JP7362015B2 - Dispatch planning system, control method, and program - Google Patents

Description

The present invention relates to a system, a control method, and a program for formulating a dispatch plan for dispatching workers to a restoration target where an event requiring restoration work has occurred.

Patent Document 1 discloses a wide-area disaster recovery support system for elevators that can predict elevator stoppages due to earthquakes.

Japanese Patent No. 6122787

When a disaster such as an earthquake causes a large number of elevator equipment to stop or break down at the same time in a wide area, a dispatch plan that specifies the targets to be dispatched to each work group and the order in which they are dispatched. The dispatch of each work group will be arranged in accordance with the above. Here, there are some types of elevator equipment whose operating status can be grasped in real time through remote monitoring, and there are other elevator equipment whose operating status cannot be grasped until they are contacted by telephone or the like because they are not subject to remote monitoring. Therefore, after a disaster occurs, especially in the early stages, it is difficult to quickly grasp the overall picture of the stopped elevator equipment, and dispatch plans drawn up at this stage may lack accuracy.

One aspect of the present invention aims to formulate a highly accurate dispatch plan that takes into consideration the operating status of equipment not subject to remote monitoring at the time of a disaster.

In order to solve the above-mentioned problem, a dispatch planning system according to one aspect of the present invention acquires the operating status from a monitoring target facility, which is the elevator facility whose operating status can be remotely monitored among a plurality of elevator facilities located at different locations. an operation status acquisition unit; an operation status estimation unit that estimates the operation status of non-monitored equipment, which is the elevator equipment whose operation status cannot be remotely monitored among the elevator equipment; and a recovery unit that identifies restoration targets from among the elevator equipment. An optimal solution or standard of a dispatch plan that determines for each group the combination of dispatch targets that are the recovery targets to which a group consisting of a target identification unit and one or more workers who perform recovery work is scheduled to dispatch, and the order of dispatch. a search unit that searches for an optimal dispatch plan that is an optimal solution, and the recovery target identification unit identifies the recovery target from among the monitored equipment according to the operating status obtained when a disaster occurs. At the same time, when the disaster occurs, the restoration target is specified from among the non-monitored equipment according to the estimated operating status based on the degree of the disaster.

In order to solve the above-mentioned problems, a control method according to one aspect of the present invention is a control method executed by one or more information processing devices, the control method remotely controlling the operating status of a plurality of elevator equipment located at different locations. an operation status acquisition step of acquiring the operating status from the equipment to be monitored, which is the elevator equipment that can be monitored; and an operation status estimation step of estimating the operating status of equipment other than the equipment to be monitored, which is the elevator equipment whose operating status cannot be remotely monitored among the elevator equipment. A combination of an estimation process, a recovery target identification process of identifying a recovery target from among the elevator equipment, and a dispatch target that is the recovery target to which a group consisting of one or more workers and performing restoration work is scheduled to be dispatched; The recovery target identification step includes a search step of searching for an optimal dispatch plan that is an optimal solution or a semi-optimal solution of the dispatch plan in which dispatch order is determined for each group, and the recovery target identification step includes the operation status acquired when the disaster occurs. The restoration target is specified from among the monitored equipment according to the disaster, and the restoration target is identified from among the non-monitored equipment according to the estimated operating status when the disaster occurs.

The dispatch planning system according to each aspect of the present invention may be realized by a computer, and in this case, the dispatch planning system is realized by the computer by operating the computer as each part (software element) included in the dispatch planning system. A program for a dispatch planning system and a computer-readable recording medium on which the program is recorded also fall within the scope of the present invention.

According to one aspect of the present invention, it is possible to formulate a highly accurate dispatch plan that takes into account the operating status of equipment not subject to remote monitoring at the time of a disaster.

1 is a block diagram showing an example of a schematic configuration of a learning model generation device according to Embodiment 1 of the present invention. FIG. FIG. 3 is a diagram showing an example of data stored in a disaster record database. It is a figure showing an example of the data structure of seismic intensity of each place. FIG. 3 is a diagram showing an example of a data structure of shaking characteristics. It is a figure which shows the example of the data memorize|stored in the elevator equipment installation information database. It is an image diagram showing the positional relationship between the epicenter of an earthquake that occurred in the past and elevator equipment. FIG. 3 is a diagram showing an example of data stored in a driving situation information database. FIG. 3 is a diagram showing an example of a data structure of driving conditions. 3 is a flowchart illustrating an example of the flow of learning model generation processing by the learning model generation device. FIG. 2 is a block diagram showing an example of a schematic configuration of a learning model generation device according to Embodiment 2 of the present invention. It is a figure which shows the example of the data memorize|stored in the elevator equipment installation information database. It is an image diagram showing the positional relationship between the epicenter of an earthquake that occurred in the past and elevator equipment. FIG. 3 is a diagram showing an example of data stored in a driving situation information database. It is a block diagram showing an example of a schematic structure of a driving situation estimation device concerning Embodiment 3 of the present invention. FIG. 2 is a diagram showing an example of a data structure of disaster occurrence information distributed from a disaster occurrence information distribution device. It is a figure showing an example of the data structure of seismic intensity of each place. FIG. 3 is a diagram showing an example of a data structure of shaking characteristics. It is a figure showing an example of location information, building information, and equipment information of attention elevator equipment. It is an image diagram showing the positional relationship between the epicenter of the latest earthquake and elevator equipment. It is a flowchart which shows an example of the flow of the process of estimating the operating status of the attention elevator equipment by the operating status estimating device. It is a figure showing an example of an estimation result outputted from a driving situation estimation device. It is a block diagram showing an outline of a maintenance system concerning Embodiment 4 of the present invention. It is a flowchart which shows the process performed by the dispatch support system based on Embodiment 4 of this invention. It is a block diagram showing principal part composition of a dispatch planning system concerning Embodiment 4 of the present invention. FIG. 3 is a diagram illustrating an example of a data structure of a recovery target list. It is a figure which shows an example of the procedure of a search performed by a search part. It is a figure which shows an example of the procedure of a search performed by a search part. It is a figure showing an example of the data structure of an optimal dispatch plan. It is a figure showing an example of a secondment plan screen. FIG. 1 is a block diagram showing an example of the hardware configuration of a system or device according to the present invention.

[Embodiment 1]
Hereinafter, one embodiment of the present invention will be described in detail with reference to FIGS. 1 to 9.

(Configuration of learning model generation device 4)
First, FIG. 1 shows the configuration of the learning model generation device 4 that performs machine learning to generate a learning model 451 that outputs the operating status of the elevator equipment using the location information of the elevator equipment and disaster information as input data. I will explain using FIG. 1 is a block diagram showing an example of a schematic configuration of a learning model generation device 4 according to Embodiment 1 of the present invention. Here, the elevator equipment is typically a passenger conveyor, an elevator, etc.

The learning model generation device 4 may include a control section 40 and a storage section 45. The control unit 40 may be, for example, a CPU (Central Processing Unit). The control unit 40 controls the learning model generation device 4 to execute processing of each function. The storage unit 45 is a storage device that stores various computer programs read by the control unit 40, data used in various processes executed by the control unit 40, and the like.

In this specification, the learning model generation device 4 that can generate teacher data for generating the learning model 451 will be described as an example, but the present invention is not limited thereto. For example, the learning model generation device 4 may be configured to store teacher data (not shown) generated in advance by another device in the storage unit 45, and use this to generate the learning model 451. . When this configuration is adopted, the learning section 42 may read and use the teacher data stored in the storage section 45. Therefore, in the learning model generation device 4, the teacher data generation section 41, which will be described later, is not an essential component.

