JP2020102145A - Incident response man-hour prediction system and method for predicting incident response man-hours, and program - Google Patents

Abstract

To provide a method for predicting, when an incident occurs, response counts, response man-hours and a response time required till completion of the incident from the answer content of a questioner.SOLUTION: An incident response system 1 comprises an incident management system 50 and a response man-hour prediction system 60. A questioner 00 inquires of a person 20 in charge of service desk about various matters by electronic mail or telephone. The person 20 in charge registers the content of an inquiry accepted from the questioner 00 to the incident management system 50. The response man-hour prediction system 60 predicts the man-hours required for all steps needed for response till incident completion (response counts, response man-hours in each response, a response time for each response) in accordance with the content of answer to the accepted inquiry by the person 20 in charge.SELECTED DRAWING: Figure 1

Description

The present invention relates to an incident response man-hour prediction system, an incident response man-hour prediction method, and a program.

In Patent Document 1, the working time is calculated based on past incident information, the number of workers is calculated from the working time, and the worker in charge is determined and assigned from the number of workers in consideration of the working conditions of each worker. A management method is disclosed. According to this method, it is possible to predict the human resources required for the operation and management work of the system and appropriately allocate the personnel.

In Patent Document 2, a response request by an email sent by a customer is accepted, an expected response time required for a response with the customer is calculated based on a keyword specified in the email, and a response schedule of an operator is determined. A system for notifying a user of a time by e-mail is disclosed. According to this system, even when the call center cannot be reached due to the congestion of the telephone line, the customer can receive the callback from the operator at the scheduled time by sending the response request by e-mail. it can.

JP, 2005-118469, A JP, 2003-143301, A

In the actual incident response, the questioner often has to interact with the questioner many times and may be waiting for an answer from the questioner. When waiting for the reply, incident handling is interrupted and the man-hours of the assigned work become available. If the man-hours are free, the incident can be handled by another person in charge, but if the interruption of the incident cannot be predicted, the person in charge will be over-allocated. In addition, if the interrogator responds to the interrupted incident while responding to another incident, the interrupted incident may be restarted and the man-hours required for response may be insufficient. Patent Documents 1 and 2 do not disclose a method for predicting the number of man-hours for incident response in consideration of interruption of incident response due to waiting for an answer from a questioner.

Therefore, an object of the present invention is to provide an incident response man-hour prediction system, an incident response man-hour prediction method, and a program that solve the above problems.

According to one aspect of the present invention, the incident response man-hour prediction system is based on the content of the response to the question from the person in charge of incident response to the questioner of the incident, from the transmission of the response content to the completion of the incident. A prediction unit that predicts the man-hours of all the processes.

Further, according to another aspect of the present invention, based on the answer contents to the question from the person in charge of incident response to the questioner of the incident, all steps from the transmission of the answer contents to the completion of the incident It is an incident response man-hour prediction method for predicting man-hours of.

Further, according to another aspect of the present invention, a computer is used from the transmission of the response content to the completion of the incident, based on the response content to the question from the person in charge of incident response to the questioner of the incident. Is a program for functioning as a means for predicting the man-hours of all the processes.

According to the present invention, the number of man-hours required for incident response is predicted based on the content of the reply from the person in charge, so that highly accurate prediction can be performed.

It is a block diagram which shows an example of the incident response system by one Embodiment of this invention. It is a figure which shows the communication between the questioner and the person in charge in the incident by one Embodiment of this invention. It is a figure which shows an example of the incident table by one Embodiment of this invention. It is a figure which shows an example of the incident response log|history table by one Embodiment of this invention. It is a 1st figure which shows an example of the prediction know-how part by one Embodiment of this invention. It is a 2nd figure which shows an example of the prediction know-how part by one Embodiment of this invention. FIG. 6 is a first diagram showing an example of work assignment according to an embodiment of the present invention. FIG. 6 is a second diagram showing an example of work assignment according to an embodiment of the present invention. FIG. 6 is a third diagram showing an example of work assignment according to an embodiment of the present invention. 6 is a flowchart showing an example of man-hour prediction and work allocation processing required for incident response according to an embodiment of the present invention. It is a 1st figure explaining the example of work allocation change by one embodiment of the present invention. It is a 2nd figure explaining the example of work allocation change by one embodiment of the present invention. It is a 3rd figure explaining the example of work allocation change by one embodiment of the present invention. It is a 4th figure explaining the example of work allocation change by one embodiment of the present invention. 6 is a first flowchart showing an example of feedback processing of prediction know-how according to an embodiment of the present invention. 9 is a second flowchart showing an example of feedback processing of prediction know-how according to an embodiment of the present invention. It is a figure which shows the minimum structure of the incident management system by one Embodiment of this invention. It is a figure which shows an example of the hardware constitutions of the incident management system in one Embodiment of this invention.

