Electric power customer service work order quality sampling inspection sample data extraction method

Abstract

The invention relates to an electric power customer service work order quality sampling inspection sample data extraction method. The extraction method comprises the following steps: setting a time starting point and a time ending point of a quality inspection sample; determining the business type of the quality inspection and the respective extraction proportion; the number of work orders extracted in each customer service staff in real time is xi, and the initial value of xi is 0; forming a work order library to be selected; randomly extracting from a first region work order in the first business type to generate a first extracted sample work order; if yes, judging whether xilt is satisfied; l2, deleting all work orders under the name of the customer service staff from the work order library to be selected, otherwise, continuing to select the work orders in the work order library; and all the work orders in the Mth area of the Nth business type are extracted from the work order library to be extracted. Compared with the prior art, the method has the advantage of ensuring that the extracted work orders of each customer service personnel are relatively balanced.

Description

Technical Field

[0001] The present invention relates to big data processing technology, and in particular to a method for extracting sample data for quality random inspection of electric power customer service work orders. Background Art

[0002] In the field of power customer service, quality inspection business is an important part of operational service work. The target of quality inspection work order extraction is balanced in each province of each business type and the number of work orders of the entire plan is balanced. In view of the centralized management of customer service center personnel, the customer service telephone numbers of the covered provinces are randomly assigned to specific customer service personnel. In addition to the differences in shifts, personal vacation factors, etc., it is impossible for the two to be completely balanced; therefore, the balanced proportion of provinces is given priority. Under the condition of meeting the balanced proportion of provinces, the number of work orders of personnel is relatively balanced. Among them, the balanced proportion of provinces means that the number of work orders extracted from each business and each province in the quality inspection plan accounts for the same proportion of the total number of work orders of the business in the province; the balanced number of work orders of personnel means that all work orders extracted from the quality inspection plan cover all customer service specialists, and the number of work orders accepted / revisited by each customer service specialist is relatively even. The extracted work orders are reviewed or revisited by phone by a dedicated person as an important means of evaluating the work quality of customer service personnel.

[0003] At present, most quality inspection work relies on manual work by quality inspectors, which is a heavy task. Therefore, it is urgent to build online quality inspection tools to improve the efficiency of quality inspection work and provide a data basis for quality inspection business analysis. Summary of the invention

[0004] The present invention proposes an automated method for extracting sample data for quality inspection of power customer service work orders. The specific technical solution is as follows:

[0005] One of the methods for extracting sample data for quality inspection of power customer service work orders includes the following process:

[0006] S100: Setting the start and end time of the quality inspection sample;

[0007] S200: Determine the business types and their respective extraction ratios for this quality inspection, a total of N business types; including: extraction ratio 1 for the first business type, extraction ratio j for the i-th business type, ..., extraction ratio m for the N-th business type;

[0008] Determine the total number of work order samples D0, the total number of customer service personnel involved in this quality inspection K0, the average number of work orders selected by each customer service personnel is L0, L0 = D0 / K0, round L0 to get L1, define L2 as L1+1, and define L3 as L1+2; the real-time number of work orders selected by each customer service personnel is x i , x i The initial value of is 0;

[0009] S300: Divide the total number of samples for each business type by region. The sampling ratio of each business sample in each region is the same as the corresponding business type ratio in S200. There are a total of Q regions, namely: the 1st region, the qth region,..., the Qth region.

[0010] Form a standby selection work order library. The first-level sorting of the work order library: the 1st business type, the 2nd business type, and so on from top to bottom until the Nth business type. Within each business type, the second-level sorting: the 1st region, the 2nd region, and so on from top to bottom until the Qth region. Within each region, the work orders are arranged in chronological order of generation.

[0011] S400: Randomly select from the work orders of the 1st region in the 1st business type to generate the 1st selected sample work order.

[0012] S500: Corresponding to the identity information of the customer service staff of this work order, increment the number of selected work orders corresponding to this customer service staff by 1, that is: x i = x i + 1;

[0013] S600: Determine whether x i < L2 is satisfied. If satisfied, go to S800; otherwise, go to S700.

[0014] S700: Delete all work orders under this customer service staff from the standby selection work order library, and then go to S800.

[0015] S800: Conduct the second work order extraction in the 1st region of the 1st business type. If the number of extracted work orders in the 1st region meets the set requirements, start with the 2nd region and proceed downwards in sequence. If the number of extracted work orders in the 1st business type meets the set requirements, start with the 2nd business type and proceed downwards in sequence. After each work order is extracted, return to S500 until all work orders in the Mth region of the Nth business type are extracted from the standby selection work order library.

