WO2019019633A1 - 基于业务线的预测方法、装置、存储介质及终端 - Google Patents

基于业务线的预测方法、装置、存储介质及终端 Download PDF

Info

Publication number
WO2019019633A1
WO2019019633A1 PCT/CN2018/077347 CN2018077347W WO2019019633A1 WO 2019019633 A1 WO2019019633 A1 WO 2019019633A1 CN 2018077347 W CN2018077347 W CN 2018077347W WO 2019019633 A1 WO2019019633 A1 WO 2019019633A1
Authority
WO
WIPO (PCT)
Prior art keywords
data
service line
prediction
specified
difference
Prior art date
Application number
PCT/CN2018/077347
Other languages
English (en)
French (fr)
Inventor
万晓辉
Original Assignee
平安科技（深圳）有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 平安科技（深圳）有限公司 filed Critical 平安科技（深圳）有限公司
Priority to US16/093,628 priority Critical patent/US20210224434A1/en
Priority to SG11201808507VA priority patent/SG11201808507VA/en
Publication of WO2019019633A1 publication Critical patent/WO2019019633A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M3/00—Automatic or semi-automatic exchanges
    • H04M3/42—Systems providing special services or facilities to subscribers
    • H04M3/50—Centralised arrangements for answering calls; Centralised arrangements for recording messages for absent or busy subscribers ; Centralised arrangements for recording messages
    • H04M3/51—Centralised call answering arrangements requiring operator intervention, e.g. call or contact centers for telemarketing
    • 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
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06315—Needs-based resource requirements planning or analysis
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
    • G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
    • G06F16/28—Databases characterised by their database models, e.g. relational or object models
    • G06F16/283—Multi-dimensional databases or data warehouses, e.g. MOLAP or ROLAP
    • 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
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06311—Scheduling, planning or task assignment for a person or group
    • 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
    • G06Q10/0631—Resource planning, allocation, distributing or scheduling for enterprises or organisations
    • G06Q10/06311—Scheduling, planning or task assignment for a person or group
    • G06Q10/063112—Skill-based matching of a person or a group to a task
    • 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
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G06Q10/06398—Performance of employee with respect to a job function
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/14—Network analysis or design
    • H04L41/142—Network analysis or design using statistical or mathematical methods
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M3/00—Automatic or semi-automatic exchanges
    • H04M3/42—Systems providing special services or facilities to subscribers
    • H04M3/50—Centralised arrangements for answering calls; Centralised arrangements for recording messages for absent or busy subscribers ; Centralised arrangements for recording messages
    • H04M3/51—Centralised call answering arrangements requiring operator intervention, e.g. call or contact centers for telemarketing
    • H04M3/523—Centralised call answering arrangements requiring operator intervention, e.g. call or contact centers for telemarketing with call distribution or queueing
    • H04M3/5232—Call distribution algorithms
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F2111/00—Details relating to CAD techniques
    • G06F2111/08—Probabilistic or stochastic CAD
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F30/00—Computer-aided design [CAD]
    • G06F30/20—Design optimisation, verification or simulation
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04M—TELEPHONIC COMMUNICATION
    • H04M2203/00—Aspects of automatic or semi-automatic exchanges
    • H04M2203/40—Aspects of automatic or semi-automatic exchanges related to call centers
    • H04M2203/402—Agent or workforce management