Hereinafter, an incident management system according to an embodiment will be described with reference to FIGS. 1 to 18.
FIG. 1 is a block diagram showing an example of an incident handling system according to an embodiment of the present invention.
As shown in the figure, the incident handling system 1 includes an incident management system 50 and a handling man-hour prediction system 60. The interrogator 00 makes various inquiries to the person in charge 20 of the service desk by e-mail or telephone. The person in charge 20 registers the content of the inquiry received from the questioner 00 in the incident management system 50. The response man-hour prediction system 60 requires all processes (the number of times of handling, the number of handling steps in each response, and the response time of each response) required for the response to the completion of the incident, according to the content of the response from the person in charge 20 to the received inquiry. Predict man-hours. In addition, the response man-hour prediction system 60 performs work allocation of incident response work to the person in charge 20 based on the predicted man-hours. The support desk efficiently arranges the person in charge 20 based on the work assignment assigned by the corresponding man-hour prediction system 60.

Next, terms used in incident handling will be described with reference to FIG.
FIG. 2 is a diagram for explaining the communication between the inquirer and the person in charge in an incident according to the embodiment of the present invention.
FIG. 2 shows how the interrogator 00 and the person in charge 20 exchange. First, the inquiry person 00 transmits an inquiry to the person in charge 20. The person in charge 20 receives the inquiry. This causes an incident. The person in charge 20 confirms the content of the inquiry and responds to the questioner 00. If this answer solves the problem of the questioner 00, the incident is completed. However, if the questioner 00 has another question, or if the response from the person in charge 20 is such as "please send a log", the incident continues. The inquirer 00 transmits the confirmation result of the item requested by the person in charge 20 and the additional question item to the person in charge 20. From the point of view of the person in charge 20, this incident process cannot proceed until the questioner 00 receives an inquiry again after answering the questioner 00. This time is called "waiting time". In other words, the “waiting time” is the time from confirmation and answer to the questioner 00 until the answer comes from the questioner 00, in other words, the person in charge 20 responds once and then the response is made. It is time to restart.

Further, the questioner 00 making an inquiry and the person in charge 20 making an answer to it is called “round trip” or “correspondence”. That is, the “round trip” is a series of exchanges from the arrival of a question or answer from the questioner 00 until the person in charge 20 returns a confirmation or answer.
Further, the number of times the round trip is performed is referred to as the “round trip number” or the “corresponding number”. In addition, the response to the incident by the person in charge 20 is referred to as “response” or “work”.

Returning to FIG. 1, the incident handling system 1 will be described. The incident management system 50 includes a response history DB 51. The correspondence history DB 51 stores an incident table and an incident correspondence history table. Here, the incident table and the incident handling history table will be described.

FIG. 3 is a diagram showing an example of an incident table according to an embodiment of the present invention.
In the incident table, "incident number", "occurrence date and time", "close date and time", "question content", "desired delivery date", "round trips", "total man-hours", and "days until closing" for one incident Is registered. For example, when the person in charge 20 receives the first inquiry from the questioner 00 (when an incident occurs), the person in charge 20 registers the date and time of occurrence, the content of the question, and the desired delivery date in the incident management system 50. The incident management system 50 assigns an incident number and registers the data input by the person in charge 20 in the corresponding field of the incident table stored in the response history DB 51. Further, when the response to the incident is completed, the person in charge 20 manages the incident closing date and time, the number of round trips, the total man-hours (for example, two persons can handle three days in six days), and the number of days until closing. Register with the system 50. The incident management system 50 registers the data input by the person in charge 20 in the corresponding field of the incident table.

FIG. 4 is a diagram showing an example of the incident response history table according to the embodiment of the present invention.
In the incident handling history table, the handling contents for each round trip for each incident registered in the incident table illustrated in FIG. 3 are registered. Specifically, in the incident response history table, each information of “incident No”, “occurrence date and time”, “response date and time”, “within response (contents of response)”, and “man-hours” is registered. For example, the records of the first to fourth lines illustrated in FIG. 4 indicate four round trips to the inquiry content “incident system number is displayed.” with incident No=1. The person in charge 20 responds to the first inquiry from the questioner 00 with "Send the following log. XXXXXX", and returns the incident number, the date and time of the incident, the date and time of the response, the contents of the response, the man-hours to the incident management system 50. sign up. The incident management system 50 registers the data input by the person in charge 20 in the corresponding field of the incident response history table. Thereafter, each time one round trip for incident response is completed, data regarding the response is added to the incident response history table. As will be described later, the response man-hour prediction system 60 predicts the number of times of correspondence and man-hours in the subsequent correspondence based on the reply contents registered in the incident correspondence history table.