[0016] Preferably, in S500, add the total number of samples x T-1 extracted from this customer service staff during the previous round of sampling, calculate the average number of samples L T-1 extracted by the previous-round customer service staff; add it to the average number of samples L0 of the current-round customer service staff to get L T0 ; perform an integer operation on L T0 to get L T1 , define L T2 as L T1 + 1, define L T3 as L T1 + 2;

[0017] In S600, determine whether x i + xT-1 <L T2 , if it meets the requirement to transfer to S800, otherwise transfer to S700.

[0018] The second method for extracting sample data of power customer service work order quality inspection includes the following process:

[0019] S100: Set the start time point and end time point of the quality inspection samples;

[0020] S200: Determine the business types of this quality inspection and their respective extraction ratios. There are a total of N business types, including: the extraction ratio 1 of the first business type, the extraction ratio j of the i-th business type,..., the extraction ratio m of the N-th business type;

[0021] Determine the total number of work order samples D0, the total number of customer service personnel K0 involved in this quality inspection, and the average number of work orders randomly selected for each customer service personnel is L0, L0 = D0 / K0. Perform a rounding operation on L0 to obtain L1, define L2 as L1 + 1, and define L3 as L1 + 2; the number of work orders randomly selected for each customer service personnel in real time is x i , x i The initial value of is 0;

[0022] S300: Divide the total sample volume of each business type by region. The extraction ratio of each business sample in each region is the same as the corresponding business type ratio in S200; there are a total of Q regions, namely: the first region, the q-th region,..., the Q-th region;

[0023] Form a standby work order library for random selection. The first-level sorting of the work order library: the first business type, the second business type, and so on from top to bottom until the N-th business type; within each business type, the second-level sorting: the first region, the second region, and so on from top to bottom until the Q-th region; within each region, the work orders are arranged in the order of generation time;

[0024] S400: Randomly select from the work orders of the first region in the first business type to generate the first selected sample work order;

[0025] S500: Corresponding to the identity information of the customer service personnel of this work order, add 1 to the number of selected work orders corresponding to this customer service personnel, that is: x i = x i + 1;

[0026] S600: Judge whether it meets x i <L3 at this time. If it meets the requirement, transfer to S800, otherwise transfer to S700;

[0027] S700: Delete all the work orders under this customer service personnel from the standby work order library for random selection, and then transfer to S800;

[0028] S800: Extract the second work order in the first region of the first business type; if the number of work orders extracted in the first region meets the set requirements, start with the second region and proceed downward in sequence; if the number of work orders extracted in the first business type meets the set requirements, start with the second business type and proceed downward in sequence; return to S500 after extracting each work order; until all work orders in the Mth region of the Nth business type are extracted from the standby work order library.

[0029] Preferably, in S500, the total number of samples drawn by each customer service staff in the previous round of sampling is increased by x T-1 , calculate the average total number of samples L drawn by customer service staff in the previous round T-1 ; Add it to the average number of samples L0 drawn by customer service staff in this round to get L T0 ; for L T0 Rounding operation is performed to obtain L T1 , L T2 Defined as L T1 +1, L T3 Defined as L T1 +2;

[0030] In S600, it is determined whether x is satisfied at this time. i +x T-1 <L T3 If the condition is met, go to S800; otherwise, go to S700.

[0031] The advantages of the present invention over the prior art are:

[0032] The pending work orders in the pending work order pool are stratified according to provinces and businesses, and independent sampling is performed in each stratum. At the same time, customer service specialists are used as the statistical dimension, and the number of drawn work orders in the real-time counting table of drawn work orders is referred to to sort the personnel by priority. Using the circular queue method, the work orders of customer service specialists with high priority levels are preferentially drawn. The number of work orders to be drawn in the provincial statistical table is used as the work order drawing quantity to realize the rotation of personnel, draw the work orders of the corresponding personnel, and ensure that the work orders drawn for each customer service specialist are relatively balanced. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 It is a flow chart of a method for extracting sample data for quality random inspection of electric customer service work orders according to the present invention. DETAILED DESCRIPTION

[0034] Example 1

[0035] The method for extracting sample data for quality inspection of power customer service work orders includes the following processes:

[0036] S100: Setting the start and end time of the quality inspection sample;

[0037] S200: Determine the business types of this quality inspection and their respective extraction ratios. There are a total of N business types, including: the extraction ratio 1 for the first business type, the extraction ratio j for the i-th business type, ……, the extraction ratio m for the N-th business type.

[0038] Determine the total number of master work order samples D0, the total number of customer service staff K0 involved in this quality inspection, and the average number of work orders randomly selected for each customer service staff L0, where L0 = D0 / K0. Perform rounding operation on L0 to obtain L1, define L2 as L1 + 1, and define L3 as L1 + 2. The number of work orders randomly selected for each customer service staff in real time is x i , x i The initial value of which is 0.