Definitions

  • the present application belongs to the field of communications technologies, and in particular, to a service line-based prediction method, apparatus, storage medium, and terminal.
  • the prior art mainly uses the prediction modes such as the time series prediction method and the regression prediction model to perform the scheduling prediction.
  • the time series forecasting method is a historical data extension prediction, which is extended by the development process and law reflected by the time series to predict the development trend;
  • the regression prediction model analyzes the between the independent variables and the dependent variables in the market. Correlation, a regression equation between variables is established, and the regression equation is used as a prediction model.
  • the outgoing service focuses on the number of customers called, and relates to the connection rate of the customer list and the customer's desire to purchase the product; while the incoming service focuses on the duration of the call and the number of calls, different types of services involve different call durations. , the number of calls.
  • the prior art adopts a unified prediction mode for different business scenarios, and the accuracy of scheduling prediction is not high, and it is difficult to meet the increasingly complex demands of the market.
  • An embodiment of the present application provides a service line-based prediction method, apparatus, storage medium, and terminal, where the prediction method includes:
  • the prediction model is divided into an incoming prediction model and an outgoing prediction model according to the service type, and the data warehouse is composed of call data and dialing list data in a preset historical time after the cleaning process.
  • the prediction method further includes:
  • the predicted value of the output dimension is smoothed according to the weekly date of the marketing activity and the emergency to eliminate interference of the marketing activity and the emergency event with the predicted value.
  • the predicting value of the output dimension is smoothed according to the weekly date of the marketing activity and the emergency event to eliminate the interference of the marketing activity and the emergency event on the predicted value, including:
  • the calculating, according to the predicted value, the total number of tasks and the required manpower of the specified service line in a specified time period includes:
  • the standard working time is obtained, the average working time is calculated according to the standard working time and the conversion rate, and the quotient between the total amount of the task and the average working time is determined as the required manpower.
  • the embodiment of the present application further provides a service line based prediction apparatus, where the prediction apparatus includes:
  • a first obtaining module configured to acquire a prediction model corresponding to the specified service line, and an input dimension and an output dimension of the current prediction when performing service prediction on the specified service line;
  • a second obtaining module configured to acquire, from the data warehouse, prediction data that meets the input dimension
  • An analysis module configured to perform trend analysis on the prediction data by using the Monte Carlo simulation method and the geometric Brownian motion, to obtain a predicted value of the output dimension
  • a calculation module configured to calculate, according to the predicted value, a total amount of tasks and a required amount of manpower of the specified service line in a specified time period
  • the prediction model is divided into an incoming prediction model and an outgoing prediction model according to the service type, and the data warehouse is composed of call data and dialing list data in a preset historical time after the cleaning process.
  • the device further includes:
  • a third obtaining module configured to acquire a marketing activity and an emergency event in the preset historical time after obtaining the predicted value of the output dimension, and determine a weekly date when the marketing activity and an emergency event occur;
  • a smoothing processing module configured to perform smoothing processing on the predicted value of the output dimension according to the weekly date of the marketing activity and the emergency event, to eliminate the interference of the marketing activity and the emergency event on the predicted value.
  • the smoothing processing module includes:
  • a screening unit configured to traverse each output dimension, and select a predicted value having the same week date as the base data from the predicted value of the output dimension;
  • a statistical processing unit configured to calculate an average value and a standard deviation of the base data
  • a comparing unit configured to calculate a difference between each of the base data and the average value, and compare an absolute value of the difference with the standard deviation
  • a smoothing processing unit configured to: when the absolute value of the difference is greater than the standard deviation, reduce the base data corresponding to the difference if the difference is a positive number, and increase if the difference is a negative number The difference corresponds to the base data.
  • calculation module includes:
  • a total amount calculation unit configured to sum the predicted values of the specified service line in a specified time period, and obtain a total amount of tasks of the specified service line in a specified time period;
  • the conversion rate calculation unit is configured to acquire the call duration and the attendance data of the plurality of agents, calculate the work efficiency of each agent according to the call duration and the attendance data, and obtain an average value of the work efficiency to obtain a conversion rate;
  • the human computing unit is configured to obtain a standard working time, calculate an average working time according to the standard working time and a conversion rate, and obtain a quotient between the total amount of the task and the average working time as the required amount of labor.
  • the embodiment of the present application further provides a computer readable storage medium, where the computer readable instructions are stored, and when the computer readable instructions are executed by the processor, the following steps are implemented:
  • the prediction model is divided into an incoming prediction model and an outgoing prediction model according to the service type, and the data warehouse is composed of call data and dialing list data in a preset historical time after the cleaning process.
  • the embodiment of the present application further provides a terminal, including a memory, a processor, and computer readable instructions stored on the memory and executable on the processor, and the processor implements the following steps when the computer readable instructions are executed:
  • the prediction model is divided into an incoming prediction model and an outgoing prediction model according to the service type, and the data warehouse is composed of call data and dialing list data in a preset historical time after the cleaning process.
  • the embodiment of the present application constructs different prediction models according to different service types, including an incoming prediction model and an outgoing prediction model.
  • the corresponding service line is obtained.
  • the trend analysis obtains the predicted value of the output dimension.
  • the total number of tasks and the required labor input of the specified service line in the specified time period are calculated according to the predicted value; thereby implementing the adoption for different business scenarios.
  • Different prediction modes improve the accuracy of forecasting different lines of business.
  • FIG. 1 is a first implementation flowchart of a service line-based prediction method provided by an embodiment of the present application
  • step S104 is a flowchart of an implementation of step S104 in a first implementation flow of a service line-based prediction method according to an embodiment of the present application
  • FIG. 3 is a second implementation flowchart of a service line-based prediction method provided by an embodiment of the present application.
  • step S305 is a flowchart of an implementation of step S305 in a second implementation flow of a service line-based prediction method according to an embodiment of the present application;
  • FIG. 5 is a structural structural diagram of a service line-based predicting apparatus according to an embodiment of the present application.
  • FIG. 6 is a schematic diagram of a terminal provided by an embodiment of the present application.
  • the prediction of the service line is a prediction of the workload and manpower of the line of business to prepare for the scheduling prediction.
  • the service line-based prediction method in the embodiment of the present application may be applied to a terminal, where the terminal includes but is not limited to a computer, a server, and a notebook computer.
  • FIG. 1 shows a first implementation flow of a service line based prediction method provided by an embodiment of the present application.
  • the service line based prediction method includes:
  • step S101 when performing service prediction on the specified service line, the prediction model corresponding to the specified service line, and the input dimension and output dimension of the current prediction are obtained.
  • the embodiment of the present application establishes a corresponding prediction model according to different service types, including an incoming prediction model and an outgoing prediction model.
  • the call prediction model is used to predict the total amount of tasks and the manpower arrangement for the incoming call type service
  • the call out prediction model is used to predict the total amount of tasks and the manpower arrangement for the outgoing call type service.
  • both the incoming prediction model and the outgoing prediction model use Monte Carlo simulation method and geometric Brownian motion as a stochastic model, and extract call data and call from a business history database in a preset historical time.
  • the list data is processed and analyzed, and then imported into an analytical database to construct a data warehouse corresponding to the incoming prediction model and the outgoing prediction model.
  • the preset historical time is preferably within a past year.
  • the incoming prediction model and the outgoing prediction model are respectively configured with corresponding input dimensions and output dimensions for the user to select.
  • the input dimension is a parameter type input to the incoming prediction model or the outgoing prediction model.
  • the output dimension is a parameter type that is output after the prediction data corresponding to the input dimension is processed by the call-in prediction model or the call-out prediction model.
  • the input dimensions include, but are not limited to, call duration, agent work utilization, call loss rate, satisfaction, agent skill level; output dimensions include, but are not limited to, call duration, number of calls, agent utilization, Call loss rate.
  • the input dimensions include, but are not limited to, the number of issued lists, the connection rate, the average call duration, the agent work utilization rate, and the agent skill level;
  • the output dimensions include, but are not limited to, the number of lists, the number of calls, and the duration of the call. , connection rate, agent utilization.
  • the embodiments of the present application implement different prediction modes for different service scenarios by establishing different prediction models and configuring different types of input dimensions and output dimensions for different prediction models. And selecting a more suitable input dimension and output dimension is beneficial to improve the accuracy of prediction for different service lines, and it is convenient for the user to select and adjust the prediction model corresponding to the specified service line and its input parameters and output parameters.
  • the user Before predicting the service line, the user may select a prediction model corresponding to the service line on the terminal in advance, and select an input dimension and an output dimension for prediction.
  • the terminal receives the scheduling prediction instruction and performs workload prediction on the specified service line, the terminal acquires the corresponding prediction model according to the current specified service line, and the input dimension and output dimension of the current prediction.
  • step S102 prediction data that satisfies the input dimension is acquired from the data warehouse.
  • the prediction model creates a data warehouse with call data and dialed list data in a preset historical time after cleaning.
  • embodiments of the present application filter out prediction data from a data warehouse based on the selected input dimensions.
  • the incoming prediction model its input dimensions include, but are not limited to, call duration, agent work utilization, call loss rate, satisfaction, agent skill level. If the selected input dimension for prediction includes three parameters: call duration, agent work utilization, and call loss rate, data that satisfies the above three input dimensions is selected from the data warehouse as prediction data.
  • step S103 the prediction data is subjected to trend analysis by using the Monte Carlo simulation method and the geometric Brownian motion to obtain a predicted value of the output dimension.
  • step S104 the total amount of tasks and the number of manpower required for the specified service line in the specified time period are calculated according to the predicted value.
  • the workload of the specified service line in the specified time period is calculated based on the predicted value.
  • the specified time period is smaller than the time span of the call data and the dialing list data in the data warehouse.
  • FIG. 2 shows a specific implementation process of step S104 in the service line-based prediction method provided by the embodiment of the present application.
  • the step S104 includes:
  • step S201 the predicted values of the specified service line in the specified time period are summed to obtain the total number of tasks of the specified service line in the specified time period.
  • the embodiment of the present application first calculates the total amount of tasks of the specified service line within a specified time period according to the predicted value obtained by the prediction model.
  • the total amount of tasks is represented by time (minutes). Assume that this prediction is an incoming prediction model.
  • the selected output dimension is the duration of the call.
  • the specified time period is from June 25 to June 29 for five days. After obtaining the predicted value of the call duration, it is obtained at 6 The sum of the predicted values of five days from the 25th of June to the 29th of June, thereby obtaining the total amount of tasks of the specified line of business within the specified time period.
  • step S202 the call duration and the attendance data of the plurality of agents are obtained, and the work efficiency of each agent is calculated according to the call duration and the attendance data, and the average of the work efficiency is obtained to obtain the conversion rate.
  • the embodiment of the present application obtains the call duration and attendance data of a number of agents; the duration of the call is the total duration after all the calls of a single agent in one day. Then, according to the duration of the call and the attendance data, the work efficiency of each agent is calculated, and the average value of the work efficiency is obtained to obtain a conversion rate.
  • the conversion rate is the average probability of effective utilization of the standard working time.
  • step S203 a standard working time length is acquired, an average working time length is calculated according to the standard working time length and a conversion rate, and a quotient between the total task amount and the average working time length is obtained as the required labor amount.
  • the product between the standard working time and the conversion rate is obtained to obtain the average working time of the agent.
  • This average working time reflects the daily working hours of a single agent.
  • the quotient between the total amount of the task and the average working time is obtained.
  • the quotient is used as the quantity of labor required to be input, thereby realizing the manpower prediction based on the service line, and the subsequent scheduling is performed. Expand according to the number of humans.
  • the average working time is calculated according to the actual duration of the call and the attendance data, thereby effectively improving the fit of the manual prediction.
  • the marketing activity, the system abnormality, the equipment failure, and the like may cause the historical data to generate large fluctuations, and the fluctuations may affect the predicted value of the scheduling. Therefore, the embodiment of the present application further includes the predicted value obtained by using the prediction model. Smoothing is performed.
  • the line-based prediction method includes:
  • Step S301 to step S303 wherein the steps S301 to S303 are the same as the steps S101 to S103 described in the embodiment of the present invention.
  • steps S301 to S303 are the same as the steps S101 to S103 described in the embodiment of the present invention.
  • steps S301 to S303 are the same as the steps S101 to S103 described in the embodiment of the present invention.
  • the prediction method further includes:
  • step S304 the marketing activities and emergencies in the preset historical time are acquired, and the weekdays when the marketing activities and emergencies occur are determined.
  • the emergency event includes, but is not limited to, a system abnormality, a device failure, and the like.
  • the preset historical time is a time span of call data and dialing list data in the data warehouse.
  • the embodiment of the present application obtains the date of occurrence of marketing activities and emergencies within the time span.
  • the occurrence date is a week date, which is a date in seven days of the week, such as Monday, Tuesday, Wednesday, Thursday, Friday, Saturday, and Sunday.
  • step S305 the predicted value of the output dimension is smoothed according to the weekly date of the marketing activity and the emergency event to eliminate interference of the marketing activity and the emergency event with the predicted value.
  • FIG. 4 shows a specific implementation process of step S305 in the second implementation process of the service line-based prediction method provided by the embodiment of the present application.
  • the step S305 includes:
  • step S401 each output dimension is traversed, and predicted values having the same week date are selected from the predicted values of the output dimensions as base data.
  • a predicted value having the same week date is filtered from the predicted values of the output dimension.
  • the weekly date of the marketing activity is Tuesday
  • the predicted value of the output dimension at each Tuesday is filtered, and the filtered predicted value is used as the base data.
  • step S402 an average value and a standard deviation of the base data are calculated.
  • the standard deviation of the base data is used as a criterion for judging whether or not the base data is smoothed.
  • step S403 a difference between each of the base data and the average value is calculated, and an absolute value of the difference is compared with the standard deviation.
  • step S402 After obtaining the average and standard deviation of the predicted values of the output dimension on Tuesday, calculating the difference between the predicted value of the output dimension on Tuesday and the average value, as described above The absolute value of the difference is compared with the standard deviation calculated in step S402 to determine whether to correct the predicted value.