Returning to FIG. 1, the response man-hour prediction system 60 of the incident response system 1 will be described. The corresponding man-hour prediction system 60 includes a prediction unit 61, a prediction know-how unit 62, a schedule allocation unit 63, and an incident analysis unit 64. The schedule allocation unit 63 includes a setting file 65. The incident analysis unit 64 includes a setting file 66.
The prediction unit 61 predicts the number of times of subsequent handling, the number of handling steps, and the timing of handling based on the content of the reply from the person in charge 20 to the inquiry from the questioner 00.
The prediction know-how section 62 has a “delivery date know-how section” and a “reply content, round trip times and man-hour know-how section”. Here, these two know-how sections will be described.

FIG. 5 is a first diagram showing an example of a know-how section according to an embodiment of the present invention.
FIG. 5 shows an example of the “delivery date know-how section”. The delivery date know-how section is, for example, a table in which the keywords associated with the delivery date included in the question content shown in FIG. 5 are associated with the delivery date. Then, the prediction know-how section 62 sets the delivery date as the current day based on the delivery date know-how portion if, for example, the inquiry content includes the keyword "Urgent". Alternatively, if the inquiry content specifies a date, that date is set as the delivery date.

FIG. 6 is a second diagram showing an example of the know-how section according to the embodiment of the present invention.
FIG. 6 shows an example of “response content, number of round trips and man-hour know-how section”. As shown in FIG. 6, the “response contents, round trip times and man-hour know-how section” is for each of the reply contents by the person in charge 20, the round trip times until the incident is completed (response times), and the response times (how many times) Is a table in which the corresponding man-hours and the waiting time for each number of correspondence are associated with each other. As illustrated in FIG. 6, for the waiting time, the waiting time may be set for each number of times of correspondence for each delivery date set by the “delivery date know-how section”. The predictive know-how section 62, based on the registered content of the “response content, round trip times and man-hour know-how section” and the response content by the person in charge 20, provides information on the man-hours related to all the processes until the incident is completed when the response is made. Output. For example, when the response from the person in charge 20 is “Please send the following log”, the prediction know-how section 62 determines the number of round trips “2.7”, For, the prediction of "2" (hours) for the first correspondence, "1" (hours) for the second correspondence, and "3" (hours) for the third correspondence is output. Further, regarding the waiting time for the above-mentioned reply contents, the prediction know-how unit 62, if the delivery date is urgent, the first response is 1 (day), the second response is 3 (day), and the third response is 2 ( Output predictions such as (day). Note that the registered content of the “response content, number of round trips, and man-hour know-how section” may be information determined based on past incident response.
The above-described prediction unit 61 inputs the reply content of the person in charge 20 to the prediction know-how unit 62 and makes an inquiry about the number of times of correspondence. The prediction know-how section 62 outputs the predicted value of the number of times of correspondence, the corresponding number of man-hours, and the corresponding time by using the “delivery date know-how section” and the “response content, round trip times and man-hour know-how section”. The prediction unit 61 predicts the number of man-hours until the incident is completed, based on the prediction value output by the prediction know-how unit 62.

The schedule allocation unit 63 allocates the person in charge 20 regarding incident response, based on the prediction made by the prediction unit 61. An example of resource allocation is shown in FIGS.
FIG. 7 is a first diagram showing an example of work assignment according to an embodiment of the present invention.
FIG. 7 shows the work assignment to the person in charge A based on the prediction of the incident in charge of the person in charge A. The schedule allocation unit 63 allocates the work of the person A in charge based on the prediction of the prediction unit 61. First, the schedule allocation unit 63 calculates the response date and time from the incident response history table of the response history DB 51. Next, the schedule allocation unit 63 acquires the delivery date set based on the “delivery date know-how unit” of the prediction know-how unit 62. Here, as an example, the delivery date is “urgent”. Next, the schedule allocating unit 63 outputs the first waiting time (for example, 1 hour when the delivery date is urgent, which the prediction know-how unit 62 outputs on the basis of the “response contents, the number of round trips and the man-hour know-how unit” at the reply date and time. ) Is added to predict when the next question will come (response time). Further, the schedule allocation unit 63 adds the first man-hour (for example, 2H) output by the prediction know-how unit 62, and calculates the reply date and time when the person in charge A answers the interrogator 00 next time. Thereafter, similarly, the schedule allocating unit 63 calculates the waiting time until the next question comes by adding the second waiting time when the delivery date is urgent. Further, the man-hours of the second time are added to calculate the next reply date and time. The prediction know-how section 62 repeats the same processing as many times as the number of round-trips (rounding up to the decimal point is rounded up) based on the “contents of reply, the number of round trips and the man-hour know-how section”, and the person A in charge of the incident A in the frame 71 in FIG. Create a work schedule.