[0039] S300: Divide the total sample quantity of each business type by region. The extraction ratio of each business sample in each region is the same as the corresponding business type ratio in S200. There are a total of Q regions, namely: the first region, the q-th region, ……, the Q-th region.

[0040] Form a standby work order library for random selection. The first-level sorting of the work order library: the first business type, the second business type, and so on from top to bottom until the N-th business type; within each business type, the second-level sorting: the first region, the second region, and so on from top to bottom until the Q-th region; within each region, the work orders are arranged in the order of generation time.

[0041] S400: Randomly select from the work orders of the first region in the first business type to generate the first selected sample work order.

[0042] S500: Identify the identity information of the customer service staff corresponding to this work order, and increment the number of selected work orders corresponding to this customer service staff by 1, that is: x i = x i + 1;

[0043] S600: Judge whether x i < L2 is satisfied. If satisfied, go to S800; otherwise, go to S700.

[0044] S700: Delete all the work orders under this customer service staff from the standby work order library for random selection, and then go to S800.

[0045] S800: Conduct the second work order extraction for the first region in the first business type; if the number of work orders extracted from the first region meets the set requirements, start the second region and proceed downward in sequence; if the number of work orders extracted from the first business type meets the set requirements, start the second business type and proceed downward in sequence; after each work order extraction, return to S500; until all the work orders of the M-th region in the N-th business type are extracted from the standby work order library for random selection.

[0046] In this embodiment, the deviation value of the employee in the single sample extraction process is controlled within 1.

[0047] Embodiment 2

[0048] In this embodiment, S100 - S400, S700, and S800 are exactly the same as those in Embodiment 1.

[0049] S500: Define the total number of samples x selected from each customer service staff in the previous round of sampling inspection. T-1 , calculate the average total number of samples L selected by the customer service staff in the previous round. T-1 ; Add it to the average number of samples L0 selected by the customer service staff in this round to get L. T0 ; Perform an integer operation on L. T0 to get L. T1 , take L. T2 and define it as L. T1 + 1, take L. T3 and define it as L. T1 + 2; Corresponding to the identity information of the customer service staff of this work order, add 1 to the number of selected work orders corresponding to this customer service staff, that is: x. i = x. i + 1;

[0050] S600: Judge whether x i + x T-1 < L T2 is satisfied at this time. If satisfied, go to S800; otherwise, go to S700.

[0051] In this embodiment, the deviation value of the employee in the cumulative sample extraction times is controlled within 1. If a certain customer service staff was not selected in the previous sampling inspection, then, in this sampling inspection, he will be selected multiple times in the selected pool, and the cumulative selection amount will be kept consistent with others to control, ensuring a greater range of balance.

[0052] Embodiment 3

[0053] In this embodiment, S100 - S500, S700, and S800 are exactly the same as those in Embodiment 1.

[0054] S600: Judge whether x i < L3 is satisfied at this time. If satisfied, go to S800; otherwise, go to S700.

[0055] In this embodiment, the deviation value of the employee in the single sample extraction process is controlled within 2.

[0056] Embodiment 4

[0057] In this embodiment, S100 - S400, S700, and S800 are exactly the same as those in Embodiment 1.

[0058] S500: Defines the total number of samples x drawn by each customer service staff in the previous round of sampling T-1 , calculate the average total number of samples L drawn by customer service staff in the previous round T-1 ; Add it to the average number of samples L0 drawn by customer service staff in this round to get L T0 ; for L T0 Rounding operation is performed to obtain L T1 , L T2 Defined as L T1 +1, L T3 Defined as L T1 +2;

[0059] S600: Determine whether x is satisfied at this time i +x T-1 <L T3 If the condition is met, go to S800; otherwise, go to S700.

[0060] In this embodiment, the deviation value of the cumulative number of sample draws by employees is controlled within 2. If a customer service staff member was not selected in the last random inspection, then in this random inspection, he will be in the sample pool multiple times, and the cumulative number of draws will be controlled to be consistent with that of others to ensure a wider range of balance.

