JP7831016B2 - Emergency employee call system and emergency employee call program - Google Patents

Description

This invention relates to an emergency employee call system and program for calling in employees and personnel to respond to emergencies such as earthquakes.

For example, if an earthquake of magnitude 4 or higher occurred at a dam site, or if the dam's seismometer detected a magnitude of 4 or higher, it was traditionally necessary to conduct an emergency inspection (primary inspection) of the dam within one hour. This meant that employees of the power company, etc., had to report to their offices to check for any abnormalities such as large cracks in the dam structure, or any abnormal water discharge from gates, etc., using surveillance cameras, and to check for any alarms or abnormalities in dam data in the dam monitoring and control system. Calling employees to conduct such inspections traditionally required contacting each department and employee individually to confirm their schedules and locations, which was time-consuming.

On the other hand, a system for creating call-out lists is known that automatically lists call-out employees based on their schedules, health condition, and familiarity with the work site, when extracting call-out employees from accident information (see, for example, Patent Document 1). This system manages employee information, including their schedules and health condition, and determines the priority of call-out employees based on accident data and employee information input from the power distribution automation system. It then creates a call-out list based on this determined priority.

Japanese Patent Publication No. 2012-242870

By the way, disasters such as earthquakes and typhoons can occur at any time, and can certainly happen outside of holidays or normal working hours. In such cases, employees must be selected to be called in in a way that does not exceed the maximum overtime hours stipulated in the labor-management agreement (36 Agreement) based on Article 36 of the Labor Standards Act. However, it is not easy for people and managers to consider these factors and quickly select the appropriate employees in an emergency. Furthermore, while the system described in Patent Document 1 determines the priority of employees who can be called in by considering their schedules and physical condition, it does not take working hours into consideration.

Therefore, the present invention aims to provide an emergency employee call system and an emergency employee call program that enable the appropriate and rapid selection of employees to be called in during emergencies.

To solve the above problems, the invention of claim 1 comprises: a communication terminal used by each employee; employee information storage means for storing employee information including the number of working hours and the days on which the employee came to work in an emergency ; employee selection means for inquiring about the availability of each employee to come to work from the communication terminal of each employee in the event of an emergency, and selecting employees to be asked to come to work from among those who have replied that they are available, based on the number of working hours ; and location information acquisition means for acquiring the location information of the employees who have replied that they are available to come to work , wherein the employee selection means selects employees whose number of working hours during a predetermined period does not exceed a predetermined upper limit of working hours, This emergency employee call system is characterized by comprehensively considering a second condition, which is to select employees so that their working hours are averaged among the aforementioned employees, and a third condition, which is to select employees who can meet the response deadline, which is the time it takes to complete the response after an emergency occurs. If there are multiple employees who can meet the response deadline, the system prioritizes selecting employees who arrive later but have fewer working hours than employees who arrive earlier but have more working hours than the aforementioned maximum working hours. Furthermore, even if an employee arrives later than other employees, the system also prioritizes selecting employees who have fewer emergency call-outs than other employees .

The invention of claim 2 is characterized in that, in the emergency employee call system described in claim 1 , the employee selection means uses a machine learning model for employee selection that has been trained on past performance data, such that when an employee who has replied that he or she is able to come to work is input, an employee to be requested to come to work is output.

The invention of claim 3 provides a computer that functions as an employee information storage means for storing employee information including the number of working hours for each employee and the days on which they came to work in an emergency , and an employee selection means for inquiring about the ability to come to work from the communication terminal used by each employee in the event of an emergency, and selecting employees to be asked to come to work from among those who have replied that they are able to come to work, based on the number of working hours , wherein the employee selection means has a first condition of selecting employees whose number of working hours during a predetermined period does not exceed a predetermined upper limit of working hours, and a second condition of selecting employees so that the number of working hours among the employees is averaged, and This emergency employee call-in program is characterized by comprehensively considering the location information of employees who have responded indicating they are available to come to work, a third condition of selecting employees who can meet the response deadline (the time by which a response must be completed after an emergency occurs), and, if there are multiple employees who can meet the response deadline, prioritizing the selection of employees who arrive later but have fewer working hours over employees who arrive earlier but have more working hours than the aforementioned maximum working hours, and also selecting employees who arrive later than other employees but have fewer emergency call-outs than the aforementioned employees.