  • step S404 when the absolute value of the difference is greater than the standard deviation, if the difference is a positive number, the base data corresponding to the difference is reduced, and if the difference is a negative number, the base is increased. The base data corresponding to the difference.
  • the predicted value whose error is not within the standard deviation is abnormal data, that is, the predicted value is corrected when the absolute value of the difference is greater than the standard deviation, including: determining the Positive/negative difference, if the difference is positive, indicating that the base data corresponding to the difference is larger, the corresponding base data is reduced, and if the difference is negative, the base data corresponding to the difference is small Then, the corresponding base data is increased, thereby completing the smoothing process of the predicted value of the output dimension.
  • the base data during the marketing campaign is usually significantly higher or larger than the average value, and therefore, the difference between the base data and the average value during the marketing campaign is a positive number and the difference is greater than the standard deviation, At this time, the base data during the marketing campaign is reduced, so that the base data during the marketing campaign converges to the regular historical time rule, thereby eliminating the fluctuation interference of the marketing activity on the predicted value.
  • the base data during the occurrence of an emergency is typically significantly lower or smaller than the average, so the difference between the base data and the average during the emergency is negative and the absolute value of the difference is greater than the standard Poor, at this time, the base data during the emergency event is increased, so that the base data during the emergency event converges to the regular time in the preset historical time, thereby eliminating the fluctuation interference of the unexpected event on the predicted value.
  • step S306 the total amount of tasks and the number of manpower required for the specified service line in the specified time period are calculated according to the predicted value.
  • the smoothed processed prediction value removes the fluctuation interference caused by the marketing activity and the emergent event, predicting the total amount of the task and the manpower of the specified service line based on the predicted value after the smoothing process can effectively improve the accuracy of the prediction result. And the degree of adaptation.
  • the size of the serial number of each step does not mean the order of execution order, and the order of execution of each step should be determined by its function and internal logic, and should not constitute any implementation process of the embodiment of the present application. limited.
  • the storage medium may be a read only memory, a magnetic disk or an optical disk, or the like.
  • FIG. 5 is a structural diagram of a service line-based prediction apparatus provided by an embodiment of the present application. For convenience of description, only parts related to the embodiment of the present application are shown.
  • the service line-based prediction apparatus is used to implement the service line-based prediction method described in the foregoing embodiments of FIG. 1 to FIG. 4, and may be a software unit, a hardware unit, or a soft A unit that combines hardware.
  • the terminal includes, but is not limited to, a computer, a server, a notebook computer.
  • the traffic line based prediction apparatus includes:
  • the first obtaining module 51 is configured to acquire, when performing service prediction on the specified service line, a prediction model corresponding to the specified service line, and an input dimension and an output dimension of the current prediction;
  • a second obtaining module 52 configured to acquire, from the data warehouse, prediction data that meets the input dimension
  • An analysis module 53 is configured to perform trend analysis on the predicted data by using the Monte Carlo simulation method and the geometric Brownian motion to obtain a predicted value of the output dimension;
  • the calculating module 54 is configured to calculate, according to the predicted value, a total amount of tasks and a required amount of manpower of the specified service line in a specified time period;
  • the prediction model is divided into an incoming prediction model and an outgoing prediction model according to the service type, and the incoming prediction model is used for predicting a total amount of tasks and a human arrangement for the incoming call type service, and the outgoing prediction model is used. Forecasting the total amount of tasks and manpower arrangements for outbound type of business. Both the incoming prediction model and the exhalation prediction model use Monte Carlo simulation and geometric Brownian motion as stochastic models.
  • the data warehouse is composed of call data and dialing list data in a preset historical time after the cleaning process.
  • the incoming prediction model and the outgoing prediction model are respectively configured with corresponding input dimensions and output dimensions for the user to select.
  • the input dimension is a parameter type input to the incoming prediction model or the outgoing prediction model.
  • the output dimension is a parameter type that is output after the prediction data corresponding to the input dimension is processed by the call-in prediction model or the call-out prediction model.
  • the input dimensions include, but are not limited to, call duration, agent work utilization, call loss rate, satisfaction, agent skill level; output dimensions include, but are not limited to, call duration, number of calls, agent utilization, Call loss rate.
  • the input dimensions include, but are not limited to, the number of issued lists, the connection rate, the average call duration, the agent work utilization rate, and the agent skill level;
  • the output dimensions include, but are not limited to, the number of lists, the number of calls, and the duration of the call. , connection rate, agent utilization.
  • the embodiments of the present application implement different predictions for different service scenarios by establishing different prediction models and configuring different types of input and output dimensions for different prediction models.
  • the mode and the selection of more suitable input and output dimensions are beneficial to improve the accuracy of prediction for different service lines, and it is convenient for the user to select and adjust the prediction model corresponding to the specified service line and its input parameters and output parameters.
  • calculation module 54 includes:
  • the total amount calculating unit 541 is configured to sum the predicted values of the specified service line in a specified time period, and obtain the total number of tasks of the specified service line in a specified time period;
  • the conversion rate calculation unit 542 is configured to acquire the call duration and attendance data of the plurality of agents, calculate the work efficiency of each agent according to the call duration and the attendance data, and obtain an average value of the work efficiency to obtain a conversion rate. ;
  • the human computing unit 543 is configured to obtain a standard working time, calculate an average working time according to the standard working time and a conversion rate, and obtain a quotient between the total amount of the task and the average working time as the required amount of labor .
  • the embodiment of the present application calculates the average working time according to the actual call duration and the attendance data, and effectively improves the fit of the manual prediction.
  • the device further includes:
  • a third obtaining module 55 configured to acquire a marketing activity and an emergency event in the preset historical time after obtaining the predicted value of the output dimension, and determine a weekly date when the marketing activity and an emergency event occur;
  • the smoothing processing module 56 is configured to perform smoothing processing on the predicted value of the output dimension according to the weekly date of the marketing activity and the emergency event to eliminate the interference of the marketing activity and the emergency event on the predicted value.
  • the method includes:
  • a screening unit 561 configured to traverse each output dimension, and select a predicted value having the same week date as the base data from the predicted value of the output dimension;
  • a statistical processing unit 562 configured to calculate an average value and a standard deviation of the base data
  • a comparison unit 563 configured to calculate a difference between each of the base data and the average value, and compare an absolute value of the difference with the standard deviation
  • the smoothing processing unit 564 is configured to: when the absolute value of the difference is greater than the standard deviation, reduce the base data corresponding to the difference if the difference is a positive number, and increase if the difference is a negative number The base data corresponding to the difference is large.
  • the embodiment of the present application obtains the base data by the week date, and expands the coverage of the sample data, and the smoothed predicted value can effectively remove the fluctuation interference caused by the marketing activity and the emergency event.
  • the task total amount and manpower of the specified service line are predicted based on the smoothed predicted value, and the accuracy and adaptability of the prediction result are improved.
  • the terminal in the embodiment of the present application may be used to implement all the technical solutions in the foregoing method embodiments. It will be clearly understood by those skilled in the art that, for convenience and brevity of description, only the division of each functional unit and module described above is exemplified. In practical applications, the above functions may be assigned to different functional units according to needs.
  • the module is completed by dividing the internal structure of the device into different functional units or modules to perform all or part of the functions described above.
  • Each functional unit and module in the embodiment may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit, and the integrated unit may be implemented by hardware. Formal implementation can also be implemented in the form of software functional units.
  • FIG. 6 is a schematic diagram of a terminal according to an embodiment of the present application.
  • the terminal 6 of this embodiment includes a processor 60, a memory 61, and computer readable instructions 62 stored in the memory 61 and operable on the processor 60.
  • the processor 60 executes the computer readable instructions 62, the steps in the above-described traffic line based prediction device embodiment are implemented, such as steps S101 to S104 shown in FIG. 1 and steps S301 to S306 shown in FIG.
  • the processor 60 when executing the computer readable instructions 62, implements the functions of the various modules/units in the above-described traffic line based predictive device embodiment, such as the functions of the modules 51 through 56 shown in FIG.
  • the computer readable instructions 62 may be partitioned into one or more modules/units that are stored in the memory 61 and executed by the processor 60, To complete this application.
  • the one or more modules/units may be an instruction segment of a series of computer readable instructions capable of performing a particular function, the instruction segments being used to describe the execution of the computer readable instructions 62 in the terminal 6.
  • the computer readable instructions 62 may be divided into a first acquisition module, a second acquisition module, an analysis module, and a calculation module, and the specific functions of each module are as follows:
  • a first obtaining module configured to acquire a prediction model corresponding to the specified service line, and an input dimension and an output dimension of the current prediction when performing service prediction on the specified service line;
  • a second obtaining module configured to acquire, from the data warehouse, prediction data that meets the input dimension
  • An analysis module configured to perform trend analysis on the prediction data by using the Monte Carlo simulation method and the geometric Brownian motion, to obtain a predicted value of the output dimension
  • a calculation module configured to calculate, according to the predicted value, a total amount of tasks and a required amount of manpower of the specified service line in a specified time period
  • the prediction model is divided into an incoming prediction model and an outgoing prediction model according to the service type, and the data warehouse is composed of call data and dialing list data in a preset historical time after the cleaning process.
  • computer readable instructions 62 can also be segmented:
  • a third obtaining module configured to acquire a marketing activity and an emergency event in the preset historical time after obtaining the predicted value of the output dimension, and determine a weekly date when the marketing activity and an emergency event occur;
  • a smoothing processing module configured to perform smoothing processing on the predicted value of the output dimension according to the weekly date of the marketing activity and the emergency event, to eliminate the interference of the marketing activity and the emergency event on the predicted value.
  • the smoothing processing module includes:
  • a screening unit configured to traverse each output dimension, and select a predicted value having the same week date as the base data from the predicted value of the output dimension;
  • a statistical processing unit configured to calculate an average value and a standard deviation of the base data
  • a comparing unit configured to calculate a difference between each of the base data and the average value, and compare an absolute value of the difference with the standard deviation
  • a smoothing processing unit configured to: when the absolute value of the difference is greater than the standard deviation, reduce the base data corresponding to the difference if the difference is a positive number, and increase if the difference is a negative number The difference corresponds to the base data.
  • calculation module includes:
  • a total amount calculation unit configured to sum the predicted values of the specified service line in a specified time period, and obtain a total amount of tasks of the specified service line in a specified time period;
  • the conversion rate calculation unit is configured to acquire the call duration and the attendance data of the plurality of agents, calculate the work efficiency of each agent according to the call duration and the attendance data, and obtain an average value of the work efficiency to obtain a conversion rate;
  • the human computing unit is configured to obtain a standard working time, calculate an average working time according to the standard working time and a conversion rate, and obtain a quotient between the total amount of the task and the average working time as the required amount of labor.
  • the terminal 6 can be a computing device such as a desktop computer, a notebook, a palmtop computer, and a cloud server.
  • the terminal may include, but is not limited to, a processor 60, a memory 61. It will be understood by those skilled in the art that FIG. 6 is merely an example of the terminal 6, and does not constitute a limitation of the terminal 6, and may include more or less components than those illustrated, or combine some components, or different components, such as
  • the terminal may also include an input/output device, a network access device, a bus, and the like.
  • the processor 60 may be a central processing unit (CPU), or may be another general-purpose processor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), Field-Programmable Gate Array (FPGA) or other programmable logic device, discrete gate or transistor logic device, discrete hardware components, etc.
  • the general purpose processor may be a microprocessor or the processor or any conventional processor or the like, which is a control center of the terminal, and connects various parts of the entire terminal using various interfaces and lines.
  • the memory 61 can be used to store the computer readable instructions and/or modules by running or executing computer readable instructions and/or modules stored in the memory, and recalling data stored in the memory Implementing various functions of the terminal.
  • the memory may mainly include a storage program area and a storage data area, wherein the storage program area may store an operating system, an application required for at least one function (such as a sound playing function, an image playing function, etc.), and the like; the storage data area may be stored. Data created based on the use of the terminal, etc.
  • the memory may include a high-speed random access memory, and may also include non-volatile memory such as a hard disk, a memory, a plug-in hard disk, a smart memory card (SMC), and a secure digital card (Secure Digital, SD). , Flash Card, at least one disk storage device, flash memory device, or other volatile solid-state storage device.
  • non-volatile memory such as a hard disk, a memory, a plug-in hard disk, a smart memory card (SMC), and a secure digital card (Secure Digital, SD).
  • SD Secure Digital
  • Flash Card at least one disk storage device, flash memory device, or other volatile solid-state storage device.
  • the disclosed apparatus/terminal and method may be implemented in other manners.
  • the device/terminal device embodiments described above are merely illustrative.
  • the division of the modules or units is only a logical function division.
  • there may be another division manner for example, multiple units.
  • components may be combined or integrated into another system, or some features may be omitted or not performed.
  • the mutual coupling or direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection through some interface, device or unit, and may be in electrical, mechanical or other form.
  • the units described as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units, that is, may be located in one place, or may be distributed to multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of the embodiment.
  • each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.
  • the above integrated unit can be implemented in the form of hardware or in the form of a software functional unit.
  • the integrated modules/units if implemented in the form of software functional units and sold or used as separate products, may be stored in a computer readable storage medium.
  • the present application implements all or part of the processes in the foregoing embodiments, and may also be implemented by computer readable instructions, which may be stored in a computer readable storage medium.
  • the computer readable instructions when executed by a processor, may implement the steps of the various method embodiments described above.
  • the computer readable instructions comprise computer readable instruction code, which may be in the form of source code, an object code form, an executable file or some intermediate form or the like.
  • the computer readable storage medium may include any entity or device capable of carrying the computer readable instruction code, a recording medium, a USB flash drive, a removable hard disk, a magnetic disk, an optical disk, a computer memory, a read only memory (ROM, Read- Only Memory), Random Access Memory (RAM), electrical carrier signals, telecommunications signals, and software distribution media. It should be noted that the content contained in the computer readable storage medium may be appropriately increased or decreased according to the requirements of legislation and patent practice in a jurisdiction, for example, in some jurisdictions, according to legislation and patent practice, computer readable The storage medium does not include an electrical carrier signal and a telecommunication signal.