Claims

1. A method for extracting sample data for quality inspection of power customer service work orders, characterized in that: The process includes the following: S100: Setting the start and end time of the quality inspection sample; S200: Determine the business types and their respective extraction ratios for this quality inspection, a total of N business types; including: extraction ratio 1 for the first business type, extraction ratio j for the i-th business type, ..., extraction ratio m for the N-th business type; Determine the total number of work order samples D0, the total number of customer service personnel involved in this quality inspection K0, the average number of work orders selected by each customer service personnel is L0, L0 = D0 / K0, round L0 to get L1, define L2 as L1+1, and define L3 as L1+2; the real-time number of work orders selected by each customer service personnel is x i , x i The initial value of is 0; S300: Divide the total amount of samples of each business type by region, and the extraction ratio of each business sample in each region is the same as the corresponding business type ratio in S200; there are Q regions in total, namely: the first region, the qth region, ..., the Qth region; A work order database for selection is formed. The first-level sorting of the work order database is: the first business type, the second business type, and so on from top to bottom until the Nth business type. Within each business type, the second-level sorting is: the first region, the second region, and so on from top to bottom until the Qth region. Within each region, the work orders are arranged in order of generation time. S400: Randomly select from the first regional work order in the first business type to generate the first sample work order; S500: Corresponding to the customer service staff identity information of the work order, the number of selected work orders corresponding to the customer service staff is increased by 1, that is: x i =x i +1; S600: Determine whether x is satisfied at this time i <L2. If satisfied, go to S800; otherwise, go to S700; S700: Delete all work orders under the customer service staff from the work order database for selection, and then go to S800; S800: Extract the second work order in the first region of the first business type; if the number of work orders extracted in the first region meets the set requirements, start with the second region and proceed downward in sequence; if the number of work orders extracted in the first business type meets the set requirements, start with the second business type and proceed downward in sequence; return to S500 after extracting each work order; until all work orders in the Mth region of the Nth business type are extracted from the standby work order library.

2. According to the method for extracting sample data for quality inspection of electric customer service work orders according to claim 1, it is characterized in that: In S500, increase the total number of samples taken by each customer service staff in the previous round of sampling x T-1 , calculate the average total number of samples L drawn by customer service staff in the previous round T-1 ; Add it to the average number of samples L0 drawn by customer service staff in this round to get L T0 ; for L T0 Rounding operation is performed to obtain L T1 , L T2 Defined as L T1 +1, L T3 Defined as L T1 +2; In S600, it is determined whether x is satisfied at this time. i +x T-1 <L T2 If the condition is met, go to S800; otherwise, go to S700.

3. A method for extracting sample data for quality inspection of power customer service work orders, characterized in that: The process includes the following: S100: Setting the start and end time of the quality inspection sample; S200: Determine the business types and their respective extraction ratios for this quality inspection, a total of N business types; including: extraction ratio 1 for the first business type, extraction ratio j for the i-th business type, ..., extraction ratio m for the N-th business type; Determine the total number of work order samples D0, the total number of customer service personnel involved in this quality inspection K0, the average number of work orders selected by each customer service personnel is L0, L0 = D0 / K0, round L0 to get L1, define L2 as L1+1, and define L3 as L1+2; the real-time number of work orders selected by each customer service personnel is x i , x i The initial value of is 0; S300: Divide the total amount of samples of each business type by region, and the extraction ratio of each business sample in each region is the same as the corresponding business type ratio in S200; there are Q regions in total, namely: the first region, the qth region, ..., the Qth region; A work order database for selection is formed. The first-level sorting of the work order database is: the first business type, the second business type, and so on from top to bottom until the Nth business type. Within each business type, the second-level sorting is: the first region, the second region, and so on from top to bottom until the Qth region. Within each region, the work orders are arranged in order of generation time. S400: Randomly select from the first regional work order in the first business type to generate the first sample work order; S500: Corresponding to the customer service staff identity information of the work order, the number of selected work orders corresponding to the customer service staff is increased by 1, that is: x i =x i +1; S600: Determine whether x is satisfied at this time i <L3, if satisfied, go to S800, otherwise go to S700; S700: Delete all work orders under the customer service staff from the work order database for selection, and then go to S800; S800: Extract the second work order in the first region of the first business type; if the number of work orders extracted in the first region meets the set requirements, start with the second region and proceed downward in sequence; if the number of work orders extracted in the first business type meets the set requirements, start with the second business type and proceed downward in sequence; return to S500 after extracting each work order; until all work orders in the Mth region of the Nth business type are extracted from the standby work order library.

4. According to the method for extracting sample data for quality inspection of electric customer service work orders according to claim 3, it is characterized in that: In S500, increase the total number of samples taken by each customer service staff in the previous round of sampling x T-1 , calculate the average total number of samples L drawn by customer service staff in the previous round T-1 ; Add it to the average number of samples L0 drawn by customer service staff in this round to get L T0 ; for L T0 Rounding operation is performed to obtain L T1 , L T2 Defined as L T1 +1, L T3 Defined as L T1 +2; In S600, it is determined whether x is satisfied at this time. i +x T-1 <L T3 If the condition is met, go to S800; otherwise, go to S700.

Images