In addition, when the delivery date does not fall within the fixed time (for example, 17:00) of the work day having a delivery date, the scheduled allocation unit 63 may allocate the shortage of the work time as the highest priority work within the fixed time of the next day. Alternatively, the schedule allocation unit 63 may allocate the work for the current day even if it exceeds the fixed time. Such an allocation method is set in the setting file 65. In the setting file 65, for example, when the waiting time is 1H and the answer is made just before the scheduled time, the work is performed at a time (1H after) that exceeds the scheduled time of the day, or at the next morning. Can be set.

FIG. 8 is a second diagram showing an example of work assignment according to an embodiment of the present invention.
The schedule assignment unit 63 also assigns work to persons other than the person in charge A. FIG. 8 shows the result of the work assignment performed by the schedule assigning unit 63 for the personnel A, personnel B, and personnel C.

FIG. 9 is a third diagram showing an example of work assignment according to an embodiment of the present invention.
The schedule allocating unit 63 also allocates work to the persons A to C for incidents newly generated on and after the next day.
For example, the prediction unit 61 predicts the number of newly generated incidents based on the past response history accumulated in the response history DB 51. For example, the prediction unit 61 calculates the average number of incidents per response content per day in the past predetermined period, and predicts that the same number of incidents as the calculated average number of incidents will occur every day after the next day. Good. Alternatively, the average number of incidents per day for each month and each customer may be calculated, and it may be predicted that an incident of that value will occur. More specifically, the prediction unit 61 predicts the number of incidents, the number of man-hours, and the number of responses according to the delivery date as follows.
Expedited delivery times are x, response times are x, initial response time is x hours Normal delivery times are x, response times are x, initial response time is x hours

Then, the schedule allocation unit 63 executes the allocation of each work of A to C based on the prediction of the prediction unit 61, as described in FIG. This makes it possible to determine whether the number of persons in charge 20 increases or decreases. For example, when there is a shortage of personnel, when the instruction to add the person in charge D is given to the scheduled assignment unit 63, the scheduled assignment unit 63 adds the person in charge D and assigns the work to the person in charge D. Further, when an incident actually occurs on the next day or later and each person in charge A or the like makes a reply, the prediction unit 61 makes a prediction of man-hours and the like based on the contents of the reply, and the scheduled allocation unit 63 updates the work allocation. To do.
In addition, in FIG. 9, the number of newly generated incidents is described only until two days after the next day, but the prediction unit 61 predicts a few weeks and a few months ahead, and the scheduled allocation unit 63 calculates a number of weeks and several days. You may assign work up to a month ahead.

Next, the flow of processing in which the incident handling system 1 predicts the number of times to respond to the incident, the number of handling steps, and the corresponding time based on the response content of the person in charge 20 and allocates the person in charge 20 will be described with reference to FIG. 10. To do.
FIG. 10 is a flowchart showing an example of incident management processing according to the embodiment of the present invention.
When the inquirer 00 makes an inquiry to the person in charge 20 prior to the prediction and allocation of the corresponding man-hour prediction system 60, an incident occurs. The person in charge 20 responds. Specifically, the person in charge 20 inputs the question content “a system error is displayed.”, the desired delivery date “5/2”, etc., to the incident management system 50. The incident management system 50 registers the input information in the incident table of the correspondence history DB 51 (record of incident No10 in FIG. 3). The person in charge 20 sends the reply “Please send the following log” to the interrogator 00 in the first response. The person in charge 20 inputs the reply content and the reply date and time into the incident management system 50. The incident management system 50 registers the contents of correspondence in the incident correspondence history table of the correspondence history DB 51 (record of incident No. 10 in FIG. 4).

Then, the prediction unit 61 predicts the delivery date (step S11). Specifically, the prediction unit 61 extracts keywords related to the delivery date such as “desired delivery date” information and “urgent” in “question content” of the incident table of the response history DB 51 from the inquiry content, and delivers the delivery date to the prediction know-how unit 62. Contact us. The prediction know-how unit 62 sets the delivery date based on the “delivery date know-how unit” and outputs the delivery date to the prediction unit 61. The prediction unit 61 sets the delivery date information (“emergency”, XX days, etc.) output by the prediction know-how unit 62 as the delivery date. In addition, when the keyword regarding the delivery date is not included, the prediction unit 61 sets “ordinary” in the delivery date.