Landscapes

  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Engineering & Computer Science (AREA)
  • Strategic Management (AREA)
  • Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Educational Administration (AREA)
  • Theoretical Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Development Economics (AREA)
  • Marketing (AREA)
  • Game Theory and Decision Science (AREA)
  • Operations Research (AREA)
  • Quality & Reliability (AREA)
  • Tourism & Hospitality (AREA)
  • General Business, Economics & Management (AREA)
  • Databases & Information Systems (AREA)
  • Signal Processing (AREA)
  • General Engineering & Computer Science (AREA)
  • Data Mining & Analysis (AREA)
  • Computer Hardware Design (AREA)
  • Evolutionary Computation (AREA)
  • Geometry (AREA)
  • Mathematical Optimization (AREA)
  • Mathematical Analysis (AREA)
  • Algebra (AREA)
  • Mathematical Physics (AREA)
  • Probability & Statistics with Applications (AREA)
  • Pure & Applied Mathematics (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Telephonic Communication Services (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

本申请适用于通信技术领域，提供了一种基于业务线的预测方法、装置、存储介质及终端，所述方法包括：在对指定业务线进行业务预测时，获取该指定业务线对应的预测模型，以及本次预测的输入维度和输出维度；从数据仓库中获取满足所述输入维度的预测数据；通过所述预测模型采用蒙特卡罗模拟法和几何布朗运动对所述预测数据进行趋势分析，得到所述输出维度的预测值；根据所述预测值计算所述指定业务线在指定时间段内的任务总量和所需投入的人力数量；其中，所述预测模型根据业务类型划分为呼入预测模型和呼出预测模型。本申请实现了针对不同的业务场景采用不同的预测模式，提高了对不同业务线进行预测的准确度。