The control section 40 includes a teacher data generation section 41 and a learning section 42.

The learning unit 42 generates a learning model 451 that outputs the operating status of the elevator equipment of interest using the location information of the elevator equipment of interest and disaster information as input data by performing machine learning using teacher data. The teacher data includes location information indicating the location of the elevator equipment, disaster information including the degree of disaster at the location, and operating status information indicating the operating status of the elevator equipment when a disaster occurs. Here, the term "elevator equipment of interest" refers to the elevator equipment whose operation status is to be estimated by the learning model when a disaster occurs.

The degree of disaster is determined by (1) the seismic intensity at the location of the elevator equipment, or (2) the magnitude of the earthquake at the location of the elevator equipment, which is calculated from the magnitude of the epicenter, the depth of the epicenter, and the distance from the epicenter. It may be the size.

The teacher data generation unit 41 generates teacher data for generating the learning model 451. For example, the teacher data generation unit 41 may use disaster records regarding past disasters acquired from the disaster record database 6a to generate disaster information including the degree of disaster at the location of the elevator equipment. The disaster record database 6a may be, for example, a database managed by the Japan Meteorological Agency. Further, the teacher data generation unit 41 may acquire location information indicating the location of the building in which the elevator equipment is installed (that is, the location of the elevator equipment) from the elevator equipment installation information database 6b. The elevator equipment installation information database 6b may be, for example, a database managed by a business that undertakes the installation of elevator equipment. Further, the teacher data generation unit 41 may acquire operating status information indicating the operating status of the elevator equipment when a past disaster occurred from the operating status information database 6c. The operating status information database 6c may be, for example, a database managed by a business that undertakes management and inspection of elevator equipment.

Hereinafter, disaster information, location information, and driving status information will be explained using specific examples.

[Disaster Information]
First, disaster information will be explained using FIG. 2 while referring to FIGS. 3 and 4. FIG. 2 is a diagram showing an example of data stored in the disaster record database 6a. Here, the explanation will be given using an earthquake as an example, but the present invention is not limited to this. The disaster may be, for example, heavy rain, typhoon, tsunami, volcanic activity, etc. These can be disasters that may cause the power supply to the elevator equipment to stop or the elevator equipment to come to an emergency stop. For each disaster, the disaster record database 6a records data related to the date and time of occurrence, epicenter name, maximum seismic intensity, epicenter and depth, seismic intensity 61 of each place, and shaking characteristics 62 in association with each other. Note that, as shown in FIG. 2, a disaster ID, which is a unique identification number, may be assigned to each disaster. The disaster ID may be a name (nickname) given to each disaster. For example, if the disaster is a typhoon, a unique name may be given to each typhoon.

Here, a more specific explanation will be given by taking as an example an earthquake to which a disaster ID of "EQ2003**" is assigned in FIG. 2 (hereinafter referred to as earthquake EQ2003**). Earthquake EQ2003** is an earthquake that occurred "off the coast of Z prefecture" on "March 25, 2020." The maximum seismic intensity of Earthquake EQ2003** is "4", and the magnitude indicating the scale of the earthquake is "6.2". The epicenter of Earthquake EQ2003** is "Zzz degrees north latitude, Zzz degrees east longitude", and the depth of the epicenter is "40 km".

The seismic intensity 61 in each region may be information on the seismic intensity observed at observation points in each region when the earthquake EQ2003** occurred, and as shown in Figure 3, the seismic intensity observed in each region and the observation point. It may also be associated with a city, ward, town, or village. According to Figure 3, a maximum seismic intensity of 4 was observed in "Prefecture Z** City" and "City A, Prefecture Z", etc., and a seismic intensity of 3 was observed in "City D, Prefecture Z" and "City B, Prefecture Z". .

The shaking characteristics 62 may include information indicating the direction of shaking, acceleration, and period of shaking observed in earthquake EQ2003**. As shown in FIG. 4, in the shaking characteristics 62, the acceleration of shaking observed when earthquake EQ2003** occurred is recorded for each of the north-south component, east-west component, and vertical component. In the example shown in FIG. 4, the shaking characteristic 62 includes the combined maximum acceleration of shaking calculated based on the north-south component “**Gal”, the east-west component “**Gal”, and the vertical component “**Gal”. **Gal)' is also recorded. The shaking characteristics 62 may include the magnitude (amplitude) of the observed shaking, the period of the shaking, and the like. In the example shown in FIG. 4, it is recorded in the shaking characteristic 62 that the shaking period is "0.8 sec." Note that the shaking characteristics 62 may include the shaking characteristics in each area where shaking was observed, recorded for each area.

Some elevator equipment may be provided with a shaking sensor that detects the magnitude of shaking and the acceleration of shaking. Therefore, the teacher data generation unit 41 may use the magnitude of the shaking and the acceleration of the shaking detected by the shaking sensor provided in the elevator equipment as they are as the degree of the disaster. On the other hand, in the case of elevator equipment that is not provided with such a sensor, the teacher data generation unit 41 may calculate the degree of disaster based on the disaster record and location information described below.

[Location information]
Next, location information will be explained using FIG. 5 with reference to FIG. 6. FIG. 5 is a diagram showing an example of data stored in the elevator equipment installation information database 6b. In the elevator equipment installation information database 6b, the location of the building where the elevator equipment is installed is recorded in association with each installed elevator equipment. Note that, as shown in FIG. 5, each installed elevator equipment may be given an elevator equipment ID, which is a unique identification number. According to Figure 5, the elevator equipment whose elevator equipment ID is "EV001" (hereinafter referred to as elevator equipment EV001) is installed in "B building" of "1-chome, A-cho, *city, Z prefecture, address *". has been done. The elevator equipment whose elevator equipment ID is "EV026" (hereinafter referred to as elevator equipment EV026) is installed at "L Heights" in "2-chome, D City, Z Prefecture *No.*".

FIG. 6 is an image diagram showing the positional relationship between the epicenter of an earthquake that occurred in the past and elevator equipment. Figure 6 shows the epicenter of Earthquake EQ2003** (see Figure 2) (indicated by "X" in the figure), as well as "Z Prefecture ** City" and "Z Prefecture ** City" facing the "XX Sea". "Z Prefecture D City" and "Z Prefecture F City" on the inland side of "Z Prefecture F City" are shown. Further, FIG. 6 shows the locations of the elevator equipment EV001 and the elevator equipment EV026. Note that major roads are also illustrated in FIG. In FIG. 6, the distance from the epicenter to elevator equipment EV001 in earthquake EQ2003** is shorter than the distance from the epicenter to elevator equipment EV026 in earthquake EQ2003**. In "Z Prefecture ** City," which is close to the epicenter, a seismic intensity of "4" was observed, and in "Z Prefecture, D City" a seismic intensity of "3" was observed (see Figure 3).

The teacher data generation unit 41 may use the seismic intensity at the location of each elevator facility, which is specified based on the seismic intensity of each place included in the disaster information, as the degree of the disaster. Alternatively, the teacher data generation unit 41 may calculate (1) the location of the epicenter included in the disaster information, the magnitude indicating the scale of the earthquake, and the depth of the epicenter, and (2) the distance between the location of each elevator facility and the epicenter. The magnitude of shaking at the location of each elevator facility calculated from the following may be used as the degree of disaster.

[Driving status information]
Next, driving situation information will be explained using FIG. 7 with reference to FIG. 8. FIG. 7 is a diagram showing an example of data stored in the driving situation information database 6c. The operating status information database 6c records the operating status of elevator equipment for each disaster that occurred in the past.