The invention of claim 4 is characterized in that, in the emergency employee call program described in claim 3 , the employee selection means uses a machine learning model for employee selection that has been trained on past performance data, such that when an employee who has replied that he or she is able to come to work is input, an employee to be requested to come to work is output.

According to the inventions described in claims 1 and 3 , when an emergency occurs, each employee is automatically inquired about their ability to come to work, and employees to be requested to come to work are selected, making it possible to quickly select employees to call in during an emergency. Furthermore, since employees to be requested to come to work are selected from among those who have replied that they are able to come to work, it is possible to select appropriate employees that are in line with each employee's circumstances and situation at the time of the emergency (taking into account each employee's convenience). Moreover, since employees to be requested to come to work are selected based on the number of working hours of employees who have replied that they are able to come to work, it is possible to select appropriate employees that take into account the number of working hours of each employee.

According to the inventions described in claims 1 and 3 , employees are selected whose working hours during a predetermined period do not exceed a predetermined upper limit on working hours, making it possible to appropriately select employees to call in during emergencies. For example, even if an emergency occurs on a holiday or outside of normal working hours, it is possible to select employees whose working hours do not exceed the upper limit on overtime work stipulated in the 36 Agreement.

According to the inventions described in claims 1 and 3 , employees are selected in such a way that their working hours are averaged out among employees, thereby reducing unfairness in workload among employees and making it possible to appropriately select employees to call in during emergencies.

According to the inventions described in claims 1 and 3 , based on the location information of employees who have responded that they are able to come to work, employees who can meet the deadline for responding to an emergency are selected, making it possible to select appropriate employees who can respond by the deadline. For example, if an emergency inspection of a dam must be carried out within a certain time after an earthquake of a certain magnitude or higher occurs, it is possible to select appropriate employees who can meet the deadline, making it possible to carry out the work properly and reliably.

According to the inventions described in claims 2 and 4 , since the employees to be requested to come to work are output using a machine learning model for employee selection, it becomes possible to select more appropriate employees.

This is a schematic diagram showing an emergency employee call system according to an embodiment of the present invention. Figure 1 is a schematic block diagram of the management computer for the emergency employee call system. Figure 2 is a data configuration diagram of the employee information database on the management computer. Figure 2 is a functional block diagram showing the schematic configuration of the employee selection learning model for the management computer.

The present invention will be described below based on the illustrated embodiments.

Figure 1 is a schematic diagram showing an emergency employee call system 1 according to an embodiment of this invention. This emergency employee call system 1 is a system for selecting and calling employees M/personnel to respond in the event of an emergency such as an earthquake. This embodiment primarily describes the case where an employee M of a power company is called to inspect (respond to) a dam in the event of an earthquake near the dam. However, an emergency is not limited to an earthquake; it could also be during a typhoon or heavy rain. Furthermore, the equipment/facilities to be addressed are not limited to dams; they could be power generation facilities or transmission facilities, and the response could be limited to inspections, repairs, or customer service. Also, the employee M is not limited to a company member; it can be any person who constitutes an organization or group.

The emergency employee call system 1 comprises a communication terminal (location information acquisition means) 2 used by each employee M, and a management computer (employee information storage means, employee selection means) 3 installed in the company's facilities, and these systems are capable of free communication with each other.

Communication terminal 2 is a terminal equipped with communication functions, including smartphones (multifunctional mobile terminals) and computers, and also includes a function (location information acquisition means) for acquiring employee M's location information. Specifically, it is equipped with GPS (Global Positioning System), and in this embodiment, when employee M replies indicating they are able to come to work (as described later), the communication terminal 2, i.e., employee M's current location (latitude and longitude), is acquired, calculated, and transmitted to the management computer 3. Here, if GPS location information cannot be acquired or transmitted, as described later, it is also possible to acquire employee M's location information by reading the address/residence of employee M stored in address 3413.