Next, the prediction unit 61 predicts the number of round trips, man-hours, and waiting time (step S12). Specifically, the prediction unit 61 outputs the delivery date predicted in step S11 to the prediction know-how unit 62 and the “response content (response content)” of the incident response history table of the response history DB 51 to inquire about the number of round trips and the like. .. The prediction know-how unit 62 searches the table illustrated in FIG. 6 (the answer contents, the number of round trips and the man-hour know-how unit) for a record in which the acquired answer contents and the value of the “answer contents keyword” field match. To do. When a matching record is found, the prediction know-how unit 62 outputs to the prediction unit 61 the number of round trips of the record, the man-hour for each correspondence count, and the waiting time for each inquiry count corresponding to the acquired delivery date. For example, the prediction know-how section 62 uses the number of round trips of the record that matches "Please send the following log""2.7" (times), the first man-hour "2" (time), the second man-hour "1". (Hours), if the third man-hour is “3” (hours), and the delivery time is “normal”, wait time for the first answer “3” (hours), second wait time “6” (hours), etc. The response time for each response frequency is calculated by adding the waiting time “3” for the first response to the first response date and time acquired and recorded in the response history DB 51. The prediction know-how unit 62 outputs the calculated response time and the like to the prediction unit 61. The prediction unit 61 predicts the number of round trips, man-hours, and response time required for future incident response according to the information acquired from the forecast know-how unit 62.
In addition, when a date is specified in the delivery date predicted by the prediction unit 61, the prediction unit 61 compares the number of days from the date and time of the incident to the date with a predetermined threshold value to determine the delivery date as “urgent”, Classify as "normal" or "slow".

Next, the schedule allocation unit 63 allocates the work schedule of the person in charge 20 (step S13). The schedule allocation unit 63 allocates the work to the person in charge 20 based on the number of round trips, the number of man-hours, and the waiting time predicted by the prediction unit 61 (FIGS. 7 to 9). The schedule allocating unit 63 may allocate the work for other incidents to the waiting time for the reply. For example, the schedule allocating unit 63 performs a process of allocating the work to the vacant time of each person in charge, in units of man-hours corresponding to the number of times of correspondence in the table of FIG. As a result, the work can be assigned in consideration of the interruption time of the incident process due to the waiting for the reply from the questioner 00.

(effect)
In an actual incident response, when the person in charge 20 requests confirmation of an inquiry, the inquirer 00 may wait for an answer. When waiting for a reply, incident handling is interrupted and the manpower of the assigned work becomes idle.However, in a general incident management system, the person in charge is assigned without considering this waiting time, so the waiting time The work to be performed is not understood and is wasteful, or when other work is performed by the person in charge 20, the subsequent schedule management may deviate from the schedule. Also, if a questioner responds while responding to another incident and resumes responding to the interrupted incident, unexpected work will occur, resulting in a shortage of man-hours for handling and lack of personnel. It can happen.
On the other hand, according to the response man-hour prediction system 60, since the reply waiting time is predicted and another work is assigned to the reply waiting time, the person in charge does not waste the waiting time and works on other incidents. It can be performed. Further, since the work related to the incident waiting for the reply is assigned after the predicted reply waiting time, it is possible to prevent a shortage of man-hours for handling and a shortage of persons in charge due to the reply coming at an unexpected timing.

Further, if the man-hours are available, the work of another person in charge can be carried out, but in a general incident management system, since the waiting time is not predicted, the person in charge may be over-allocated.
On the other hand, according to the response man-hour prediction system 60, based on the content of the reply from the person in charge 20, the number of times of correspondence, the number of man-hours for each number of times of correspondence, and the response waiting time for each number of times of correspondence are predicted, and the work is carried out to the person in charge 20. Since the work is assigned, the work can be assigned to each person in charge without excess or shortage of man-hours. Further, as a result of the work assignment, if the number of persons in charge is large, the number of persons in charge can be reduced, and if the number of persons in charge is insufficient, the number of persons in charge can be increased, so that the personnel can be efficiently arranged, for example, The personnel can be assigned such that the person in charge is large when the number of incidents to be dealt with is large and the person in charge is small when the number of incidents is small.

Further, the corresponding man-hour prediction system 60 predicts future man-hours and the like based on the “content of the answer” that the person in charge answers the question, not the “content of the question” sent from the questioner. For example, as a past response, the inquiry content: "A system error is displayed", the log output status: "A "DB connection error" has been output to the log", and the other status: "The network was disconnected" Is recorded in the incident response history table.
If the new inquiry is only the inquiry content: "A system error is displayed", the person in charge 20 asks the questioner 00 to send a log or check the status in order to check the log and other conditions. Need to make a request. If the new inquiry includes not only the content of the inquiry: "A system error is displayed" but also the information "A "DB connection error" has been output to the log", take action without requesting to send the log. May be able to complete. In this way, even if the event of the inquiry content is the same, the subsequent response and man-hours will change depending on whether or not there is other information. Therefore, the prediction unit 61 predicts future correspondence from the “answer content” instead of the “question content”. As a result, the accuracy of prediction can be improved.