Description

基于业务线的预测方法、装置、存储介质及终端
本专利申请以2017年07月26日提交的申请号为201710615821.X，名称为“基于业务线的预测方法、装置、存储介质及终端”的中国发明专利申请为基础，并要求其优先权。
技术领域
本申请属于通信技术领域，尤其涉及一种基于业务线的预测方法、装置、存储介质及终端。
背景技术
在针对呼入业务和呼出业务的排班中，现有技术主要采用的时间序列预测法、回归预测模型等预测模式进行排班预测。其中，时间序列预测法是一种历史资料延伸预测，以时间数列所能反映的发展过程和规律进行引伸外推，预测发展趋势；回归预测模型通过分析市场上的自变量和因变量之间的相关关系，建立变量之间的回归方程，以所述回归方程作为预测模型。然而，呼出业务侧重所呼叫的客户量，同时涉及到客户名单的接通率以及客户对产品的购买欲望；而呼入业务侧重通话时长、通话次数，不同的业务类型会涉及到不同的通话时长、通话次数。现有技术针对不同的业务场景采用统一的预测模式，排班预测的准确度不高，难以满足市场日趋复杂的需求。
发明内容
本申请实施例提供了一种基于业务线的预测方法、装置、存储介质及终端，所述预测方法包括：
在对指定业务线进行业务预测时，获取该指定业务线对应的预测模型，以及本次预测的输入维度和输出维度；
从数据仓库中获取满足所述输入维度的预测数据；
通过所述预测模型采用蒙特卡罗模拟法和几何布朗运动对所述预测数据进行趋势分析，得到所述输出维度的预测值；
根据所述预测值计算所述指定业务线在指定时间段内的任务总量和所需投入的人力数量；
其中，所述预测模型根据业务类型划分为呼入预测模型和呼出预测模型，所述数据仓 库由清洗处理后的预设历史时间内的通话数据、拨打名单数据构成。
进一步地，在得到所述输出维度的预测值之后，所述预测方法还包括：
获取所述预设历史时间内的营销活动和突发事件，确定所述营销活动和突发事件发生时的周日期；
根据所述营销活动和突发事件的周日期对所述输出维度的预测值进行平滑处理，以消除所述营销活动、突发事件对所述预测值的干扰。
进一步地，所述根据所述营销活动和突发事件的周日期对所述输出维度的预测值进行平滑处理，以消除所述营销活动、突发事件对所述预测值的干扰包括：
遍历每一个输出维度，从所述输出维度的预测值中筛选出具有相同周日期的预测值作为基底数据；
计算所述基底数据的平均值和标准差；
计算每一个基底数据与所述平均值之间的差值，比对所述差值的绝对值与所述标准差；
当所述差值的绝对值大于所述标准差时，若所述差值为正数则缩小所述差值对应的基底数据，若所述差值为负数则增大所述差值对应的基底数据。
进一步地，所述根据所述预测值计算所述指定业务线在指定时间段内的任务总量和所需投入的人力数量包括：
对所述指定业务线在指定时间段内的预测值进行求和，得到所述指定业务线在指定时间段内的任务总量；
获取若干坐席人员的通话日时长、考勤数据，根据所述通话日时长、考勤数据计算每一个坐席人员的工作效率，求取所述工作效率的平均值得到折算率；
获取标准工作时长，根据所述标准工作时长和折算率计算平均工作时长，求取所述任务总量与所述平均工作时长之间的商作为所需投入的人力数量。
本申请实施例还提供了一种基于业务线的预测装置，所述预测装置包括：
第一获取模块，用于在对指定业务线进行业务预测时，获取该指定业务线对应的预测模型，以及本次预测的输入维度和输出维度；
第二获取模块，用于从数据仓库中获取满足所述输入维度的预测数据；
分析模块，用于通过所述预测模型采用蒙特卡罗模拟法和几何布朗运动对所述预测数据进行趋势分析，得到所述输出维度的预测值；
计算模块，用于根据所述预测值计算所述指定业务线在指定时间段内的任务总量和所需投入的人力数量；
其中，所述预测模型根据业务类型划分为呼入预测模型和呼出预测模型，所述数据仓 库由清洗处理后的预设历史时间内的通话数据、拨打名单数据构成。
进一步地，所述装置还包括：
第三获取模块，用于在得到所述输出维度的预测值之后，获取所述预设历史时间内的营销活动和突发事件，确定所述营销活动和突发事件发生时的周日期；
平滑处理模块，用于根据所述营销活动和突发事件的周日期对所述输出维度的预测值进行平滑处理，以消除所述营销活动、突发事件对所述预测值的干扰。
进一步地，所述根据所述平滑处理模块包括：
筛选单元，用于遍历每一个输出维度，从所述输出维度的预测值中筛选出具有相同周日期的预测值作为基底数据；
统计处理单元，用于计算所述基底数据的平均值和标准差；
比较单元，用于计算每一个基底数据与所述平均值之间的差值，比对所述差值的绝对值与所述标准差；
平滑处理单元，用于当所述差值的绝对值大于所述标准差时，若所述差值为正数则缩小所述差值对应的基底数据，若所述差值为负数则增大所述差值对应的基底数据。
进一步地，所述计算模块包括：
总量计算单元，用于对所述指定业务线在指定时间段内的预测值进行求和，得到所述指定业务线在指定时间段内的任务总量；
折算率计算单元，用于获取若干坐席人员的通话日时长、考勤数据，根据所述通话日时长、考勤数据计算每一个坐席人员的工作效率，求取所述工作效率的平均值得到折算率；
人力计算单元，用于获取标准工作时长，根据所述标准工作时长和折算率计算平均工作时长，求取所述任务总量与所述平均工作时长之间的商作为所需投入的人力数量。
本申请实施例还提供了一种计算机可读存储介质，其上存储有计算机可读指令，该计算机可读指令被处理器执行时实现如下步骤：
在对指定业务线进行业务预测时，获取该指定业务线对应的预测模型，以及本次预测的输入维度和输出维度；
从数据仓库中获取满足所述输入维度的预测数据；
通过所述预测模型采用蒙特卡罗模拟法和几何布朗运动对所述预测数据进行趋势分析，得到所述输出维度的预测值；
根据所述预测值计算所述指定业务线在指定时间段内的任务总量和所需投入的人力数量；
其中，所述预测模型根据业务类型划分为呼入预测模型和呼出预测模型，所述数据仓 库由清洗处理后的预设历史时间内的通话数据、拨打名单数据构成。
本申请实施例还提供了一种终端，包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机可读指令，所述处理器执行所述计算机可读指令时实现以下步骤：
在对指定业务线进行业务预测时，获取该指定业务线对应的预测模型，以及本次预测的输入维度和输出维度；
从数据仓库中获取满足所述输入维度的预测数据；
通过所述预测模型采用蒙特卡罗模拟法和几何布朗运动对所述预测数据进行趋势分析，得到所述输出维度的预测值；
根据所述预测值计算所述指定业务线在指定时间段内的任务总量和所需投入的人力数量；
其中，所述预测模型根据业务类型划分为呼入预测模型和呼出预测模型，所述数据仓库由清洗处理后的预设历史时间内的通话数据、拨打名单数据构成。
与现有技术相比，本申请实施例根据不同的业务类型构建了不同的预测模型，包括呼入预测模型和呼出预测模型；在对指定业务线进行业务预测时，获取该指定业务线对应的预测模型，以及本次预测的输入维度和输出维度；然后从数据仓库中获取满足所述输入维度的预测数据；通过所述预测模型采用蒙特卡罗模拟法和几何布朗运动对所述预测数据进行趋势分析，得到所述输出维度的预测值；最后根据所述预测值计算所述指定业务线在指定时间段内的任务总量和所需投入的人力数量；从而实现了针对不同的业务场景采用不同的预测模式，提高了对不同业务线进行预测的准确度。
附图说明
为了更清楚地说明本申请实施例或现有技术中的技术方案，下面将对实施例或现有技术描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本申请的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他附图。
图1是本申请实施例提供的基于业务线的预测方法的第一实现流程图；
图2是本申请实施例提供的基于业务线的预测方法的第一实现流程中步骤S104的实现流程图；
图3是本申请实施例提供的基于业务线的预测方法的第二实现流程图；
图4是本申请实施例提供的基于业务线的预测方法的第二实现流程中步骤S305的实现 流程图；
图5是本申请实施例提供的基于业务线的预测装置的组成结构图；
图6是本申请实施例提供的终端的示意图。
具体实施方式
为了使本申请的目的、技术方案及优点更加清楚明白，以下结合附图及实施例，对本申请进行进一步详细说明。应当理解，此处所描述的具体实施例仅仅用以解释本申请，并不用于限定本申请。
在本申请实施例中，对业务线的预测为对业务线的工作量和人力进行的预测，以为排班预测做准备。可选地，本申请实施例所述的基于业务线的预测方法可应用于终端，所述终端包括但不限于计算机、服务器、笔记本电脑。图1示出了本申请实施例提供的基于业务线的预测方法的第一实现流程。
参阅图1，所述基于业务线的预测方法包括：
在步骤S101中，在对指定业务线进行业务预测时，获取该指定业务线对应的预测模型，以及本次预测的输入维度和输出维度。
在这里，本申请实施例根据不同的业务类型建立了对应的预测模型，包括呼入预测模型和呼出预测模型。其中，所述呼入预测模型用于对呼入类型的业务进行任务总量和人力安排的预测，所述呼出预测模型用于对呼出类型的业务进行任务总量和人力安排的预测。在本申请实施例中，所述呼入预测模型和呼出预测模型均以蒙特卡罗模拟法和几何布朗运动作为随机模型，并从业务性数据库中提取出预设历史时间内的通话数据、拨打名单数据进行加工、分析后导入分析性数据库，构建出所述呼入预测模型和呼出预测模型对应的数据仓库。可选地，所述预设历史时间优选为过去一年内。
在本申请实施例中，所述呼入预测模型和呼出预测模型还分别配置了对应的输入维度和输出维度供用户选择。其中，所述输入维度为输入至呼入预测模型或者呼出预测模型的参数种类。所述输出维度为输入维度对应的预测数据经呼入预测模型或者呼出预测模型处理后输出的参数种类。
对于所述呼入预测模型，其输入维度包括但不限于通话时长、坐席工作利用率、呼损率、满意度、坐席技能等级；输出维度包括但不限于通话时长、通话次数、坐席利用率、呼损率。
对于所述呼出预测模型，其输入维度包括但不限于下发名单量、接通率、平均通话时 长、坐席工作利用率、坐席技能等级；输出维度包括但不限于名单量、拨打次数、拨打时长、接通率，坐席利用率。
与以往采用统一的预测模式进行预测，本申请实施例通过建立不同的预测模型，以及为不同的预测模型配置不同种类的输入维度和输出维度，实现了针对不同的业务场景采用不同的预测模式，以及选择更适配的输入维度和输出维度，有利于提高对不同业务线进行预测的准确度，且方便了用户选择和调整指定业务线对应的预测模型及其输入参数、输出参数。
在对业务线进行预测前，用户可以预先在终端上选择所述业务线对应的预测模型，以及选定用于预测的输入维度和输出维度。终端在接收到排班预测指令对指定业务线进行工作量预测时，则根据当前的指定业务线获取对应的预测模型，以及本次预测的输入维度和输出维度。
在步骤S102中，从数据仓库中获取满足所述输入维度的预测数据。
如前文所述，所述预测模型以经过清洗后的预设历史时间内的通话数据、拨打名单数据创建了数据仓库。因此，在进行预测时，本申请实施例基于所选定的输入维度从数据仓库中筛选出预测数据。
示例性地，对于所述呼入预测模型，其输入维度包括但不限于通话时长、坐席工作利用率、呼损率、满意度、坐席技能等级。若选定的用于预测的输入维度包括通话时长、坐席工作利用率、呼损率三种参数时，则从数据仓库中筛选出满足上述三种输入维度的数据作为预测数据。
在步骤S103中，通过所述预测模型采用蒙特卡罗模拟法和几何布朗运动对所述预测数据进行趋势分析，得到所述输出维度的预测值。
示例性地，在使用蒙特卡罗模拟法和几何布朗运动对预测数据进行趋势分析时，先选择合适的先验分布模型，然后基于上述预测数据，利用给定的规则进行快速充分大量的随机抽样，对抽样的数据进行数学计算和统计学处理，再根据上述统计学处理结果生成概率分布曲线和累计概率曲线，通常为基于正太分布的概率累计S曲线，依据所述累计概率曲线进行趋势分析，得到预测值，最后筛选出满足所选定的输出维度的预测值。示例性地，若选定的输出维度为通话时长，则经过预测模型之后，得到该通话时长的预测值。
在步骤S104中，根据所述预测值计算所述指定业务线在指定时间段内的任务总量和所需投入的人力数量。
在得到满足输出维度的预测值之后，则基于所述预测值计算所述指定业务线在指定时间段内的工作量。其中，所述指定时间段小于所述数据仓库中的通话数据、拨打名单数据的时间跨度。
可选地，图2示出了本申请实施例提供的基于业务线的预测方法中步骤S104的具体实现流程。参阅图2，所述步骤S104包括：
在步骤S201中，对所述指定业务线在指定时间段内的预测值进行求和，得到所述指定业务线在指定时间段内的任务总量。
在这里，本申请实施例首先根据通过预测模型得到的预测值来计算指定业务线在指定时间段内的任务总量。示例性地，对于坐席人员，其任务总量通过时间(分钟)来表示。假设本次预测为呼入预测模型，所选的输出维度为通话时长，指定时间段为6月25日至6月29日供五天，则在得到通话时长的预测值之后，求取在6月25日至6月29日五天的预测值之和，从而得到所述指定业务线在所述指定时间段内的任务总量。
在步骤S202中，获取若干坐席人员的通话日时长、考勤数据，根据所述通话日时长、考勤数据计算每一个坐席人员的工作效率，求取所述工作效率的平均值得到折算率。
在这里，尽管标准工作时长是规定的，但是每个坐席人员在上班的时候并不能够保证100％充分利用该标准工作时长，在工作时间内会出现会议、小憩、请假、休假等情况。鉴于此，本申请实施例获取若干坐席人员的通话日时长以及考勤数据；该通话日时长为单个坐席人员在一天内所有通话相加后的总时长。然后根据所述通话日时长、考勤数据计算每一个坐席人员的工作效率，求取所述工作效率的平均值得到折算率。所述折算率为标准工作时长有效利用的平均概率。
在步骤S203中，获取标准工作时长，根据所述标准工作时长和折算率计算平均工作时长，求取所述任务总量与所述平均工作时长之间的商作为所需投入的人力数量。
在得到所述折算率后，求取所述标准工作时长和折算率之间的乘积，以得到坐席人员的平均工作时长。该平均工作时长反映了单个坐席人员的日工作时长。最后求取所述任务总量与所述平均工作时长之间的商，本申请实施例以所述商值则作为需要投入的人力数量，从而实现了基于业务线的人力预测，后续排班则根据所述人力数量来展开。本申请实施例根据实际的通话日时长及考勤数据来计算平均工作时长，有效地提高了人力预测的适配度。
进一步地，营销活动、系统异常、设备故障等情况会导致历史数据产生较大的波动，这些波动会影响到排班的预测值，因此，本申请实施例还包括对通过预测模型得到的预测值进行平滑处理。
基于上述图1实施例所述的基于业务线的预测方法的第一实现流程，提出本申请实施例所述的基于业务线的预测方法的第二实现流程。参阅图3，所述基于业务线的预测方法包括：
步骤S301至步骤S303，其中，步骤S301至步骤S303与图1实施例中所述的步骤S101至步骤S103相同，具体请参见上述实施例的叙述，此处不再赘述。
在得到所述输出维度的预测值之后，所述预测方法还包括：
在步骤S304中，获取所述预设历史时间内的营销活动和突发事件，确定所述营销活动和突发事件发生时的周日期。
在这里，所述突发事件包括但不限于系统异常、设备故障等情况。所述预设历史时间为数据仓库内的通话数据、拨打名单数据的时间跨度。本申请实施例获取该时间跨度范围内的营销活动和突发事件的发生日期。所述发生日期为周日期，所述周日期为一周七天中的日期，比如周一、周二、周三、周四、周五、周六、周日。
在步骤S305中，根据所述营销活动和突发事件的周日期对所述输出维度的预测值进行平滑处理，以消除所述营销活动、突发事件对所述预测值的干扰。
可选地，图4示出了本申请实施例提供的基于业务线的预测方法的第二实现流程中步骤S305的具体实现流程。参阅图4，所述步骤S305包括：
在步骤S401中，遍历每一个输出维度，从所述输出维度的预测值中筛选出具有相同周日期的预测值作为基底数据。