Furthermore, this communication terminal 2 has an application program installed for operating the emergency employee call system 1, and can constantly receive notifications (inquiry notifications) from the management computer 3 via email, SNS (Social Networking Service), etc., asking about the employee's ability to come to work. Upon receiving an inquiry notification, the following response options are displayed on the touch panel or display.

"Safe and able to come to work" (Both the individual and their family are safe and able to come to work)
"Safe, but unable to come to work" (Both the employee and their family are safe, but they are unable to come to work.)
"Disability present, but able to come to work" (The individual or a family member has a disability, but is able to come to work.)
"Disability; unable to come to work" (The employee or a family member has a disability and is unable to come to work.)
Then, depending on the circumstances and situation of employee M at that time, the employee selects and clicks an answer from the options, and reply information including whether the employee and their family are safe and whether they are able to come to work is sent to the management computer 3. In this way, in this embodiment, not only is it possible to come to work, but also whether employee M and their family are safe (safety confirmation) can be notified to the management computer 3, that is, the company.

The management computer 3 is a computer used to select employees M to be requested to come to work. As shown in Figure 2, it mainly comprises an input unit 31, a display unit 32, a communication unit 33, a storage unit 34, an employee selection task (employee selection means) 35, a learning task 36, and a central processing unit 37 that controls these.

The input unit 31 is an interface for inputting various information and commands. Specifically, it is used to input employee information into the employee information database 341 (described later) and to input the activation command for the employee selection task 35. The display unit 32 is a display that shows various data and information. Specifically, it displays the identification information and call list of employees M selected by the employee selection task 35, as described later.

The communication unit 33 is an interface for communicating with the outside world via the internet, telephone network, etc. Specifically, it communicates with each communication terminal 2 and receives earthquake information and other emergency information from emergency information sources. That is, it receives weather and emergency information such as earthquakes, typhoons, tsunamis, warnings, and advisories from the Japan Meteorological Agency (emergency information source), and also receives seismic intensity, acceleration, etc. from seismometers (emergency information sources) installed around the dam.

The memory unit 34 primarily comprises an employee information database (employee information storage means) 341, an employee selection learning model 342, and an employee selection performance database 343. The employee selection learning model 342 and the employee selection performance database 343 will be described later. The employee information database 341 is a database that stores employee information, including the identification information and working hours of each employee M. As shown in Figure 3, for each employee ID 3411, the department 3412, address 3413, contact information 3414, working hours 3415, emergency response 3416, and other information 3417 are stored.

Employee ID 3411 stores employee M's identification information, such as employee number and name. Department 3412 stores the departments employee M currently and has previously belonged to. This information allows for the determination of what actions employee M can take in an emergency. Address 3413 stores employee M's address/residence, that is, the place where employee M is usually located on holidays or after returning to the office. This information allows for the determination of how long it will take for employee M to return to work in the event of an emergency on a holiday or other emergency. Therefore, reading this information also constitutes a location information acquisition means for obtaining employee M's location. Contact information 3414 stores contact methods for contacting employee M, such as telephone number and email address, and in particular, the IP address for sending and receiving the above-mentioned inquiry notifications and reply information.

The "Working Hours 3415" field stores information about employee M's working hours. Specifically, it stores the daily working hours from the past to the present, and can calculate the working hours for each fiscal year (including the current year), each month (including the current month), and each week (including the current week). The calculation results may also be stored. Furthermore, it can determine the days when M worked during emergencies or on holidays, and the working hours on those days, and these results may also be stored. In this embodiment, this information is obtained and stored from a server that manages working hours; however, in the employee selection task 35, this information may be read directly from the server instead of using "Working Hours 3415." The "Emergency Response 3416" field stores information such as the date (including whether it was a holiday or not), the working hours (including whether it was during normal working hours or not), and the details of the response when employee M responded to past emergencies.