(Review the forecast based on the contents of the second and subsequent answers)
In the above example, based on the response contents of the first response, the number of subsequent responses, the number of response man-hours, and the response time are predicted, and the work is assigned. However, the work assignment is performed based on the response contents of the second and subsequent responses. You may change it. An example of changing the work assignment based on the contents of the second reply is shown in FIGS. 11 and 12.
11 and 12 are first and second diagrams illustrating an example of changing work assignment according to an embodiment of the present invention.
FIG. 11 shows a result of the scheduled allocation unit 63 performing work allocation to the persons A to C based on the number of times of correspondence predicted by the prediction unit 61 based on the content of the first reply. The response 11a in FIG. 11 is the first response by the person A in charge of a certain incident α. Based on the contents of the first reply transmitted to the interrogator 00 at the end of the correspondence 11a, the work schedule 11b for the next day, the work schedule 11c for the next day and the work schedule 11d for the second and subsequent days are assigned.

It is assumed that the next day, the work schedule 11b is dealt with. For example, in this response, it is assumed that the person in charge A makes a second confirmation request to the interrogator 00. Then, the prediction unit 61 makes another prediction of the work schedule for the third time and thereafter based on the reply contents of the second time and the prediction know-how unit 62, and the schedule allocation unit 63 redistributes the work. FIG. 12 shows the result of the work allocation being redone by the schedule allocation unit 63. The work schedules 11c′ to 11e′ are work schedules newly assigned to the incident α. Further, in the reassignment of the work, the work schedule 12a originally assigned as the work of the person A in the next day in FIG. 11 is reassigned as the work of the person C in the next day in FIG.
Similarly, the prediction unit 61 may predict the subsequent man-hours and the like based on the reply contents for the reply contents after the third time, and the scheduled allocation unit 63 may update the work allocation based on the prediction.

(Adjustment of waiting time)
In the above description, the waiting time is predicted by referring to the waiting time for each number of times of delivery for each delivery date defined in “Response contents, round trip times and man-hours know-how section” of the prediction know-how section 62. It is possible to weight the waiting time based on the information such as the response history and the generation system. For example, when it is known from the past correspondence history that the respondent is the late-answering questioner 00, the prediction unit 61 determines that the waiting time registered in the “response content, number of round trips and man-hours know-how unit” is 1.5. The multiplied value may be used as the predicted value of the waiting time. Alternatively, when it is known that the source system is a system of high importance that is greatly affected by a long downtime, the predicting unit 61 causes the waiting unit registered in the “response content and round trip times and man-hour know-how unit”. A value obtained by multiplying the time by 0.8 may be used as the predicted value of the waiting time.

(Adjustment of man-hours)
In the above description, the man-hours are predicted with reference to the man-hours for each number of times of correspondence defined in the “response contents and round trip times and man-hour know-how department” of the prediction know-how section 62. , Manpower, knowledge, etc.) may be added to the man-hours. For example, when a person with high skill corresponds, the prediction unit 61 may set a value obtained by multiplying the man-hours registered in the “response content, number of round trips and man-hour know-how unit” by 0.8 as the man-hour predicted value. Alternatively, the value multiplied by 1.5 is used as the estimated value of the man-hour in the case of "Product A" in which the generation source system takes a long time to respond, and the value multiplied by 0.8 in the case of "Product B" which is relatively easy to handle. May be used as the estimated man-hour value.

(Adjustment of work allocation due to holidays)
Further, the corresponding man-hour prediction system 60 can change the work assignment according to the vacation of the person in charge 20.
FIG. 13 is a third diagram illustrating an example of changing work assignment according to the embodiment of the present invention.
The work schedules 15b, 15c, 15c' shown in FIG. 13 are work schedules assigned to the person A in charge based on the correspondence 15a. For example, when the person in charge A takes a leave on the day after the next day, the leave schedule of the person in charge A is input to the corresponding man-hour prediction system 60 in advance, and then the schedule allocation unit 63 sends the person A in charge of the work schedule 15c on the day after the next day. No assignment is made, and the work is assigned two days after the person A in charge. Alternatively, when the delivery date is imminent and it cannot be allocated two days after next, the scheduled allocation unit 63 may allocate the incident response work to another person in charge 20. For example, the schedule allocating unit 63 selects a person in charge B who has a work schedule for the next day within the scheduled time and has a margin, and allocates the work whose assignment to the person A is canceled to the person in charge B. When changing the person in charge of the incident and assigning the work, the scheduled assignment unit 63 performs the handover assignment. The takeover assignment means that work assignment is performed by multiplying the man-hours predicted by the prediction unit 61 by the "takeover coefficient" as the man-hours required for handing over to another person in charge. In the example of FIG. 13, in order to allocate from the person in charge A to the person in charge C, the work is assigned to the person in charge C with a man-hour of 2.5 h obtained by multiplying the original expected man-hour 2 h by 1.25. As a result, it is possible to secure a work transfer time from the person A to the person B.
Even when the worker A is not scheduled to take a leave and is to take a leave on the day due to poor physical condition or the like, the scheduled assignment unit 63 performs the same handover assignment and postpones the corresponding date and time. be able to.