The operation situation 63 when earthquake EQ2003** occurs will be explained using FIG. 8 as an example. FIG. 8 is a diagram showing an example of the data structure of the driving situation 63. Earthquake EQ2003** is an earthquake that occurred "off the coast of Z prefecture" on "March 25, 2020" as shown in Figure 2. The operating status 63 records the operating status of each elevator facility when EQ2003** occurs. Here, the operating status 63 may include at least normal operation and stoppage. Note that the operating status 63 may record the results of work (for example, restoration work) performed on each elevator facility. For example, in FIG. 8, it is recorded that the elevator equipment EV001 was stopped when the earthquake EQ2003** occurred, and that restoration work was performed on March 27, 2020. Furthermore, in Figure 8, it is recorded that elevator equipment EV026 was not stopped when earthquake EQ2003** occurred and was operating normally, and that no restoration work was performed and only inspection was performed. .

Elevator equipment includes monitored equipment whose operating status can be remotely monitored, and non-monitored equipment whose operating status cannot be monitored remotely. There is no need to actually go to the location and check the operating status of the equipment to be monitored. On the other hand, in the case of facilities that are not subject to monitoring, operators may receive calls (callbacks) from owners or users of the facility regarding the operating status via telephone or other communication means, or workers at the facility may actually visit the location. It is necessary to check the driving situation and damage situation.

(Processing to generate learning model 451)
Next, the process of generating the learning model 451 will be explained using FIG. 9. FIG. 9 is a flowchart illustrating an example of the process of generating the learning model 451 by the learning model generation device 4.

First, the teacher data generation unit 41 acquires a disaster record, location information indicating the location of the elevator equipment, and operating status information indicating the operating status of the elevator equipment when the disaster occurred (step S41).

Subsequently, the teacher data generating unit 41 generates teacher data including location information, disaster information including the degree of disaster at the location of the elevator equipment, and driving status information (step S42).

Next, the learning unit 42 inputs the disaster information and location information into the learning model candidates. Here, the learning model candidate is intended to be a learning model for which machine learning has not been completed. When the predetermined machine learning is completed, the learning model candidate is stored in the storage unit 45 as a learning model 451 (learned model). Here, as a machine learning algorithm for generating the learning model 451, for example, a neural network may be used, although it is not limited thereto.

The learning unit 42 uses the estimation result obtained by the learning model candidate to estimate the operating status of the elevator equipment corresponding to the input disaster information and location information, and the operation status of the elevator equipment associated with the input disaster information and location information. The error is calculated by comparing with the situation information (step S44). The learning unit 42 updates the learning model candidate so that the error calculated in step S44 is minimized (step S45).

Updating the learning model candidate may include performing the following so that the error between the estimation result estimated by the learning model candidate and the driving situation information is minimized.
- Updating learning model candidates by updating weights between nodes, hyperparameters, etc. - Generating new learning model candidates To update learning model candidates, an error backpropagation method or the like may be applied.

When the predetermined machine learning is completed, the learning unit 42 stores the learning model candidate at that time in the storage unit 45 as the learning model 451.

The learning model generation device 4 includes location information indicating the location of the elevator equipment, disaster information including the degree of disaster at the location, and operation status information indicating the operating status of the elevator equipment when a disaster occurred in the past. A learning model 451 is generated by machine learning using data. When a new disaster occurs, the learning model 451 generated in this way estimates the operating status of the elevator equipment in response to input of the location information of the elevator equipment and the disaster information of the disaster that has occurred. The results can be output. The learning model 451 can be applied to the operating status estimation device 5 that estimates the operating status of each elevator equipment when a disaster occurs based on the operating status of each elevator equipment when a disaster occurred in the past. The driving situation estimation device 5 will be explained later (see FIG. 14).

[Embodiment 2]
A learning model generation device 4a capable of generating a learning model 451a capable of estimating the operating status of elevator equipment when a disaster occurs with higher accuracy will be explained using FIGS. 10 to 13. . For convenience of explanation, members having the same functions as the members described in the above embodiment are given the same reference numerals, and the description thereof will not be repeated.

(Configuration of learning model generation device 4a)
The configuration of a learning model generation device 4a that performs machine learning to generate a learning model 451a that outputs the operating status of the elevator equipment using the location information of the elevator equipment and disaster information as input data will be explained using FIG. explain. FIG. 10 is a block diagram showing an example of a schematic configuration of a learning model generation device 4a according to Embodiment 2 of the present invention.

The learning model generation device 4a may include a control section 40a and a storage section 45. The control unit 40a may be, for example, a CPU. The control unit 40a controls the learning model generation device 4a to execute processing of each function. The storage unit 45 is a storage device that stores various computer programs read by the control unit 40a, data used in various processes executed by the control unit 40a, and the like.

The control section 40a includes a teacher data generation section 41a and a learning section 42a.

The learning unit 42a generates a learning model 451a that outputs the operating status of the elevator equipment of interest using the location information of the elevator equipment of interest and disaster information as input data by performing machine learning using teacher data. In addition to location information, disaster information, and driving status information, the teacher data may further include at least one of equipment information, building information, and surrounding driving status information.

Here, the equipment information is information indicating at least one of the type and specifications of the elevator equipment. The building information is information indicating at least one of the completion date and seismic specifications of the building in which the elevator equipment is installed. The peripheral operating status information is information indicating the operating status of peripheral equipment, which is elevator equipment that is located within a predetermined range from the elevator equipment and whose operating status can be remotely monitored, when a disaster occurs.

FIG. 10 shows a learning model generation device 4a that can generate teacher data including location information, disaster information, driving status information, equipment information, building information, and surrounding driving status information. The teacher data generation unit 41a generates teacher data for generating the learning model 451a. For example, the teacher data generation unit 41a may acquire location information, equipment information, and building information indicating the location of the building in which the elevator equipment is installed (i.e., the location of the elevator equipment) from the elevator equipment installation information database 6b. Further, the teacher data generation unit 41 may acquire driving situation information and surrounding driving situation information from the driving situation information database 6c.

There are two types of elevator equipment: elevator equipment that is equipped with a controlled operation function that makes an emergency stop in the event of a disaster, and elevator equipment that is not equipped with a controlled operation function. Therefore, the teacher data used to generate the learning model 451 may further include equipment information indicating at least one of the type and specifications of the elevator equipment.

The seismic standards that buildings must meet are subject to review and change. Buildings that were completed before the earthquake resistance standards were changed are buildings that meet the old earthquake resistance standards, and buildings that were completed after the earthquake resistance standards were changed are buildings that meet the new earthquake resistance standards. In this way, if the completion date of a building is known, it can be determined whether the earthquake resistance standards applied to that building are the old earthquake resistance standards or the new earthquake resistance standards. Seismic specifications include seismic isolation structures and earthquake-resistant structures. Earthquake resistance grades 1 to 3 may be applied to the earthquake resistance specifications of houses. Even buildings to which the old earthquake resistance standards were applied may meet the new earthquake resistance standards by performing seismic reinforcement work. Therefore, the teacher data used to generate the learning model 451 may include building information indicating at least one of the completion date and seismic specifications of the building in which the elevator equipment is installed.

Furthermore, when a disaster occurs, the extent to which the disaster affects the operational status of elevator equipment is also influenced by the structural type of each building. Examples of the structure types include reinforced concrete construction (RC construction), steel frame construction (S construction), and steel reinforced concrete construction (SRC construction). Buildings with different types of structures experience differences in the magnitude of shaking during an earthquake. Therefore, the building information used for the teacher data used to generate the learning model 451 may include the structural type of the building in which the elevator equipment is installed.