The employee selection task 35 is a task program that, in the event of an emergency, inquires about the availability of each employee M to report to work via their communication terminal 2. From among the employees M who respond that they are available, it selects those employees M to report based on factors such as their working hours. Specifically, it is activated when the Japan Meteorological Agency (or its information distribution equipment, etc.) or seismometers around the dam receive notification of an earthquake of a predetermined seismic intensity (e.g., intensity 4) or higher observed near the dam. It then selects the identification information of employees M to conduct dam inspections. Here, for each type of emergency, the content of the response, the number of people required, the deadline for completing the response after the emergency occurs (sometimes there is no deadline), and the standard time required for the response are pre-set and stored in the system.

Specifically, first, based on the department 3412 of each employee M stored in the employee information database 341, employee M in the department responsible for dam inspection (for example, a technical department such as the civil engineering department) is extracted, and an inquiry notification is sent simultaneously to the communication terminal 2 of each extracted employee M. Then, from among the employee M who reply that they are able to come to work ("Safe and able to come to work" or "Unable to come to work, but can come to work"), employee M is selected in accordance with the following conditions.

Firstly, select employees M whose working hours during the specified period do not exceed the prescribed maximum working hours. That is, when requiring overtime work beyond statutory working hours or work on statutory holidays, the maximum overtime hours stipulated in the labor-management agreement (36 Agreement) based on Article 36 of the Labor Standards Act must not be exceeded. Therefore, employees M are selected so as not to exceed the maximum working hours stipulated in this 36 Agreement. Specifically, the 36 Agreement stipulates and stores the maximum overtime hours for one month and one year as a general rule, the maximum overtime hours for one month and one year in cases of temporary special circumstances, the maximum average hours over a 2-6 month period, and the number of months per year in which the maximum monthly hours can be exceeded as a general rule. Then, based on the information and working hours stored in the working hours 3415 of each employee M who has responded indicating their availability to come to work, employees M who satisfy these provisions are selected. At this time, even if the response is carried out based on the standard time required for the response, employees M who satisfy these provisions are selected.

Secondly, employees M are selected to ensure that their working hours are averaged out among all employees M. That is, employees M are selected to be requested to come to work in a way that prevents any particular employee M from having significantly more or less working hours than other employees M. Specifically, employees M with fewer overtime hours (less than the average) compared to other employees M, or employees M with fewer emergency call-outs compared to other employees M, are selected based on the information and working hours stored in each employee M's working hours 3415.

Thirdly, based on the location information of each employee M who has responded indicating their availability to come to work, employees M who can meet the response deadline (the time within which the response must be completed after the emergency occurs) are selected. That is, as described above, a response deadline is pre-set and stored for each response to an emergency, and employees M who can meet this deadline are selected based on the location information of each employee M obtained as described above. For example, if an earthquake occurs near a dam, it is pre-set and stored that an emergency inspection (primary inspection) of the dam must be conducted within one hour of the earthquake, a report/preliminary report must be sent to the designated reporting destination, and further, a prescribed inspection must be conducted and a report sent within three hours or 24 hours. Then, the arrival time to the dam is calculated based on each employee M's location information, and employees M who can meet these response deadlines are selected based on this arrival time and the pre-assumed and set response/inspection times. Therefore, even if there are multiple responses with different deadlines for the same emergency, different employees M may be selected for each response.

Then, based on a comprehensive assessment of these conditions, an employee M will be selected to be requested to come to work. For example, if there are multiple employees M who can meet the deadline, an employee M who arrives later but works fewer hours will be prioritized over an employee M who arrives earlier but whose working hours are close to the upper limit. Furthermore, an employee M who has fewer emergency visits than other employees M will be selected even if their arrival time is later than other employees M, as long as they can meet the deadline. Additionally, an employee M who responded "unaffected" may be prioritized over an employee M who responded "problem occurred."