(Assignment of work to available personnel)
Regardless of whether the person in charge A is on vacation, the corresponding man-hour prediction system 60 can take over and assign the work of the person in charge A to another person in charge.
FIG. 14 is a fourth diagram illustrating an example of changing work assignment according to an embodiment of the present invention.
It is assumed that the work schedule 16a shown in FIG. 14 is a work having a high priority, and the work schedule 16c is a work having a relatively low priority. The schedule assigning unit 63 assigns the work schedule 16a to the person in charge B in place of the work schedule 16c (takeover coefficient=1.25). As a result, the work schedule 16a can be completed several hours earlier than the original schedule. Further, the worker A can be assigned another work 16b having a high priority.
In this way, when another person in charge is vacant or there is an incident that should be prioritized, it is possible to take over and assign it and deal with it as soon as possible.

(Update forecasting know-how)
Next, the function of the incident analysis unit 64 will be described. The incident analysis unit 64 provides feedback to the “response content, round trip times, and man-hour know-how unit”.
FIG. 15 is a first flowchart showing an example of the feedback process of the prediction know-how according to the embodiment of the present invention.
The person in charge 20 responds to the incident according to the schedule assigned by the schedule assigning unit 63, and inputs the result to the incident management system 50. The incident analysis unit 64 extracts the information registered in the incident response history table of the response history DB 51 for each “incident No” (step S21). Next, the incident analysis unit 64 compares the extracted data with the registered contents of the “response contents and the number of round trips and man-hours know-how unit” of the prediction know-how unit 62 (step S22). Then, the incident analysis unit 64, based on the comparison result, if there is a content to be updated in any of the number of round trips, the number of handling steps for each response number, and the waiting time for each number of response times, the response content and the number of round trips and the number of steps are updated. The registered contents of the "know-how department" are updated (step S23). For example, in response to an incident whose response content is "Please send the log below", the response time is "2" (hours) until the first response when the delivery time is "normal" If so, the waiting time until the first reply is updated from "3" to "2". Alternatively, the incident analysis unit 64 calculates the average waiting time until the first response in the incident whose response content in the past response is “Please send the log below”, and the average value is registered now. If the calculated value is different from the current value, the current registered content may be updated with the calculated average value.
By the processing of the incident analysis unit 64, it is possible to maintain the accuracy of the registered contents of the “response contents, round trip times and man-hour know-how department”. It should be noted that the feedback processing by the incident analysis unit 64 may be performed at predetermined time intervals, or may be performed each time one incident is completed, with respect to the registered content related to the incident.

(Addition of know-how)
In addition, the incident analysis unit 64 discovers new answer contents useful for predicting the number of incident response man-hours, and displays the found answer contents, the number of times of response to the answer contents, the number of man-hours, and the time of response as “answer content, number of round trips and man-hours”. It has the function of registering in the “know-how department”.
FIG. 16 is a second flowchart showing an example of update processing of prediction know-how according to an embodiment of the present invention.
The incident analysis unit 64 extracts the response history information registered in the incident response history table of the response history DB 51 (step S31). Next, the incident analysis unit 64 extracts, from the extracted data, a word that frequently appears in the content of the first reply and information regarding correspondence thereafter (step S32). The content of the first response is the value of the “response content (corresponding content)” field of the record with the oldest response date in each “incident No” record. The incident analysis unit 64 extracts frequently-used words that are not registered in the “response content, round trip count, and man-hour know-how unit”. In addition, the incident analysis unit 64 waits from the time when the first response is transmitted to the time when a response arrives for the incident in which the extracted frequently-used word is included in the response content, the number of response times after that, and the wait time for each response time. , Analyze the man-hours for each number of correspondence. The incident analysis unit 64 sends a sentence including the extracted frequently-used word and the word included in the question content (for example, the frequently-used word is “operation result” and the sentence is “send operation result below”) to the person in charge 20. Present. The person in charge 20 judges whether or not the presented sentence may be newly registered, and inputs the judgment result to 60. When the judgment result of the person in charge 20 is “new registration”, the incident analysis unit 64 sets the sentence including the extracted frequently-used word, the number of subsequent responses, the waiting time for each number of responses, and the man-hour for each number of responses “ Register the answer contents, round trip times and man-hour know-how section. When the judgment result of the person in charge 20 is “not newly registered”, the incident analysis unit 64 does not newly register the data regarding the frequently-used word. As a result, it is possible to make know-how about the exchange of incidents for which the subsequent man-hours can be estimated based on the content of the first response.
The reason for confirming the registration with the person in charge 20 is to confirm whether or not information that should not be registered, such as the name of the person in charge, is erroneously extracted. For example, the setting file 66 included in the incident analysis unit 64 may be set to perform new registration without confirmation of the person in charge 20.