对于每一个输出维度，基于所述周日期，从所述输出维度的预测值中筛选出具有相同周日期的预测值。示例性地，若营销活动的周日期为周二，则筛选出输出维度在每个周二时的预测值，以所筛选出来的预测值作为基底数据。
在步骤S402中，计算所述基底数据的平均值和标准差。
在这里，所述基底数据的标准差作为是否对所述基底数据进行平滑处理的判断标准。
在步骤S403中，计算每一个基底数据与所述平均值之间的差值，比对所述差值的绝对值与所述标准差。
如前所述，在得到输出维度在每周二的预测值的平均值和标准差之后，计算所述输出维度在每周二时的预测值与所述平均值之间的差值，比对所述差值的绝对值与步骤S402中计算得到的标准差，以确定是否对所述预测值进行修正。
在步骤S404中，当所述差值的绝对值大于所述标准差时，若所述差值为正数则缩小所述差值对应的基底数据，若所述差值为负数则增大所述差值对应的基底数据。
在这里，本申请实施例以误差不在所述标准差范围内的预测值为异常数据，即所述差值的绝对值大于所述标准差时对所述预测值进行修正，包括：判断所述差值的正负，若所述差值为正数表明所述差值对应的基底数据偏大则缩小对应的基底数据，若所述差值为负数表明所述差值对应的基底数据偏小则增大对应的基底数据，从而完成了对输出维度的预测值的 平滑处理。在这里，营销活动期间的基底数据通常明显高于或大于所述平均值，因此，营销活动期间的基底数据与平均值之间的差值为正数且所述差值大于所述标准差，此时缩小所述营销活动期间的基底数据，使得营销活动期间的基底数据趋同于所述预设历史时间内的规律，从而剔除营销活动对预测值产生的波动干扰。突发事件发生期间的基底数据通常明显低于或小于所述平均值，因此，突发事件期间的基底数据与平均值之间的差值为负数且所述差值的绝对值大于所述标准差，此时增大所述突发事件期间的基底数据，使得突发事件期间的基底数据趋同于所述预设历史时间内的规律，从而剔除突发事件对预测值产生的波动干扰。
在步骤S306中，根据所述预测值计算所述指定业务线在指定时间段内的任务总量和所需投入的人力数量。
由于平滑处理后的预测值去掉了营销活动、突发事件产生的波动干扰，基于所述平滑处理后的预测值来预测指定业务线的任务总量和人力，可有效地提高预测结果的准确性和适配度。
应理解，在上述实施例中，各步骤的序号的大小并不意味着执行顺序的先后，各步骤的执行顺序应以其功能和内在逻辑确定，而不应对本申请实施例的实施过程构成任何限定。
需要说明的是，本领域普通技术人员可以理解实现上述实施例的全部或部分步骤可以通过硬件来完成，也可以通过计算机可读指令来指令相关的硬件完成，所述的计算机可读指令可以存储于一种计算机可读存储介质中，所述存储介质可以是只读存储器，磁盘或光盘等。
图5示出了本申请实施例提供的基于业务线的预测装置的组成结构图，为了便于说明，仅示出了与本申请实施例相关的部分。
在本申请实施例中，所述基于业务线的预测装置用于实现上述图1至图4实施例中所述的基于业务线的预测方法，可以是内置于终端的软件单元、硬件单元或者软硬件结合的单元。所述终端包括但不限于计算机、服务器、笔记本电脑。
参阅图5，所述基于业务线的预测装置包括：
第一获取模块51，用于在对指定业务线进行业务预测时，获取该指定业务线对应的预测模型，以及本次预测的输入维度和输出维度；
第二获取模块52，用于从数据仓库中获取满足所述输入维度的预测数据；
分析模块53，用于通过所述预测模型采用蒙特卡罗模拟法和几何布朗运动对所述预测数据进行趋势分析，得到所述输出维度的预测值；
计算模块54，用于根据所述预测值计算所述指定业务线在指定时间段内的任务总量和所需投入的人力数量；
其中，所述预测模型根据业务类型划分为呼入预测模型和呼出预测模型，所述呼入预测模型用于对呼入类型的业务进行任务总量和人力安排的预测，所述呼出预测模型用于对呼出类型的业务进行任务总量和人力安排的预测。所述呼入预测模型和呼出预测模型均以蒙特卡罗模拟法和几何布朗运动作为随机模型。所述数据仓库由清洗处理后的预设历史时间内的通话数据、拨打名单数据构成。
在本申请实施例中，所述呼入预测模型和呼出预测模型还分别配置了对应的输入维度和输出维度供用户选择。其中，所述输入维度为输入至呼入预测模型或者呼出预测模型的参数种类。所述输出维度为输入维度对应的预测数据经呼入预测模型或者呼出预测模型处理后输出的参数种类。
对于所述呼入预测模型，其输入维度包括但不限于通话时长、坐席工作利用率、呼损率、满意度、坐席技能等级；输出维度包括但不限于通话时长、通话次数、坐席利用率、呼损率。
对于所述呼出预测模型，其输入维度包括但不限于下发名单量、接通率、平均通话时长、坐席工作利用率、坐席技能等级；输出维度包括但不限于名单量、拨打次数、拨打时长、接通率，坐席利用率。
与以往采用统一的预测模式进行预测相比，本申请实施例通过建立不同的预测模型，以及为不同的预测模型配置不同种类的输入维度和输出维度，实现了针对不同的业务场景采用不同的预测模式，以及选择更适配的输入维度和输出维度，有利于提高对不同业务线进行预测的准确度，且方便了用户选择和调整指定业务线对应的预测模型及其输入参数、输出参数。
进一步地，所述计算模块54包括：
总量计算单元541，用于对所述指定业务线在指定时间段内的预测值进行求和，得到所述指定业务线在指定时间段内的任务总量；
折算率计算单元542，用于获取若干坐席人员的通话日时长、考勤数据，根据所述通话日时长、考勤数据计算每一个坐席人员的工作效率，求取所述工作效率的平均值得到折算率；
人力计算单元543，用于获取标准工作时长，根据所述标准工作时长和折算率计算平均工作时长，求取所述任务总量与所述平均工作时长之间的商作为所需投入的人力数量。
在这里，本申请实施例根据实际的通话日时长及考勤数据来计算平均工作时长，有效地提高了人力预测的适配度。
进一步地，营销活动、系统异常、设备故障等情况会导致历史数据产生较大的波动，这些波动会影响到排班的预测值。鉴于此，所述装置还包括：
第三获取模块55，用于在得到所述输出维度的预测值之后，获取所述预设历史时间内的营销活动和突发事件，确定所述营销活动和突发事件发生时的周日期；
平滑处理模块56，用于根据所述营销活动和突发事件的周日期对所述输出维度的预测值进行平滑处理，以消除所述营销活动、突发事件对所述预测值的干扰。
进一步地，所述根据所述平滑处理模块56包括：
筛选单元561，用于遍历每一个输出维度，从所述输出维度的预测值中筛选出具有相同周日期的预测值作为基底数据；
统计处理单元562，用于计算所述基底数据的平均值和标准差；
比较单元563，用于计算每一个基底数据与所述平均值之间的差值，比对所述差值的绝对值与所述标准差；
平滑处理单元564，用于当所述差值的绝对值大于所述标准差时，若所述差值为正数则缩小所述差值对应的基底数据，若所述差值为负数则增大所述差值对应的基底数据。
在这里，本申请实施例以周日期来获取基底数据，扩大了样本数据的覆盖范围，平滑处理后的预测值可有效地去掉营销活动、突发事件产生的波动干扰。基于所述平滑处理后的预测值来预测指定业务线的任务总量和人力，提高了预测结果的准确性和适配度。
需要说明的是，本申请实施例中的终端可以用于实现上述方法实施例中的全部技术方案。所属领域的技术人员可以清楚地了解到，为了描述的方便和简洁，仅以上述各功能单元、模块的划分进行举例说明，实际应用中，可以根据需要而将上述功能分配由不同的功能单元、模块完成，即将所述装置的内部结构划分成不同的功能单元或模块，以完成以上描述的全部或者部分功能。实施例中的各功能单元、模块可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中，上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。另外，各功能单元、模块的具体名称也只是为了便于相互区分，并不用于限制本申请的保护范围。上述单元、模块的具体工作过程，可以参考前述方法实施例中的对应过程，在此不再赘述。
在上述实施例中，对各个实施例的描述都各有侧重，某个实施例中没有详述或记载的部分，可以参见其它实施例的相关描述。
图6是本申请实施例提供的一种终端的示意图。如图6所示，该实施例的终端6包括：处理器60、存储器61以及存储在所述存储器61中并可在所述处理器60上运行的计算机可 读指令62。所述处理器60执行所述计算机可读指令62时实现上述基于业务线的预测装置实施例中的步骤，例如图1所示的步骤S101至S104、图3所示的步骤S301至S306。或者，所述处理器60执行所述计算机可读指令62时实现上述基于业务线的预测装置实施例中各模块/单元的功能，例如图5所示模块51至56的功能。
示例性的，所述计算机可读指令62可以被分割成一个或多个模块/单元，所述一个或者多个模块/单元被存储在所述存储器61中，并由所述处理器60执行，以完成本申请。所述一个或多个模块/单元可以是能够完成特定功能的一系列计算机可读指令的指令段，该指令段用于描述所述计算机可读指令62在所述终端6中的执行过程。例如，所述计算机可读指令62可以被分割成第一获取模块、第二获取模块、分析模块、计算模块，各模块具体功能如下：
第一获取模块，用于在对指定业务线进行业务预测时，获取该指定业务线对应的预测模型，以及本次预测的输入维度和输出维度；
第二获取模块，用于从数据仓库中获取满足所述输入维度的预测数据；
分析模块，用于通过所述预测模型采用蒙特卡罗模拟法和几何布朗运动对所述预测数据进行趋势分析，得到所述输出维度的预测值；
计算模块，用于根据所述预测值计算所述指定业务线在指定时间段内的任务总量和所需投入的人力数量；
其中，所述预测模型根据业务类型划分为呼入预测模型和呼出预测模型，所述数据仓库由清洗处理后的预设历史时间内的通话数据、拨打名单数据构成。
进一步地，所述计算机可读指令62还可以分割出：
第三获取模块，用于在得到所述输出维度的预测值之后，获取所述预设历史时间内的营销活动和突发事件，确定所述营销活动和突发事件发生时的周日期；
平滑处理模块，用于根据所述营销活动和突发事件的周日期对所述输出维度的预测值进行平滑处理，以消除所述营销活动、突发事件对所述预测值的干扰。
进一步地，所述根据所述平滑处理模块包括：
筛选单元，用于遍历每一个输出维度，从所述输出维度的预测值中筛选出具有相同周日期的预测值作为基底数据；
统计处理单元，用于计算所述基底数据的平均值和标准差；
比较单元，用于计算每一个基底数据与所述平均值之间的差值，比对所述差值的绝对值与所述标准差；
平滑处理单元，用于当所述差值的绝对值大于所述标准差时，若所述差值为正数则缩 小所述差值对应的基底数据，若所述差值为负数则增大所述差值对应的基底数据。
进一步地，所述计算模块包括：
总量计算单元，用于对所述指定业务线在指定时间段内的预测值进行求和，得到所述指定业务线在指定时间段内的任务总量；
折算率计算单元，用于获取若干坐席人员的通话日时长、考勤数据，根据所述通话日时长、考勤数据计算每一个坐席人员的工作效率，求取所述工作效率的平均值得到折算率；
人力计算单元，用于获取标准工作时长，根据所述标准工作时长和折算率计算平均工作时长，求取所述任务总量与所述平均工作时长之间的商作为所需投入的人力数量。
所述终端6可以是桌上型计算机、笔记本、掌上电脑及云端服务器等计算设备。所述终端可包括，但不仅限于，处理器60、存储器61。本领域技术人员可以理解，图6仅仅是终端6的示例，并不构成对终端6的限定，可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件，例如所述终端还可以包括输入输出设备、网络接入设备、总线等。
所称处理器60可以是中央处理单元(Central Processing Unit，CPU)，还可以是其他通用处理器、数字信号处理器(Digital Signal Processor，DSP)、专用集成电路(Application Specific Integrated Circuit，ASIC)、现成可编程门阵列(Field-Programmable Gate Array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器或者该处理器也可以是任何常规的处理器等，所述处理器是所述终端的控制中心，利用各种接口和线路连接整个终端的各个部分。
所述存储器61可用于存储所述计算机可读指令和/或模块，所述处理器通过运行或执行存储在所述存储器内的计算机可读指令和/或模块，以及调用存储在存储器内的数据，实现所述终端的各种功能。所述存储器可主要包括存储程序区和存储数据区，其中，存储程序区可存储操作系统、至少一个功能所需的应用程序(比如声音播放功能、图像播放功能等)等；存储数据区可存储根据终端的使用所创建的数据等。此外，存储器可以包括高速随机存取存储器，还可以包括非易失性存储器，例如硬盘、内存、插接式硬盘、智能存储卡(Smart Media Card，SMC)、安全数字卡(Secure Digital，SD)、闪存卡(Flash Card)，至少一个磁盘存储器件、闪存器件或其他易失性固态存储器件。
本领域普通技术人员可以意识到，结合本文中所公开的实施例描述的各示例的单元及算法步骤，能够以电子硬件、或者计算机软件和电子硬件的结合来实现。这些功能究竟以硬件还是软件方式来执行，取决于技术方案的特定应用和设计约束条件。专业技术人员可以对每个特定的应用来使用不同方法来实现所描述的功能，但是这种实现不应认为超出本申请的范 围。
在本申请所提供的实施例中，应该理解到，所揭露的装置/终端和方法，可以通过其它的方式实现。例如，以上所描述的装置/终端设备实施例仅仅是示意性的，例如，所述模块或单元的划分，仅仅为一种逻辑功能划分，实际实现时可以有另外的划分方式，例如多个单元或组件可以结合或者可以集成到另一个系统，或一些特征可以忽略，或不执行。另一点，所显示或讨论的相互之间的耦合或直接耦合或通讯连接可以是通过一些接口，装置或单元的间接耦合或通讯连接，可以是电性，机械或其它的形式。
所述作为分离部件说明的单元可以是或者也可以不是物理上分开的，作为单元显示的部件可以是或者也可以不是物理单元，即可以位于一个地方，或者也可以分布到多个网络单元上。可以根据实际的需要选择其中的部分或者全部单元来实现本实施例方案的目的。
另外，在本申请各个实施例中的各功能单元可以集成在一个处理单元中，也可以是各个单元单独物理存在，也可以两个或两个以上单元集成在一个单元中。上述集成的单元既可以采用硬件的形式实现，也可以采用软件功能单元的形式实现。
所述集成的模块/单元如果以软件功能单元的形式实现并作为独立的产品销售或使用时，可以存储在一个计算机可读存储介质中。基于这样的理解，本申请实现上述实施例方法中的全部或部分流程，也可以通过计算机可读指令来指令相关的硬件来完成，所述的计算机可读指令可存储于一计算机可读存储介质中，该计算机可读指令在被处理器执行时，可实现上述各个方法实施例的步骤。其中，所述计算机可读指令包括计算机可读指令代码，所述计算机可读指令代码可以为源代码形式、对象代码形式、可执行文件或某些中间形式等。所述计算机可读存储介质可以包括：能够携带所述计算机可读指令代码的任何实体或装置、记录介质、U盘、移动硬盘、磁碟、光盘、计算机存储器、只读存储器(ROM，Read-Only Memory)、随机存取存储器(RAM，Random Access Memory)、电载波信号、电信信号以及软件分发介质等。需要说明的是，所述计算机可读存储介质包含的内容可以根据司法管辖区内立法和专利实践的要求进行适当的增减，例如在某些司法管辖区，根据立法和专利实践，计算机可读存储介质不包括是电载波信号和电信信号。
以上所述实施例仅用以说明本申请的技术方案，而非对其限制；尽管参照前述实施例对本申请进行了详细的说明，本领域的普通技术人员应当理解：其依然可以对前述各实施例所记载的技术方案进行修改，或者对其中部分技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本申请各实施例技术方案的精神和范围，均应包含在本申请的保护范围之内。