This employee selection task 35 uses a machine learning model 342, which is based on past performance data, to output employee M to be requested to come to work when employee M's employee information and location information, which have responded indicating they are available to come to work, are input. This employee selection model 342 is created by the learning task 36.

In other words, the learning task 36 creates an employee selection learning model 342 using a known machine learning algorithm such as a neural network, based on past performance data recorded and stored in the employee selection performance database 343. This employee selection performance database 343 is a database that records and stores performance data, including the results of selecting employee M to be requested to come to work by experts and specialists in emergency response and labor management, based on the employee information and location information of employee M who responded indicating they were available to come to work as input information. The past performance data includes not only data created based on the actual employee information and location information of employee M, and the results of selections by experts and specialists in emergency response and labor management, but also data created through pre-training, etc.

As shown in Figure 4, this learning task 36 uses machine learning and deep learning with a neural network to create a neural network based on performance data recorded in the employee selection performance database 343. For example, the input layer contains employee information and location information of employee M who have responded indicating they are available to come to work, the output layer contains identification information of employee M who are being requested to come to work, and the hidden layer handles the analysis processing from the input layer to the output layer. Then, the learning task 36 uses the performance data of the employee selection learning model 342 as training data to learn various parameters in the hidden layer. In other words, the learning task 56 learns various parameters in the hidden layer so that employee M who are being requested to come to work are appropriately selected and output based on employee information and location information.

Then, including the identification information, telephone number, and email address of the selected employees M, a list of employees M to be called is created with priority assigned to those who meet the above conditions. This list may include all employees M who meet the above conditions, potentially exceeding the number of employees required for response. Alternatively, only the number of employees M required for response may be listed. This list can be displayed on the display unit 32 or transmitted to the designated administrator's communication terminal 2, allowing the administrator to contact the employees M and request them to report to work. Alternatively, the management computer 3 may automatically contact the employees M and request them to report to work.

As described above, with this emergency employee call system 1, when an emergency occurs, each employee M is automatically inquired about their ability to come to work, and employees M to be requested to come to work are selected. Therefore, it is possible to quickly select employees M to be called in during an emergency. Furthermore, since employees M to be requested to come to work are selected from among those who have replied that they are able to come to work, it is possible to select appropriate employees M that are appropriate to each employee's circumstances and situation (considering each employee's convenience) during an emergency. Moreover, since employees M to be requested to come to work are selected based on the working hours of employees M who have replied that they are able to come to work, it is possible to select appropriate employees M that take into account the working hours of each employee M.

In other words, since an employee M whose working hours during a specified period do not exceed the specified maximum working hours is selected, it becomes possible to appropriately select an employee M to call in during an emergency. For example, even if an emergency occurs on a holiday or outside of normal working hours, it is possible to select an employee M whose working hours do not exceed the maximum overtime hours stipulated in the 36 Agreement.

Furthermore, since employees M are selected in a way that averages their working hours, it reduces unfairness in workload among employees M, making it possible to appropriately select employees M to call in during emergencies.

Furthermore, based on the location information of employee M who has responded indicating they are available to come to work, an employee M who can meet the deadline for responding to the emergency is selected. This ensures that a suitable employee M capable of responding by the deadline is chosen. For example, if an emergency inspection of a dam needs to be conducted within a specified timeframe after an earthquake of a certain magnitude or higher, it is possible to select a suitable employee M who can meet the deadline, ensuring that the work is carried out appropriately and reliably.

Furthermore, since the employee M to be requested to come to work is output using the machine learning-based employee selection model 342, it becomes possible to select a more appropriate employee M.

The embodiments of this invention have been described in detail above, but the specific configuration is not limited to these embodiments, and any design changes, etc., that do not depart from the gist of this invention are also included. For example, in the above embodiment, the department 3412 of employee M determines whether or not they are capable of responding to an emergency, but it is also possible to store employee M's career history, skills, training history, etc., as employee information and make the decision based on this information. Furthermore, while employee M who meets the provisions of the 36 Agreement is selected, it is also possible to select employee M who does not exceed the maximum working hours set for each department or for each individual employee, instead of the 36 Agreement, or in addition to the 36 Agreement.