As described above, elevator equipment includes monitored equipment, which is elevator equipment whose operating status can be remotely monitored, and non-monitored equipment, which is elevator equipment whose operating status cannot be monitored remotely. Therefore, the training data used by the learning model generation device 4a to generate the learning model 451a is used when a disaster occurs for peripheral equipment, which is elevator equipment located within a predetermined range from the elevator equipment and whose operating status can be remotely monitored. The vehicle may also include surrounding driving situation information indicating the driving situation of the vehicle.

The operating status of unmonitored equipment is not known until the building owner or user contacts us or actually checks it on-site. However, when a disaster occurs, the damage situation and driving situation of surrounding facilities located close to each other can be used as a reference for determining the damage situation and driving situation. Since the operating status of monitored equipment is constantly monitored remotely, business offices are also notified of the operating status in the event of an earthquake, etc. Therefore, if the operating status of peripheral equipment of the non-monitored equipment is also considered, it is possible to improve the reliability of estimating the operating status of the non-monitored equipment.

The equipment information, building information, and surrounding driving status information will be described below using specific examples.

[Equipment information and building information]
First, equipment information and building information will be explained using FIG. 11. FIG. 11 is a diagram showing an example of data stored in the elevator equipment installation information database 6b. In the elevator equipment installation information database 6b, in addition to the location of the building where the elevator equipment is installed, equipment information and building information are recorded in association with each elevator equipment. Note that each elevator facility may include information indicating whether it is a monitored facility that can be remotely monitored or a non-monitored facility that cannot be remotely monitored.

According to FIG. 11, the elevator equipment EV001 is installed in a "reinforced concrete (RC)" building that was completed on "**Monday**, 1979." This building complies with the old earthquake resistance standards. The model of the elevator equipment EV001 is "A03", and its specifications include a "controlled operation function". Elevator equipment EV002 is installed in a reinforced steel concrete (SRC) building that was completed on **Monday**, 1979. This building complies with the old earthquake resistance standards. The model of the elevator equipment EV002 is "C05", and its specifications include a "controlled operation function". Elevator equipment EV010 is installed in a "reinforced concrete (RC)" building that was completed on "**Monday**, 2015." This building complies with the new earthquake resistance standards. The model of the elevator equipment EV010 is "D01", and its specifications include a "controlled operation function". Elevator equipment EV026 is installed in a "steel frame (S)" building that was completed on "**Monday**, 2018." This building complies with the new earthquake resistance standards. The model of the elevator equipment EV026 is "P12", and its specifications do not include a "controlled operation function".

[Surrounding driving situation information]
According to FIG. 11, the elevator equipment EV002 and the elevator equipment EV010 are equipment to be monitored, and the elevator equipment EV001 and the elevator equipment EV026 are equipment not to be monitored. The surrounding operating status information will be explained using the positional relationship of these elevator facilities. FIG. 12 is an image diagram showing the positional relationship between the epicenter of an earthquake that occurred in the past and elevator equipment EV001, EV002, EV010, and EV026. Both elevator equipment EV001 and elevator equipment EV002 are located in ** City, Z Prefecture. The location of the elevator facility EV010 is F City, Z Prefecture, and the location of the elevator facility EV026 is D City, Z Prefecture. Facilities to be monitored are indicated by black circles, and facilities not to be monitored are indicated by white circles.

As shown in FIG. 12, the elevator equipment EV002 is located within a predetermined range (indicated by a broken circle in the figure) from the elevator equipment EV001. Furthermore, since the elevator equipment EV002 is the equipment to be monitored, the elevator equipment EV002 is a peripheral equipment of the elevator equipment EV001. Similarly, the elevator equipment EV010 is located within a predetermined range (indicated by a broken line circle in the figure) from the elevator equipment EV026. Further, since the elevator equipment EV010 is the equipment to be monitored, the elevator equipment EV010 is a peripheral equipment of the elevator equipment EV026.

The teacher data generation unit 41a may acquire driving situation information and surrounding driving situation information from the driving situation information database 6c. FIG. 13 is a diagram showing an example of data stored in the driving situation information database 6c. The operating status of the elevator equipment EV002 and the elevator equipment EV010 is peripheral operating status information of the elevator equipment EV001 and the elevator equipment EV026, respectively.

[Embodiment 3]
By employing the learning models 451 and 451a generated by the learning model generation devices 4 and 4a according to the embodiments described above, it is possible to realize the driving situation estimation device 5 that can estimate the driving situation of elevator equipment. The driving situation estimation device 5 will be explained below using FIGS. 14 to 21. For convenience of explanation, members having the same functions as the members described in the above embodiment are given the same reference numerals, and the description thereof will not be repeated.

(Configuration of driving situation estimation device 5)
First, using the learning models 451 and 451a, the configuration of the operating status estimation device 5 that estimates the operating status of the elevator equipment of interest using the location information of the elevator equipment of interest and disaster information as input data will be explained using FIG. 14. explain. FIG. 14 is a block diagram showing an example of a schematic configuration of the driving situation estimation device 5 according to the third embodiment of the present invention. Here, the driving situation estimation device 5 to which the learning model 451a is applied will be described as an example. Note that each function of the driving situation estimation device 5 can be realized even if the learning model 451a in the following explanation is replaced with the learning model 451.

The driving situation estimation device 5 may include a control section 50 and a storage section 54. The control unit 50 may be, for example, a CPU. The control unit 50 controls the driving situation estimation device 5 to execute processing of each function. The storage unit 54 is a storage device that stores various computer programs read by the control unit 50, data used in various processes executed by the control unit 50, and the like.

In this specification, the driving situation estimation device 5 having a function of generating input data will be described as an example, but the present invention is not limited thereto. For example, the driving situation estimation device 5 may be configured to acquire input data generated in advance by another device and input this to the learning model 451a. When this configuration is adopted, the learning section 42 may read and use the teacher data stored in the storage section 45. Therefore, in the driving situation estimating device 5, the input data generating section 51, which will be described later, is not an essential component.

The control section 50 includes an input data generation section 51, an estimation section 52, and an output control section 53.

The input data generation unit 51 generates input data to be input to the learning model 451a. For example, the input data generation unit 51 may use disaster occurrence information about a disaster that has occurred from the disaster occurrence information distribution device 6d to generate disaster information including the degree of disaster at the location of the attention elevator equipment. The disaster occurrence information distribution device 6d may be a device that quickly distributes information regarding the disaster that has occurred. The disaster occurrence information may be, for example, information included in an emergency bulletin distributed by the Japan Meteorological Agency. Further, the input data generation unit 51 may acquire peripheral operating status information indicating the operating status of peripheral equipment located within a predetermined range from the elevator equipment of interest from the elevator equipment remote monitoring device 6e. The elevator equipment remote monitoring device 6e may be, for example, a device managed by a business office that undertakes management of equipment to be monitored. Further, the input data generation unit 51 may acquire location information, building information, and equipment information of the elevator equipment of interest from the elevator equipment installation information database 6b.

The estimation unit 52 estimates the probability of the operating status of the elevator equipment of interest. The estimation unit 52 inputs input data to the learning model 451a, and outputs output data that is an estimation result of the operating status of the elevator equipment of interest.