Claims (20)

  1. 一种基于业务线的预测方法，其特征在于，所述预测方法包括：
    在对指定业务线进行业务预测时，获取该指定业务线对应的预测模型，以及本次预测的输入维度和输出维度；
    从数据仓库中获取满足所述输入维度的预测数据；
    通过所述预测模型采用蒙特卡罗模拟法和几何布朗运动对所述预测数据进行趋势分析，得到所述输出维度的预测值；
    根据所述预测值计算所述指定业务线在指定时间段内的任务总量和所需投入的人力数量；
    其中，所述预测模型根据业务类型划分为呼入预测模型和呼出预测模型，所述数据仓库由清洗处理后的预设历史时间内的通话数据、拨打名单数据构成。
  2. 如权利要求1所述的基于业务线的预测方法，其特征在于，在得到所述输出维度的预测值之后，所述预测方法还包括：
    获取所述预设历史时间内的营销活动和突发事件，确定所述营销活动和突发事件发生时的周日期；
    根据所述营销活动和突发事件的周日期对所述输出维度的预测值进行平滑处理，以消除所述营销活动、突发事件对所述预测值的干扰。
  3. 如权利要求2所述的基于业务线的预测方法，其特征在于，所述根据所述营销活动和突发事件的周日期对所述输出维度的预测值进行平滑处理，以消除所述营销活动、突发事件对所述预测值的干扰包括：
    遍历每一个输出维度，从所述输出维度的预测值中筛选出具有相同周日期的预测值作为基底数据；
    计算所述基底数据的平均值和标准差；
    计算每一个基底数据与所述平均值之间的差值，比对所述差值的绝对值与所述标准差；
    当所述差值的绝对值大于所述标准差时，若所述差值为正数则缩小所述差值对应的基底数据，若所述差值为负数则增大所述差值对应的基底数据。
  4. 如权利要求1所述的基于业务线的预测方法，其特征在于，所述根据所述预测值计算所述指定业务线在指定时间段内的任务总量和所需投入的人力数量包括：
    对所述指定业务线在指定时间段内的预测值进行求和，得到所述指定业务线在指定时间段内的任务总量；
    获取若干坐席人员的通话日时长、考勤数据，根据所述通话日时长、考勤数据计算每一个坐席人员的工作效率，求取所述工作效率的平均值得到折算率；
    获取标准工作时长，根据所述标准工作时长和折算率计算平均工作时长，求取所述任务总量与所述平均工作时长之间的商作为所需投入的人力数量。
  5. 如权利要求2或3所述的基于业务线的预测方法，其特征在于，所述根据所述预测值计算所述指定业务线在指定时间段内的任务总量和所需投入的人力数量包括：
    对所述指定业务线在指定时间段内的预测值进行求和，得到所述指定业务线在指定时间段内的任务总量；
    获取若干坐席人员的通话日时长、考勤数据，根据所述通话日时长、考勤数据计算每一个坐席人员的工作效率，求取所述工作效率的平均值得到折算率；
    获取标准工作时长，根据所述标准工作时长和折算率计算平均工作时长，求取所述任务总量与所述平均工作时长之间的商作为所需投入的人力数量。
  6. 一种基于业务线的预测装置，其特征在于，所述预测装置包括：
    第一获取模块，用于在对指定业务线进行业务预测时，获取该指定业务线对应的预测模型，以及本次预测的输入维度和输出维度；
    第二获取模块，用于从数据仓库中获取满足所述输入维度的预测数据；
    分析模块，用于通过所述预测模型采用蒙特卡罗模拟法和几何布朗运动对所述预测数据进行趋势分析，得到所述输出维度的预测值；
    计算模块，用于根据所述预测值计算所述指定业务线在指定时间段内的任务总量和所需投入的人力数量；
    其中，所述预测模型根据业务类型划分为呼入预测模型和呼出预测模型，所述数据仓库由清洗处理后的预设历史时间内的通话数据、拨打名单数据构成。
  7. 如权利要求6所述的基于业务线的预测装置，其特征在于，所述装置还包括：
    第三获取模块，用于在得到所述输出维度的预测值之后，获取所述预设历史时间内的营销活动和突发事件，确定所述营销活动和突发事件发生时的周日期；
    平滑处理模块，用于根据所述营销活动和突发事件的周日期对所述输出维度的预测值进行平滑处理，以消除所述营销活动、突发事件对所述预测值的干扰。
  8. 如权利要求7所述的基于业务线的预测装置，其特征在于，所述平滑处理模块包括：
    筛选单元，用于遍历每一个输出维度，从所述输出维度的预测值中筛选出具有相同周日期的预测值作为基底数据；
    统计处理单元，用于计算所述基底数据的平均值和标准差；
    比较单元，用于计算每一个基底数据与所述平均值之间的差值，比对所述差值的绝对值与所述标准差；
    平滑处理单元，用于当所述差值的绝对值大于所述标准差时，若所述差值为正数则缩小所述差值对应的基底数据，若所述差值为负数则增大所述差值对应的基底数据。
  9. 如权利要求6所述的基于业务线的预测装置，其特征在于，所述计算模块包括：
    总量计算单元，用于对所述指定业务线在指定时间段内的预测值进行求和，得到所述指定业务线在指定时间段内的任务总量；
    折算率计算单元，用于获取若干坐席人员的通话日时长、考勤数据，根据所述通话日时长、考勤数据计算每一个坐席人员的工作效率，求取所述工作效率的平均值得到折算率；
    人力计算单元，用于获取标准工作时长，根据所述标准工作时长和折算率计算平均工作时长，求取所述任务总量与所述平均工作时长之间的商作为所需投入的人力数量。
  10. 如权利要求7或8所述的基于业务线的预测装置，其特征在于，所述计算模块包括：
    总量计算单元，用于对所述指定业务线在指定时间段内的预测值进行求和，得到所述指定业务线在指定时间段内的任务总量；
    折算率计算单元，用于获取若干坐席人员的通话日时长、考勤数据，根据所述通话日时长、考勤数据计算每一个坐席人员的工作效率，求取所述工作效率的平均值得到折算率；
    人力计算单元，用于获取标准工作时长，根据所述标准工作时长和折算率计算平均工作时长，求取所述任务总量与所述平均工作时长之间的商作为所需投入的人力数量。
  11. 一种计算机可读存储介质，其上存储有计算机可读指令，其特征在于，该计算机可读指令被处理器执行时实现如下步骤：
    在对指定业务线进行业务预测时，获取该指定业务线对应的预测模型，以及本次预测的输入维度和输出维度；
    从数据仓库中获取满足所述输入维度的预测数据；
    通过所述预测模型采用蒙特卡罗模拟法和几何布朗运动对所述预测数据进行趋势分析，得到所述输出维度的预测值；
    根据所述预测值计算所述指定业务线在指定时间段内的任务总量和所需投入的人力数量；
    其中，所述预测模型根据业务类型划分为呼入预测模型和呼出预测模型，所述数据仓库由清洗处理后的预设历史时间内的通话数据、拨打名单数据构成。
  12. 如权利要求11所述的计算机可读存储介质，其特征在于，所述计算机可读指令被处理器执行时实现如下步骤：
    获取所述预设历史时间内的营销活动和突发事件，确定所述营销活动和突发事件发生时的周日期；
    根据所述营销活动和突发事件的周日期对所述输出维度的预测值进行平滑处理，以消除所述营销活动、突发事件对所述预测值的干扰。
  13. 如权利要求12所述的计算机可读存储介质，其特征在于，所述根据所述营销活动和突发事件的周日期对所述输出维度的预测值进行平滑处理，以消除所述营销活动、突发事件对所述预测值的干扰包括：
    遍历每一个输出维度，从所述输出维度的预测值中筛选出具有相同周日期的预测值作为基底数据；
    计算所述基底数据的平均值和标准差；
    计算每一个基底数据与所述平均值之间的差值，比对所述差值的绝对值与所述标准差；
    当所述差值的绝对值大于所述标准差时，若所述差值为正数则缩小所述差值对应的基底数据，若所述差值为负数则增大所述差值对应的基底数据。
  14. 如权利要求11所述的计算机可读存储介质，其特征在于，所述根据所述预测值计算所述指定业务线在指定时间段内的任务总量和所需投入的人力数量包括：
    对所述指定业务线在指定时间段内的预测值进行求和，得到所述指定业务线在指定时间段内的任务总量；
    获取若干坐席人员的通话日时长、考勤数据，根据所述通话日时长、考勤数据计算每一个坐席人员的工作效率，求取所述工作效率的平均值得到折算率；
    获取标准工作时长，根据所述标准工作时长和折算率计算平均工作时长，求取所述任务总量与所述平均工作时长之间的商作为所需投入的人力数量。
  15. 如权利要求12或13所述的计算机可读存储介质，其特征在于，所述根据所述预测值计算所述指定业务线在指定时间段内的任务总量和所需投入的人力数量包括：
    对所述指定业务线在指定时间段内的预测值进行求和，得到所述指定业务线在指定时间段内的任务总量；
    获取若干坐席人员的通话日时长、考勤数据，根据所述通话日时长、考勤数据计算每一个坐席人员的工作效率，求取所述工作效率的平均值得到折算率；
    获取标准工作时长，根据所述标准工作时长和折算率计算平均工作时长，求取所述任务总量与所述平均工作时长之间的商作为所需投入的人力数量。
  16. 一种终端，包括存储器、处理器及存储在存储器上并可在处理器上运行的计算机可读指令，其特征在于，所述处理器执行所述计算机可读指令时实现以下步骤：
    在对指定业务线进行业务预测时，获取该指定业务线对应的预测模型，以及本次预测的输入维度和输出维度；
    从数据仓库中获取满足所述输入维度的预测数据；
    通过所述预测模型采用蒙特卡罗模拟法和几何布朗运动对所述预测数据进行趋势分析，得到所述输出维度的预测值；
    根据所述预测值计算所述指定业务线在指定时间段内的任务总量和所需投入的人力数量；
    其中，所述预测模型根据业务类型划分为呼入预测模型和呼出预测模型，所述数据仓库由清洗处理后的预设历史时间内的通话数据、拨打名单数据构成。
  17. 如权利要求16所述的终端，其特征在于，所述处理器执行所述计算机可读指令时还实现如下步骤：
    获取所述预设历史时间内的营销活动和突发事件，确定所述营销活动和突发事件发生时的周日期；
    根据所述营销活动和突发事件的周日期对所述输出维度的预测值进行平滑处理，以消除所述营销活动、突发事件对所述预测值的干扰。
  18. 如权利要求17所述的终端，其特征在于，所述根据所述营销活动和突发事件的周日期对所述输出维度的预测值进行平滑处理，以消除所述营销活动、突发事件对所述预测值的干扰包括：
    遍历每一个输出维度，从所述输出维度的预测值中筛选出具有相同周日期的预测值作为基底数据；
    计算所述基底数据的平均值和标准差；
    计算每一个基底数据与所述平均值之间的差值，比对所述差值的绝对值与所述标准差；
    当所述差值的绝对值大于所述标准差时，若所述差值为正数则缩小所述差值对应的基底数据，若所述差值为负数则增大所述差值对应的基底数据。
  19. 如权利要求16所述的终端，其特征在于，所述根据所述预测值计算所述指定业务线在指定时间段内的任务总量和所需投入的人力数量包括：
    对所述指定业务线在指定时间段内的预测值进行求和，得到所述指定业务线在指定时间段内的任务总量；
    获取若干坐席人员的通话日时长、考勤数据，根据所述通话日时长、考勤数据计算每一个坐席人员的工作效率，求取所述工作效率的平均值得到折算率；
    获取标准工作时长，根据所述标准工作时长和折算率计算平均工作时长，求取所述任 务总量与所述平均工作时长之间的商作为所需投入的人力数量。
  20. 如权利要求17或18所述的终端，其特征在于，所述根据所述预测值计算所述指定业务线在指定时间段内的任务总量和所需投入的人力数量包括：
    对所述指定业务线在指定时间段内的预测值进行求和，得到所述指定业务线在指定时间段内的任务总量；
    获取若干坐席人员的通话日时长、考勤数据，根据所述通话日时长、考勤数据计算每一个坐席人员的工作效率，求取所述工作效率的平均值得到折算率；
    获取标准工作时长，根据所述标准工作时长和折算率计算平均工作时长，求取所述任务总量与所述平均工作时长之间的商作为所需投入的人力数量。
PCT/CN2018/077347 2017-07-26 2018-02-27 基于业务线的预测方法、装置、存储介质及终端 WO2019019633A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
US16/093,628 US20210224434A1 (en) 2017-07-26 2018-02-27 Service line-based predication method, device, storage medium and terminal
SG11201808507VA SG11201808507VA (en) 2017-07-26 2018-02-27 Service line-based predication method, device, storage medium and terminal