Alternatively, the emergency employee call system 1 and management computer 3 described above may be configured by installing the following emergency employee call program on a general-purpose computer. In other words, this emergency employee call program is characterized by having the computer function as an employee information storage means (employee information database 341) that stores employee information including the number of working hours for each employee M, and an employee selection means (employee selection task 35) that queries the communication terminal 2 used by each employee M in the event of an emergency to inquire about their ability to come to work, and selects employees M to be requested to come to work from among those employees M who have replied that they are able to come to work, based on their working hours. The employee selection means selects employees M whose working hours during a predetermined period do not exceed a predetermined upper limit of working hours, selects employees M so that their working hours are averaged among employees M, selects employees M who can meet the response deadline (the time that must be completed after the emergency occurs) based on the location information of employees M who have replied that they are able to come to work, and further uses an employee selection learning model 342 that has been machine-learned based on past performance data so that when the employee information and location information of employees M who have replied that they are able to come to work are input, the employee M to be requested to come to work is output.

1. Emergency employee call system 2. Communication terminal (means for acquiring location information)
3. Management computer 341. Employee information database (employee information storage means)
342. Learning Model for Employee Selection 35. Employee Selection Task (Method of Employee Selection)
M employee

Claims (4)

The communication terminals used by each employee,
An employee information storage means that stores employee information including the number of working hours for each employee and the days on which the employee came to work in an emergency ,
An employee selection method that, in the event of an emergency, inquires via the communication terminal of each employee whether they are able to come to work, and from among the employees who reply that they are able to come to work, selects employees to be asked to come to work based on the number of working hours,
A location information acquisition means for acquiring the location information of an employee who has responded that they are able to come to work,
Equipped with ,
The employee selection method comprehensively considers three conditions: firstly, that employees whose working hours during a specified period do not exceed a specified upper limit of working hours; secondly, that employees be selected so that their working hours are averaged out among them; and thirdly, that employees be selected to meet the response deadline, which is the time it takes to complete a response after an emergency occurs. If there are multiple employees who can meet the response deadline, priority is given to selecting an employee who arrives later but has fewer working hours than an employee who arrives earlier but whose working hours are close to the upper limit of working hours. Furthermore, even if an employee arrives later than other employees, priority is given to selecting an employee who has reported to work in emergencies fewer times than other employees.
An emergency employee call system characterized by the following features. The employee selection means uses a machine learning model for employee selection, which is based on past performance data, so that when employee information of an employee who has responded that they are available to come to work is input, the employee to be requested to come to work is output.
The emergency employee call system according to feature 1 . Computers,
An employee information storage means that stores employee information including the number of working hours for each employee and the days on which they came to work in an emergency ,
An employee selection method that, in the event of an emergency, inquires via the communication terminal used by each employee whether they are able to come to work, and from among the employees who reply that they are able to come to work, selects employees to be asked to come to work based on the number of working hours.
To make it function as ,
The employee selection method comprehensively considers the following three conditions: firstly, that employees whose working hours during a predetermined period do not exceed a predetermined maximum working hours; secondly, that employees be selected so that their working hours are averaged out among them; and thirdly, that employees be selected based on the location information of employees who have responded that they are available to come to work, so that they can meet the response deadline, which is the time it takes to complete the response after an emergency occurs. If there are multiple employees who can meet the response deadline, the method prioritizes selecting employees who arrive later but have fewer working hours than employees who arrive earlier but whose working hours are close to the maximum working hours. Furthermore, even if an employee arrives later than other employees, the method also prioritizes selecting employees who have come to work fewer times in emergencies than other employees.
An emergency employee call-out program characterized by the following features. The employee selection means uses a machine learning model for employee selection, which is based on past performance data, so that when employee information of an employee who has responded that they are available to come to work is input, the employee to be requested to come to work is output.
The emergency employee call program according to feature 3 .