The output control unit 53 outputs the output data from the estimation unit 52 to an external device. For example, the output control unit 53 causes the display device 9 to display the estimation result output by the estimation unit 52. Here, the display device 9 may be a display that is communicably connected to the driving situation estimation device 5 by wire or wirelessly. In one example, the driving situation estimation device 5 may be a personal computer and a tablet terminal, and in this case, the display device 9 may be a display included in the personal computer and the tablet terminal. Further, the output control unit 53 may cause the estimation result outputted by the estimation unit 52 to be stored in the storage unit 54 of the driving situation estimation device 5 or another storage device.

[Disaster occurrence information]
Disaster occurrence information will be described below using FIG. 15 with reference to FIGS. 16 and 17. FIG. 15 is a diagram showing an example of the data structure of disaster occurrence information distributed from the disaster occurrence information distribution device 6d. The information distributed from the disaster occurrence information distribution device 6d includes data regarding the occurrence date and time, epicenter name, maximum seismic intensity, epicenter and epicenter depth, disaster ID, seismic intensity of each place, and shaking characteristics regarding the disaster that has occurred. may be included.

FIG. 15 shows disaster occurrence information regarding an earthquake (hereinafter referred to as earthquake EQ2106**) to which a disaster ID of "EQ2106**" has been assigned. Earthquake EQ2106** is an earthquake that occurred "off the coast of Z Prefecture" on "June **, 2021". The maximum seismic intensity of earthquake EQ2106** is "4", and the magnitude indicating the scale of the earthquake is "3.8". The epicenter of earthquake EQ2106** is "nn degrees north latitude, mm degrees east longitude", and the depth of the epicenter is "30 km".

The seismic intensity 61a in each region is information on the seismic intensity observed at observation points in each region when earthquake EQ2106** occurred, and as shown in Figure 16, the seismic intensity observed in each region and the city, ward, town, or village where the observation point is may be associated with each other. According to Figure 16, a maximum seismic intensity of 4 was observed in places such as "Prefecture Z** City" and "City A, Prefecture Z", and a seismic intensity of 3 was observed in "City F, Prefecture Z" and "City B, Prefecture Z". It has been shown that

The shaking characteristics 62a may include information indicating the direction, acceleration, and period of shaking observed in the earthquake EQ2106**. As shown in FIG. 17, the shaking characteristic 62a shows the acceleration of shaking observed when earthquake EQ2106** occurred for each of the north-south component, east-west component, and up-and-down component. The shaking characteristics 62a may include the magnitude (amplitude) of the observed shaking, the period of the shaking, and the like. In the example shown in FIG. 17, the shaking characteristic 62a indicates that a shaking period of "1.2 sec" was measured. Note that the shaking characteristics 62a may be distributed for each area, such as the shaking characteristics in each area where the shaking was observed.

Next, the location information, building information, and equipment information of the elevator equipment of interest will be explained using FIG. 18 while referring to FIG. 19. FIG. 18 is a diagram showing an example of location information, building information, and equipment information of the elevator equipment of interest.

FIG. 19 is an image diagram showing the positional relationship between the epicenter of earthquake EQ2106** and elevator equipment. FIG. 19 shows the epicenter (indicated by "X" in the figure) of earthquake EQ2106** (see FIG. 15). Further, FIG. 19 shows the locations of elevator equipment EV001, elevator equipment EV002, elevator equipment EV010, elevator equipment EV026, and elevator equipment EV050. Note that major roads are also illustrated in FIG. 19.

The input data generation unit 51 may use the seismic intensity at the location of each elevator facility, which is specified based on the seismic intensity of each place included in the disaster occurrence information, as the degree of disaster. Alternatively, the teacher data generation unit 41 may calculate (1) the location of the epicenter included in the disaster information, the magnitude indicating the scale of the earthquake, and the depth of the epicenter, and (2) the distance between the location of each elevator facility and the epicenter. The magnitude of shaking at the location of each elevator facility calculated from the following may be used as the degree of disaster.

The input data may include the operating status when a disaster occurs, which is obtained by remotely monitoring peripheral equipment located within a predetermined range from the elevator equipment of interest. In FIG. 19, elevator equipment EV002 is peripheral equipment of elevator equipment EV001, and elevator equipment EV010 is peripheral equipment of elevator equipment EV026 and elevator equipment EV050. Further, the input data may further include at least one of equipment information of the elevator equipment of interest and building information of the elevator equipment of interest. By inputting this information into the learning model 451a together with the location information, disaster information, and driving situation information, the driving situation estimation device 5 can output a more reliable estimation result.

(Process executed by driving situation estimation device 5)
Next, the process executed by the driving situation estimation device 5 will be described using FIG. 20 with reference to FIG. 21. FIG. 20 is a flowchart illustrating an example of the process of estimating the operating status of the elevator equipment of interest by the operating status estimating device 5.

First, the input data generation unit 51 acquires disaster occurrence information, surrounding operating status information, location information of the elevator equipment of interest, building information, and equipment information (step S51).

Subsequently, the input data generation unit 51 generates input data including disaster information and location information of the elevator equipment of interest (step S52).

On the other hand, the estimation unit 52 reads the learning model 451a from the storage unit 54 (step S53). Subsequently, the estimation unit 52 inputs the input data generated in step S52 to the learning model 451a, and estimates the operating status of the elevator equipment of interest (step S54: estimation step).

The estimation unit 52 outputs the estimation result (step S55). FIG. 21 is a diagram showing an example of the estimation results output from the driving situation estimation device 5. The estimation result may be the probability of the operating status of the elevator equipment of interest. For example, the operating status estimating device 5 may output the operating status of each elevator equipment as a probability that it is stopped as shown in FIG. 21 or as a probability that it is operating normally. According to FIG. 21, when a disaster (for example, earthquake EQ2106**) occurs, the probability that elevator equipment EV001 is stopped is "80%", and the probability that elevator equipment EV026 is stopped is "15%". ”. The probabilities related to the operating conditions output as the estimation results serve as a reference in determining the necessity of dispatching personnel for restoration work. For example, dispatch to restore elevator equipment EV001, which has an 80% probability of being stopped, should be prioritized over dispatch to restore elevator equipment EV026, which has a 15% probability of being stopped. be. The estimation result output from the estimation unit 52 may be output to the display device 9 via the output control unit 53, or may be output to the dispatch planning system 10, which will be described later.

Although the driving situation estimation device 5 that employs the learning model 451a has been described here as an example, the driving situation estimation device 5 may also employ the learning model 451. In this case, the operating status estimating device 5 may input the location information and disaster information of the elevator equipment of interest to the learning model 451 to estimate the operating status of the elevator equipment of interest.

[Embodiment 4]
Other embodiments of the invention will be described below. For convenience of explanation, members having the same functions as the members described in the above embodiment are given the same reference numerals, and the description thereof will not be repeated.

<Overview of maintenance system>
Hereinafter, one embodiment of the present invention will be described in detail. FIG. 22 is a block diagram showing an overview of the maintenance system 1000. The maintenance system 1000 is a system for supporting recovery work for a plurality of recovery targets in different locations where an event requiring recovery work has occurred. In this embodiment, the restoration target is typically elevator equipment such as a passenger conveyor or an elevator. Furthermore, an example of an event that requires restoration work is a stoppage or failure of elevator equipment.

Maintenance system 1000 includes at least dispatch support system 100. The secondment support system 100 supports, for example, a front end system 1, a damage information providing system 200, a map information providing system 300, a business office device 2, a terminal device 3 (information terminal), etc. via a communication network such as the Internet. Connected for communication with other devices.