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN201710615821.X 2017-07-26
CN201710615821.XA CN108282586B (zh) 2017-07-26 2017-07-26 基于业务线的预测方法、装置、存储介质及终端

Publications (1)

Publication Number Publication Date
WO2019019633A1 true WO2019019633A1 (zh) 2019-01-31

Family

ID=62801163

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2018/077347 WO2019019633A1 (zh) 2017-07-26 2018-02-27 基于业务线的预测方法、装置、存储介质及终端

Country Status (4)

Country Link
US (1) US20210224434A1 (zh)
CN (1) CN108282586B (zh)
SG (1) SG11201808507VA (zh)
WO (1) WO2019019633A1 (zh)

Families Citing this family (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109672795B (zh) * 2018-11-14 2022-03-11 平安科技（深圳）有限公司 呼叫中心资源管理方法及装置、电子设备、存储介质
CN111327456A (zh) * 2020-01-21 2020-06-23 山东汇贸电子口岸有限公司 一种基于lstm的云计算资源管理方法和系统
CN111882338B (zh) * 2020-08-11 2023-06-30 网易（杭州）网络有限公司 在线人数的异常检测方法、装置及电子设备
CN113393047A (zh) * 2021-06-23 2021-09-14 中国工商银行股份有限公司 一种业务场景关键表预测方法及装置
CN114331147A (zh) * 2021-12-30 2022-04-12 贝壳找房网(北京)信息技术有限公司 一种服务的运营业务的实施效果评估方法及装置
CN117592769B (zh) * 2024-01-19 2024-04-05 四川绿豆芽信息技术有限公司 一种碳小屋站点管理方法及系统

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103095937A (zh) * 2012-12-14 2013-05-08 广东电网公司佛山供电局 基于话务预测的呼叫中心座席数量的预测方法
JP2015097334A (ja) * 2013-11-15 2015-05-21 Kddi株式会社 通信トラヒック予測装置およびプログラム
US9467567B1 (en) * 2014-04-03 2016-10-11 Amdocs Software Systems Limited System, method, and computer program for proactive customer care utilizing predictive models
CN106713677A (zh) * 2016-05-24 2017-05-24 国家电网公司客户服务中心 一种电力客户服务中心呼入话务量的预测方法

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN106682754A (zh) * 2015-11-05 2017-05-17 阿里巴巴集团控股有限公司 事件发生概率预测方法及装置

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN103095937A (zh) * 2012-12-14 2013-05-08 广东电网公司佛山供电局 基于话务预测的呼叫中心座席数量的预测方法
JP2015097334A (ja) * 2013-11-15 2015-05-21 Kddi株式会社 通信トラヒック予測装置およびプログラム
US9467567B1 (en) * 2014-04-03 2016-10-11 Amdocs Software Systems Limited System, method, and computer program for proactive customer care utilizing predictive models
CN106713677A (zh) * 2016-05-24 2017-05-24 国家电网公司客户服务中心 一种电力客户服务中心呼入话务量的预测方法

Also Published As

Publication number Publication date
SG11201808507VA (en) 2019-02-27
CN108282586B (zh) 2019-02-19
CN108282586A (zh) 2018-07-13
US20210224434A1 (en) 2021-07-22

Similar Documents

Publication Publication Date Title
WO2019019633A1 (zh) 基于业务线的预测方法、装置、存储介质及终端
US11367026B2 (en) Systems and methods for automatic scheduling of a workforce
US20180189743A1 (en) Intelligent scheduling management
US7853006B1 (en) Systems and methods for scheduling call center agents using quality data and correlation-based discovery
US7864946B1 (en) Systems and methods for scheduling call center agents using quality data and correlation-based discovery
CN107784404B (zh) 业务处理过程中进行告警方法及装置
CN103095937B (zh) 基于话务预测的呼叫中心座席数量的预测方法
US10178056B2 (en) Predicting and updating availability status of a user
US20150092936A1 (en) Request process optimization and management
US20080300955A1 (en) System and Method for Multi-Week Scheduling
US9372734B2 (en) Outage window scheduler tool
Hu et al. Prediction-driven surge planning with application in the emergency department
US20140358626A1 (en) Assessing the impact of an incident in a service level agreement
US20230121667A1 (en) Categorized time designation on calendars
CN109903079A (zh) 信息处理方法、设备及存储介质
CN113726961A (zh) 外呼数量的确定方法、装置、外呼系统及存储介质
CN110928748B (zh) 业务系统运行监测方法及装置
CN115330219A (zh) 一种资源调度的方法及装置
CN113487183A (zh) 垂直式业务场景中业务资源确定方法、装置及存储介质
US11775984B1 (en) System, method, and computer program for preempting bill related workload in a call-center
US11849064B1 (en) Techniques for detecting calling anomalies in inbound call traffic in telecommunications networks
CN114338429A (zh) 网络带宽的确定方法、装置及电子设备
US20210049531A1 (en) Automatic calendar event entry creation and modification
CN113988769B (zh) 一种智能匹配分布式资源的方法、装置和计算机设备
CN117667425A (zh) 一种异构云资源的配置优化方法及系统

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 18838675

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 18838675

Country of ref document: EP

Kind code of ref document: A1

32PN Ep: public notification in the ep bulletin as address of the adressee cannot be established

Free format text: NOTING OF LOSS OF RIGHTS PURSUANT TO RULE 112(1) EPC (EPO FORM 1205 DATED 05/08/2020)

122 Ep: pct application non-entry in european phase

Ref document number: 18838675

Country of ref document: EP

Kind code of ref document: A1