step S2.1(i)Wherein μ and σ are mean and variance, respectively, and substituting them into the formula in step S2.1, the probability density function substitution process for the ith sample is as follows:
p(ε(i))
the deformation is as follows:
the total likelihood represents the likelihood multiplication of each sample, the following formula represents the m rows of samples, and the total likelihood of the i intermediate variables is:
because the formula relates to a large number of multiplication iterations, the solving speed of a computer is low, and more resources are occupied; the multiplication iteration operation is changed into addition iteration according to the following formula, so that the solving speed is greatly improved, and the resource occupation is reduced.
The total likelihood formula is logarithmized as follows:
J (theta) is a total log likelihood algorithm, m rows and i columns of repair sample repair variables are sequentially substituted into the formula to obtain the minimum total log likelihood, namely hθThe row vector in the repair reporting variable is the optimal solution;
hθnamely a linear relation model of the report variables and the final evaluation of the user.
Through hθThe function can know the weight of each factor evaluated by the customer of the total log likelihood emergency repair work order and can also be used according to hθBefore the client is not evaluated, the intermediate variables of the substitution process are carried out at any timeAnd (4) predicting an evaluation result, and avoiding low evaluation of a user in the emergency repair command process.
According to the method, regression analysis of data is carried out by means of a machine learning algorithm, a relation model of process data and user evaluation is obtained, intermediate data with large weight is improved in a targeted mode, user evaluation is improved, user evaluation can be predicted in advance, a response can be made in time, and low-grade evaluation is avoided.
Claims (3)
1. A rush-repair work order evaluation factor analysis method based on a total log likelihood algorithm is characterized by comprising the following steps:
s1, selecting the last year as a sample period, and collecting intermediate data related to user evaluation of each link in the repair flow as a repair sample;
s2, performing regression analysis by adopting a maximum likelihood estimation method to obtain a linear relation between all intermediate data in each repair sample and final evaluation of the user;
and S3, analyzing and predicting user evaluation according to the linear relation.
2. The rush-repair work order evaluation factor analysis method based on the logarithm total likelihood algorithm as claimed in claim 1, wherein in step S1, links in the repair reporting process include user telephone repair reporting, artificial intelligence research and judgment, and work order distribution;
the repair variables in the repair report of the user telephone comprise a user mobile phone, a user number, a power failure type, a power failure cell and a power failure address;
the repair variables in the artificial intelligence research and judgment comprise a thoroughly-copied result, a recall result, a dispatch analysis result and a dispatch analysis time;
repair variables in the work order distribution include team information, personnel information, the number of work orders to be handled, arrival time, historical repair quality, repair duration, repair area, repair force configuration, vehicle configuration and repair spare parts.
3. The rush repair work order evaluation factor analysis method based on the log total likelihood algorithm as claimed in claim 1, wherein the step S2 comprises the following steps:
s2.1, the relationship between the reported correction variable and the error in each sample is as follows:
y(i)=θTx(i)+ε(i)
y(i)represents the true user rating, θ, of the ith sampleTMatrix transposition, x, representing the true value of the ith sample and the prediction evaluation(i)Represents the repair variable, θ, for the ith sampleTx(i)Denotes the predicted value of the ith sample, ∈(i)Representing the error of the ith real value and the ith predicted value;
s2.2, substituting the probability density function:
(x-. mu.) represents the error ε in step S2.1(i)Wherein μ and σ are mean and variance, respectively, and substituting them into the formula in step S2.1, the probability density function substitution process for the ith sample is as follows:
p(∈(i))
the deformation is as follows:
the total likelihood represents the likelihood multiplication of each sample, the following formula represents the m rows of samples, and the total likelihood of the i intermediate variables is:
the total likelihood formula is logarithmized as follows:
J (theta) is a total log likelihood algorithm, m rows and i columns of repair sample repair variables are sequentially substituted into the formula to obtain the minimum total log likelihood, hθNamely a linear relation model of the report variables and the final evaluation of the user.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010142596.4A CN111325475A (en) | 2020-03-04 | 2020-03-04 | Emergency repair work order evaluation factor analysis method based on total log-likelihood algorithm |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202010142596.4A CN111325475A (en) | 2020-03-04 | 2020-03-04 | Emergency repair work order evaluation factor analysis method based on total log-likelihood algorithm |
Publications (1)
Publication Number | Publication Date |
---|---|
CN111325475A true CN111325475A (en) | 2020-06-23 |
Family
ID=71171315
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202010142596.4A Pending CN111325475A (en) | 2020-03-04 | 2020-03-04 | Emergency repair work order evaluation factor analysis method based on total log-likelihood algorithm |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN111325475A (en) |
Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104112204A (en) * | 2014-07-01 | 2014-10-22 | 国家电网公司 | Evaluation method for efficient operation of power supply quality |
CN104573312A (en) * | 2014-10-22 | 2015-04-29 | 浙江中烟工业有限责任公司 | Log-based mobile application user satisfaction evaluation method |
CN104680428A (en) * | 2015-03-16 | 2015-06-03 | 朗新科技股份有限公司 | Construction method of power grid customer satisfaction model |
CN106548357A (en) * | 2016-10-27 | 2017-03-29 | 南方电网科学研究院有限责任公司 | The assessment method and system of CSAT |
CN106600455A (en) * | 2016-11-25 | 2017-04-26 | 国网河南省电力公司电力科学研究院 | Electric charge sensitivity assessment method based on logistic regression |
CN107392479A (en) * | 2017-07-27 | 2017-11-24 | 国网河南省电力公司电力科学研究院 | The power customer power failure susceptibility scorecard implementation of logic-based regression model |
CN107766316A (en) * | 2016-08-15 | 2018-03-06 | 株式会社理光 | The analysis method of evaluating data, apparatus and system |
CN109344481A (en) * | 2018-09-21 | 2019-02-15 | 昆明理工大学 | A kind of online service evaluation method based on Plackett-Luce model |
-
2020
- 2020-03-04 CN CN202010142596.4A patent/CN111325475A/en active Pending
Patent Citations (8)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN104112204A (en) * | 2014-07-01 | 2014-10-22 | 国家电网公司 | Evaluation method for efficient operation of power supply quality |
CN104573312A (en) * | 2014-10-22 | 2015-04-29 | 浙江中烟工业有限责任公司 | Log-based mobile application user satisfaction evaluation method |
CN104680428A (en) * | 2015-03-16 | 2015-06-03 | 朗新科技股份有限公司 | Construction method of power grid customer satisfaction model |
CN107766316A (en) * | 2016-08-15 | 2018-03-06 | 株式会社理光 | The analysis method of evaluating data, apparatus and system |
CN106548357A (en) * | 2016-10-27 | 2017-03-29 | 南方电网科学研究院有限责任公司 | The assessment method and system of CSAT |
CN106600455A (en) * | 2016-11-25 | 2017-04-26 | 国网河南省电力公司电力科学研究院 | Electric charge sensitivity assessment method based on logistic regression |
CN107392479A (en) * | 2017-07-27 | 2017-11-24 | 国网河南省电力公司电力科学研究院 | The power customer power failure susceptibility scorecard implementation of logic-based regression model |
CN109344481A (en) * | 2018-09-21 | 2019-02-15 | 昆明理工大学 | A kind of online service evaluation method based on Plackett-Luce model |
Non-Patent Citations (2)
Title |
---|
专注_每天进步一点点: "Al_02_线性回归深入和代码实现_04_应用正太分布概率密度函数_对数总似然", 《HTTPS:// BLOG.CSDN.NET/QQ_32649581/ARTICLE/DETAILS/89881562》 * |
张佳佳等: "新型农村合作医疗满意度影响因素的实证分析——基于河南省某市农村新型合作医疗的调查", 《商业经济》 * |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN109461025B (en) | Electric energy substitution potential customer prediction method based on machine learning | |
CN104933627A (en) | Energy efficiency combination evaluation method of machine tool product manufacture system | |
CN109818775A (en) | Short-term network method for predicting based on adaptive differential evolution algorithm Optimization of Wavelet neural network | |
CN108549907B (en) | Data verification method based on multi-source transfer learning | |
CN116681312B (en) | Ecological-oriented multi-objective reservoir optimal scheduling decision method and system | |
Lei et al. | Learning-based resource allocation: Efficient content delivery enabled by convolutional neural network | |
CN113780684A (en) | Intelligent building user energy consumption behavior prediction method based on LSTM neural network | |
CN105471647A (en) | Power communication network fault positioning method | |
CN110162390A (en) | A kind of method for allocating tasks and system of mist computing system | |
CN113676371A (en) | Network fault detection and diagnosis method based on deep Q learning | |
CN109190786A (en) | A kind of combination forecasting method of photovoltaic efficiency | |
CN116579447A (en) | Time sequence prediction method based on decomposition mechanism and attention mechanism | |
CN115358437A (en) | Power supply load prediction method based on convolutional neural network | |
CN112836876B (en) | Power distribution network line load prediction method based on deep learning | |
CN110956304A (en) | Distributed photovoltaic power generation capacity short-term prediction method based on GA-RBM | |
CN112508254B (en) | Method for determining investment prediction data of transformer substation engineering project | |
CN111325475A (en) | Emergency repair work order evaluation factor analysis method based on total log-likelihood algorithm | |
CN113837486B (en) | RNN-RBM-based distribution network feeder long-term load prediction method | |
CN116485139A (en) | Short-term photovoltaic power generation amount prediction method based on multi-feature fusion | |
CN114707422A (en) | Intelligent power check method based on load prediction | |
CN115423414A (en) | Tool and method for generating distribution scheme of goods space of maintenance equipment warehouse | |
Tanhatalab et al. | Deep ran: A scalable data-driven platform to detect anomalies in live cellular network using recurrent convolutional neural network | |
CN114444760A (en) | Industry user electric quantity prediction method based on mode extraction and error adjustment | |
CN111967689A (en) | Model and method for wind power generation prediction by combining multivariate and stepwise linear regression and artificial neural network | |
CN109816276B (en) | Method, device and equipment for evaluating reliability index of power distribution network |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination | ||
RJ01 | Rejection of invention patent application after publication | ||
RJ01 | Rejection of invention patent application after publication |
Application publication date: 20200623 |