(About secondment support system)
In the maintenance system 1000, the dispatch support system 100 has a function of dispatching each of a work group (hereinafter simply referred to as a "group") to perform recovery work on a plurality of recovery targets. As an example, the dispatch support system 100 has a dispatch plan planning function that drafts a dispatch plan for dispatching each group to a plurality of recovery targets. A configuration group of hardware and software for realizing this dispatch planning function is referred to as a dispatch planning system 10. Details of the configuration of the dispatch planning system 10 will be described in detail later in the present embodiment, but the dispatch planning system 10 according to the present embodiment typically includes the driving situation estimation device described in the above embodiment. It is equipped with 5.

In addition, in the maintenance system 1000, the dispatch support system 100 arranges the dispatch of each group based on the drafted dispatch plan, manages the progress of recovery work, and adjusts the dispatch plan according to the ever-changing disaster situation. It has the function of reviewing and arranging dispatch. A configuration group of hardware and software for realizing this dispatch arrangement function is referred to as a dispatch arrangement system 20.

The dispatch planning system 10 and the dispatch arrangement system 20 of the present disclosure may be configured by one information processing device equipped with a configuration group of hardware and software, or may be configured by a plurality of information processing devices. . A configuration example of the information processing device will be described later with reference to FIG. 30.

Note that part of the hardware and software configuration group described above may be configured to realize both the dispatch planning function and the dispatch arrangement function. That is, part of the configuration group that realizes the dispatch planning system 10 may be part of the configuration group that realizes the dispatch arrangement system 20.

(About secondment plans)
The dispatch plan drawn up by the dispatch planning system 10 is one in which the combination of dispatch targets, which are the recovery targets to which the group performing the recovery work is scheduled to dispatch, and the order of dispatch are determined for each group. The secondment planning system 10 may formulate a secondment plan for one office that has a plurality of groups in the area under its control. Further, the secondment planning system 10 may formulate a secondment plan for every office under the jurisdiction of the maintenance company.

In this embodiment, as an example, it is assumed that the secondment planning system 10 creates a secondment plan for each office under the jurisdiction of the maintenance company. The dispatch planning system 10 may operate according to instructions from a front end system 1 (output unit) operated by an operator working at a maintenance company. In this case, the dispatch planning system 10 may be incorporated into an information processing device that constitutes the front end system 1, or may be implemented by another information processing device that is communicably connected to the front end system 1 as an input/output device. may be configured.

In addition, when the secondment planning system 10 is to formulate a secondment plan for one business office in the area under its control, the secondment planning system 10 is a business office device operated by an operator working at the business office. 2 (output section). In this case, the secondment planning system 10 may be built into the office device 2 as an information processing device, or may be constituted by another information processing device communicably connected to the office device 2 as an input/output device. may be done.

(About other devices)
As described above, the front end system 1 is a system configured with one or more information processing devices operated by an operator working at a maintenance company. In this embodiment, the front end system 1 is communicably connected to a dispatch support system 100 including a dispatch planning system 10 and a dispatch arrangement system 20. As an input/output device, the front end system 1 supports the operator in inputting information into the dispatch support system 100, acquires information output from the dispatch support system 100, and presents the information to the operator. The front end system 1 and the dispatch support system 100 are connected through an arbitrary communication network. The communication network may be a LAN (Local Area Network) or a wide area network including the Internet.

As described above, the office device 2 is a device operated by an operator working at one office. In this embodiment, the office device 2 may be communicably connected to the secondment support system 100 and the terminal device 3 via an arbitrary communication network.

The terminal device 3 is a device operated by a group leader of a group. The terminal device 3 is assumed to be, for example, a mobile terminal such as a smartphone or a tablet PC. In this embodiment, the terminal device 3 is communicably connected to the secondment support system 100 via an arbitrary communication network. The terminal device 3 may be communicably connected to the office device 2.

The damage information providing system 200 monitors the elevator equipment that maintenance companies are contracted to maintain, collects information on each elevator equipment, and provides the information to the dispatch support system 100. Specifically, when a disaster such as an earthquake occurs, the damage information providing system 200 collects information on affected elevator equipment. If an event such as stoppage or failure occurs in a large number of elevator equipment at the same time, the damage information providing system 200 generates operating status information indicating the operating status of the elevator equipment in the disaster area. The operating status information generated here, similar to the operating status information shown in Figure 13, indicates whether the operating status falls under "stopped" or "normal operation" for each elevator facility located in the disaster area. It may be information. The driving situation information is transmitted from the damage information providing system 200 to the dispatch support system 100 via the communication network.

The map information providing system 300 provides map information to the dispatch support system 100. The map information providing system 300 may provide map information to the front end system 1 , the office device 2 , and the terminal device 3 . The map information providing system 300 may grasp the road traffic situation in various places in real time and provide each device with map information including information such as traffic jams or road closures, if necessary. Furthermore, the map information providing system 300 may receive a departure point and a destination from a client, and provide the movable route, travel distance, travel time, etc. from the departure point to the destination. Here, the clients include the secondment support system 100, the front end system 1, the office equipment 2, the terminal device 3, and the like.

(Processing flow of secondment support system)
FIG. 23 is a flowchart showing the processing executed by the secondment support system 100.

In step S1, the secondment support system 100 stores in advance the number of groups that can respond to emergencies for each office. The secondment support system 100 sets a group leader for each group in each office, and stores the contact information of the group leader for each office. A list storing the number of groups for each office and the contact information of the group leader of each group may be created in advance by the operator and stored in the storage device of the dispatch support system 100.

It is preferable that the process in step S1 is executed in advance before a disaster actually occurs.

When a disaster occurs and a failure or stoppage of a large number of elevator equipment is simultaneously detected, the damage information providing system 200 generates a list of all the elevator equipment that requires restoration work. When the entire list is provided from the damage information providing system 200 to the dispatch support system 100, the dispatch support system 100 starts processing from step S2 onwards.

Note that the damage information providing system 200 may directly communicate with each elevator facility that is a target of remote monitoring, and add elevator facilities that are determined to require restoration work to the above-mentioned overall list as restoration targets. Furthermore, in the damage information providing system 200, an operator may be provided to receive notifications (callbacks) of failures from owners or users via communication means such as telephones regarding elevator equipment that is not subject to remote monitoring. The operator who receives the callback registers the elevator equipment that is not subject to remote monitoring in the damage information providing system 200 as a recovery target. In this manner, the damage information providing system 200 may add the elevator equipment for which the callback has been received to the above-mentioned entire list as restoration targets.

In step S2, the dispatch support system 100 divides operation status information including the operation status of a large number of elevator facilities in the disaster area for each business establishment. Division of the driving situation information may be performed based on geographical conditions. For example, the secondment support system 100 knows in advance the region that each business office is in charge of, and assigns the elevator equipment in the operation status information to each business office depending on which region the elevator equipment is located in. Can be associated. Hereinafter, the operation status information of elevator equipment in the region corresponding to each business establishment, which is divided by business establishment, will be referred to as an operation status list by business establishment. Note that the operating status list by business establishment may be created according to manual input work by an operator. The operating status list by establishment obtained in this way shows the operating status of elevator equipment whose status is actually known. Lift equipment whose operating status is known are those that are subject to remote monitoring and those that are not subject to remote monitoring, and whose status is known by receiving a callback from the owner or user. Refers to elevator equipment. The operating status may include at least normal operation and stoppage. In other words, the operating status list by business establishment is a list that at least indicates, for each elevator equipment in the area under the jurisdiction of the business establishment, whether the elevator equipment is in normal operation or has stopped abnormally. This allows the dispatch planning system 10 to determine whether recovery work is necessary.