FIG. 17 is a diagram showing the minimum configuration of the incident management system according to the embodiment of the present invention.
As shown in FIG. 17, the incident response man-hour prediction system 100 includes at least a prediction unit 101.
The prediction unit 101 predicts the man-hours of all steps of the incident from the transmission of the response to the completion of the incident, based on the content of the reply from the person in charge 20 to the question from the questioner 00 of the incident. More specifically, the prediction unit 101 predicts the number of times of handling until the completion of an incident, the number of handling steps in each handling, and the handling time when each handling occurs. The prediction unit 101 predicts the next response time based on the waiting time from when the incident person 20 responds to the questioner 00 for a response until the response is transmitted from the questioner 00. To do.

FIG. 18 is a diagram showing an example of the hardware configuration of the incident handling system in the embodiment of the present invention.
The computer 900 includes a CPU 901, a main storage device 902, an auxiliary storage device 903, an input/output interface 904, and a communication interface 905. Each of the incident management system 50 and the corresponding man-hour prediction system 60 described above is implemented in the computer 900. Then, each operation described above is stored in the auxiliary storage device 903 in the form of a program. The CPU 901 reads the program from the auxiliary storage device 903, expands the program in the main storage device 902, and executes the above processing according to the program. Further, the CPU 901 secures a storage area in the main storage device 902 according to the program. Further, the CPU 901 secures a storage area for storing data being processed in the auxiliary storage device 903 according to the program.

Note that, in at least one embodiment, the auxiliary storage device 903 is an example of a non-transitory tangible medium. Other examples of non-transitory tangible media include magnetic disks, magneto-optical disks, CD-ROMs, DVD-ROMs, semiconductor memories, etc. connected via the input/output interface 904. Further, when this program is distributed to the computer 900 via a communication line, the computer 900 that receives the distribution may expand the program in the main storage device 902 and execute the above processing. Further, the program may be a program for realizing some of the functions described above. Further, the program may be a so-called difference file (difference program) that realizes the function described above in combination with another program already stored in the auxiliary storage device 903.
Further, each of the incident management system 50 and the corresponding man-hour prediction system 60 may be composed of a plurality of computers 900.

In addition, it is possible to appropriately replace the constituent elements in the above-described embodiments with known constituent elements without departing from the spirit of the present invention. Further, the technical scope of the present invention is not limited to the above-described embodiment, and various modifications can be made without departing from the spirit of the present invention. The response man-hour prediction system 60 is an example of an incident response man-hour prediction system.

1... Incident response system 50... Incident management system 51... Response history DB
60... Corresponding man-hour prediction system 61... Prediction unit 62... Prediction know-how unit 63... Schedule allocation unit 64... Incident analysis unit 65, 66... Setting file

Claims (10)

A prediction unit that predicts the man-hours of all processes from the transmission of the response contents to the completion of the incident, based on the response contents to the questioner of the incident from the person in charge of incident response,
Incident response man-hour forecasting system. The prediction unit predicts the number of times of handling until the completion of the incident, the number of handling steps, and the handling time,
The incident response man-hour prediction system according to claim 1. Corresponding, based on the response history to past incidents, the response content, the response frequency, the response man-hour in each response included in the response frequency, and the response time of each response included in the response frequency. Prediction know-how department that stores the attached information,
The incident response man-hour prediction system according to claim 2, further comprising: Comparing incidents related to the prediction of the prediction unit, the actual number of times of handling until the completion of the incident, the number of handling steps in each handling included in the number of handlings, and the response time of each response included in the number of handlings Then, the incident analysis unit that updates the information stored in the prediction know-how unit,
The incident response man-hour prediction system according to claim 3, further comprising: The incident analysis unit, based on the response history of past incidents, included in the response content, extracts a word that affects the incident response after the response, the response content including the word, and the response after that Of the corresponding number of times, the corresponding man-hours, and the corresponding period of time are associated with each other and registered in the prediction know-how section.
The incident response man-hour prediction system according to claim 4. The prediction unit predicts the urgency of the response time based on a keyword related to the delivery date included in the question, and predicts the response time according to the urgency.
The incident response man-hour prediction system according to any one of claims 2 to 5. Based on the prediction of the prediction unit, further comprises a scheduled allocation unit that allocates a person in charge of incident response,
The incident response man-hour prediction system according to any one of claims 1 to 6. When assigning the incident response that the first person has responded to to the second person, the scheduled assignment unit adds the work time taken from the first person to the second person. , The second person in charge calculates man-hours corresponding to the incident,
The incident response man-hour prediction system according to claim 7. Based on the answer contents to the questioner of the incident from the person in charge of incident response, predict the man-hours of all steps from the transmission of the answer contents to the completion of the incident,
Incident response man-hour prediction method. Computer,
Means for predicting the man-hours of all steps from the transmission of the response contents to the completion of the incident, based on the response contents to the questioner of the incident from the person in charge of incident response,
Program to function as.

Images