In step S3, the operating status estimating device 5 of the dispatch planning system 10 estimates the operating status of elevator equipment that is not subject to remote monitoring and whose operating status is not known. Specifically, step S3 includes an operation status acquisition step (S3-1) of acquiring the operation status from monitored equipment, which is elevator equipment whose operation status can be remotely monitored among a plurality of elevator equipment located at different locations; This includes an operating status estimation step (S3-2) of estimating the operating status of non-monitored equipment, which is elevator equipment whose operating status cannot be remotely monitored. The driving situation estimation process executed by the driving situation estimating device 5 in step S3 has been described in the third embodiment, so detailed description thereof will not be repeated in this embodiment.

In step S4, the dispatch planning system 10 of the dispatch support system 100 searches for an optimal dispatch plan from among several possible dispatch plan candidates. The optimal dispatch plan is an optimal solution or a suboptimal solution of the dispatch plan. In this embodiment, the secondment planning system 10 searches for a secondment plan with a relatively early estimated completion time for completion of recovery of all secondment targets handled by the business office as the optimal secondment plan. Although a genetic algorithm can be applied as an example of a search method, other algorithms for solving combinatorial optimization problems may also be applied.

Specifically, step S4 includes a restoration target identification step (S4-1) in which a restoration target is identified from among the elevator equipment, and a restoration target identified in the restoration target identification step as a secondment target, and the combination and secondment. and a search step (S4-2) for searching for an optimal dispatch plan with a determined order.

In the recovery target identification process, when a disaster occurs, the dispatch planning system 10 identifies recovery targets from among the monitored equipment according to the operating status acquired in the operating status acquisition process, and Sometimes, recovery targets are identified from among the non-monitored equipment according to the operating status estimated in the above-mentioned operating status estimation process.

In the search process, the dispatch planning system 10 determines the optimal dispatch plan in which the combination and dispatch order of dispatch targets to which a group consisting of one or more workers performing recovery work is scheduled to be dispatched is determined for each group. Search for an optimal dispatch plan that is a solution or a suboptimal solution.

In this embodiment, the dispatch planning system 10 searches for an optimal dispatch plan for each office owned by the maintenance company. That is, the secondment planning system 10 performs a driving status acquisition process (S3-1), a driving status estimation process (S3-2), a recovery target identification process (S4-1), and a search process (S4) for each office. -2) and repeat. In another example, the dispatch planning system 10 targets the entire disaster area and, prior to step S2, performs a driving status acquisition process (S3-1), a driving status estimation process (S3-2), and a recovery target A specifying step (S4-1) may also be performed. Then, in step S2, the secondment planning system 10 divides the list of recovery targets for the entire disaster area by area under the jurisdiction of the office, and then creates an optimal dispatch plan for the recovery targets for each office. The search step (S4-2) may be repeated for each business establishment.

In addition, the dispatch planning system 10 determines that when a business office proceeds with recovery work in accordance with the corresponding optimal dispatch plan, the business office completes the recovery work for all recovery targets assigned to the business office. The estimated time may be output as the estimated completion time of the office.

Furthermore, the secondment planning system 10 virtually changes group affiliations between business offices, simulates secondment arrangements, and develops a better optimal secondment plan that accelerates the estimated completion time for each business office. You can also explore each location. The secondment planning system 10 may suggest to the operator a group distribution pattern for each business office that will result in a better optimal secondment plan.

The dispatch planning system 10 identifies the optimal dispatch plan adopted by the operator as the final optimal dispatch plan. The operator may adopt the optimal dispatch plan searched in step S3, or may select the optimal dispatch plan to be adopted from among several simulated optimal dispatch plans.

The secondment planning system 10 stores the identified final optimal secondment plan for each office in the storage device of the secondment support system 100. In the present embodiment, the optimal dispatch plan for each office may include, for each group, a queue in which dispatch targets that are recovery targets to which the group is scheduled to be dispatched are arranged in a line in the order of dispatch.

In step S5, the dispatch arrangement system 20 of the dispatch support system 100 arranges dispatch according to the saved optimal dispatch plan. First, the secondment arrangement system 20 may confirm whether or not the secondment target can be seconded with respect to the group to which the secondment target is associated. The secondment arrangement system 20 may manage a secondment target who has been informed by the group that the secondment is possible, as a secondment confirmed target, and distinguish it from other secondment unconfirmed targets.

In step S6, the dispatch arrangement system 20 may manage the progress of the recovery work for each dispatch confirmation target. Then, the dispatch arrangement system 20 updates the above-mentioned queue according to the ever-changing situation as the restoration work progresses or new restoration targets are added.

In step S7, the dispatch planning system 10 monitors updates in the operating status list by office and determines whether it is necessary to search again for an optimal dispatch plan. If it is determined that re-search is necessary (YES in S7), the process returns to step S3, the updated operating status of the monitored equipment is acquired, the operating status of the non-monitored equipment is estimated, and restoration is performed. Respecify the target and repeat the search for the optimal dispatch plan for the target to be restored.

In step S8, the dispatch arrangement system 20 determines whether the recovery work for all dispatch targets, that is, the recovery targets, has been completed at all offices. If the recovery work has been completed for all recovery targets (YES in S8), the dispatch support system 100 ends the series of processes. Note that if there remains a restoration target for which the restoration work has not been completed (NO in S8), the dispatch arrangement system 20 returns to step S6 and continues progress management.

<Configuration of secondment planning system 10>
FIG. 24 is a block diagram showing the configuration of main parts of the dispatch planning system 10. The dispatch planning system 10 includes at least a driving situation acquisition section 11 , a driving situation estimation device 5 , a recovery target identification section 14 , and a search section 12 . Although only the estimation unit 52 of the driving situation estimation device 5 is representatively shown in FIG. 24, the driving situation estimation device 5 may include each part of the driving situation estimation device 5 shown in FIG.

The operating status acquisition unit 11 acquires operating status from monitored equipment, which is elevator equipment whose operating status can be remotely monitored, among a plurality of elevator equipment located at different locations. The operating status acquisition unit 11 may directly communicate with each of the monitored equipment to obtain the operating status of each, or may use the operating status of each monitored equipment aggregated in the damage information providing system 200 as damage information. They may be acquired all at once from the providing system 200. Further, the driving status acquisition unit 11 may acquire the driving status for each business establishment. For example, the operation status acquisition unit 11 may collect the operation status list by business establishment, which indicates the operation status of elevator equipment in the area under the jurisdiction of one business establishment, from among the operation status information provided by the damage information providing system 200. It may be acquired from the provision system 200 or the dispatch support system 100. Note that the operating status list by business facility includes the operating status of non-monitored equipment whose actual status is known based on callbacks from owners or users, even if the equipment is not subject to monitoring. It's okay.

The estimating unit 52 (operating status estimating unit) of the operating status estimating device 5 estimates the operating status of non-monitored equipment, which is elevator equipment whose operating status cannot be remotely monitored among the elevator equipment. The estimation unit 52 may estimate the operating status of the non-monitored equipment based on the operating status of the monitored equipment included in the above-mentioned operating status list by business establishment. Specifically, the input data generation unit 51 generates input data to be input to the learning model 451a using the operating status of the monitored equipment included in the above-mentioned operating status list by business establishment. The estimation unit 52 inputs the generated input data to the learning model 451a and obtains output data, thereby estimating the operating status of the non-monitored equipment. The output data is the estimation results described in Embodiments 1 to 3, and the estimation results may be information indicating the estimated operating status of each non-monitored facility. As shown in FIG. 21, the estimation result may be information indicating the estimated outage probability for each non-monitored facility. The output control unit 53 outputs the estimated operating status of the non-monitored equipment to the recovery target identification unit 14. Hereinafter, the estimated operating status of the non-monitored equipment will be referred to as an estimated operating status to distinguish it from the actually known operating status acquired by the operating status acquisition unit 11.

The restoration target specifying unit 14 identifies restoration targets from elevator equipment whose operating status or estimated operating status is shown. Specifically, the recovery target identification unit 14 identifies recovery targets from among the monitored equipment according to the operating status obtained when a disaster occurs, and also identifies recovery targets based on the degree of disaster when a disaster occurs. The equipment to be restored is identified from among the non-monitored equipment according to the estimated operating status.

For example, the recovery target identification unit 14 may generate a recovery target list indicating information about the plurality of recovery targets at different locations where an event requiring recovery work has occurred and which has been identified as described above. It may be generated separately. The recovery target specifying unit 14 outputs the recovery target list generated for each office to the search unit 12.

The search unit 12 searches for an optimal dispatch plan that is an optimal solution or a semi-optimal solution of the dispatch plan. As described above, the dispatch plan is a plan that determines for each group the combination of dispatch targets to which a group of one or more workers performing recovery work is scheduled to be dispatched, and the dispatch order. In this embodiment, the search unit 12 searches for an optimal dispatch plan using the recovery target specified by the recovery target identification unit 14 as the dispatch target.

According to the above-mentioned configuration, the operating status at the time of occurrence of a disaster is estimated for equipment that is not subject to monitoring and whose operating status cannot be grasped until it is contacted because it is not subject to remote monitoring. As a result, it is possible to specify not only equipment to be monitored whose operation status can be grasped in real time through remote monitoring, but also equipment not to be monitored to be restored. In this way, it is possible to formulate an accurate dispatch plan that takes into consideration the operational status of equipment not to be monitored at the time of a disaster.

After a disaster occurs, especially in the early stages, it is difficult to quickly grasp the overall picture of stopped elevator equipment, and dispatch plans drawn up at this stage may lack accuracy. Therefore, by utilizing the dispatch planning system 10 of the present invention in the early stages of a disaster occurrence, it is possible to quickly grasp the overall picture of the stopped elevator equipment and formulate a highly accurate dispatch plan from the early stage of the occurrence of a disaster. This is particularly advantageous because it makes it possible to

As an example, in this embodiment, a plurality of patterns are created for the dispatch plan. Each pattern differs from each other in the combination of dispatched objects and the dispatched order. Hereinafter, the above-mentioned pattern will be referred to as a secondment pattern. The process of generating a plurality of dispatch plans with different dispatch patterns may be executed by the search unit 12. In the present embodiment, the search unit 12 selects, among a plurality of dispatch plans with various dispatch patterns, a dispatch plan in which the estimated completion time for all the dispatch targets handled by the business office is relatively early. You may also explore this as a secondment plan. In the present embodiment, the estimated completion time may be estimated by taking into consideration at least the total travel time to the seconded target and the total work time required for the recovery work of the seconded target. The estimated completion time may be estimated by the search unit 12. The search unit 12 may output, for example, numerical values indicating the date and hours and minutes as the estimated completion time. The search unit 12 may output numerical values indicating the year, month, day, hour, minute, and second as the estimated completion time.

According to the above-described configuration, the search unit 12 searches for an optimal dispatch plan for a plurality of restoration targets in different locations. The optimal dispatch plan is an optimal dispatch plan or a semi-optimal dispatch plan among dispatch plans in which the combination of dispatch targets that are recovery targets to which the group is scheduled to dispatch and the dispatch order are determined for each group. That is, the search unit 12 searches for a dispatch plan with a relatively early estimated completion time at which the restoration of all the dispatch targets handled by the business office is completed, as the optimal dispatch plan. In this way, it is possible to formulate a dispatch plan that accelerates the time when all of the restoration work is completed.

The dispatch planning system 10 may further include one or more of a travel time acquisition section 13, an output control section 15, a disaster information acquisition section 16, and a confinement information acquisition section 17 as necessary.

Each of the above-mentioned parts of the dispatch planning system 10 (driving status acquisition unit 11, estimation unit 52, recovery target identification unit 14, search unit 12, travel time acquisition unit 13, output control unit 15, disaster information acquisition unit 16, and confinement information acquisition unit) 17) is a control block realized by a control device included in an information processing device that implements the dispatch planning system 10 and a storage device that stores a program.

Further, the secondment planning system 10 may store the latest secondment plan database 31 (latest secondment plan DB 31) in the above-mentioned storage device.

The travel time acquisition unit 13 acquires the travel time required for the worker to travel from the starting point to the destination. For example, the travel time acquisition unit 13 transmits to the map information providing system 300 a request that includes a query that specifies the location of the first seconded target as the departure point and the next second seconded target location as the destination. Then, the travel time acquisition unit 13 may acquire the travel time from the departure point to the destination from the map information providing system 300. Thereby, the search unit 12 can estimate the sum of the travel times required to move the seconded objects in the seconded order as the total travel time.

The travel time acquisition unit 13 may acquire the average travel time calculated by the map information providing system 300 as the above-mentioned travel time. Alternatively, the travel time acquisition unit 13 may acquire the travel time calculated by the map information providing system 300, taking into account the road traffic conditions. The travel time that takes road traffic conditions into account may be the travel time that is calculated taking into account the degree of road congestion, or the travel time that is calculated based on the travel distance of an alternative route when the shortest route is closed. It may also be the travel time.

Note that the dispatch planning system 10 may acquire position information indicating the current location of a predetermined leader among the workers belonging to the group. For example, the secondment planning system 10 may communicate with the terminal device 3 of each group leader of each office and acquire the current position information of the group leader from the terminal device 3.

Thereby, the search unit 12 can determine the current location of the group leader as the departure point of the group to which the group leader belongs, for each of a plurality of dispatch plans with various dispatch patterns. Then, the search unit 12 can estimate the travel time from the departure point to the location of the first dispatched target in the dispatch order.

The output control unit 15 outputs the results of the search performed by the search unit 12 to the output unit. In this embodiment, the output unit is the front end system 1 or the office device 2, as an example.

In this embodiment, as described above, each elevator facility that is a restoration target is associated with one of the establishments that perform maintenance of the restoration target. Each group belongs to one of the establishments. The search unit 12 searches for an optimal secondment plan for each office, targeting groups belonging to the office. Through this search, the optimal dispatch plan along with the estimated completion time is searched for each office.

The output control unit 15 outputs each of the estimated completion times in the optimal dispatch plan searched for each office to the output unit.

Thereby, the operator can easily grasp for each business establishment when the restoration work will be completed for all the restoration targets assigned to each business establishment. Presenting the estimated completion time for each office can contribute to quickly communicating the outlook for recovery work to all parties involved. By presenting the outlook for recovery in this way, it is possible to give a sense of security to the users or owners who are subject to recovery. Furthermore, the operator can contribute to quickly proceeding with dispatch arrangements based on the optimal dispatch plan. Alternatively, it may contribute to the operator quickly making a decision to revise a more optimal dispatch plan.

As an example, the output control unit 15 may output to the front end system 1 each estimated completion time in the optimal dispatch plan found for each office for all offices under the jurisdiction of the maintenance company. good. Furthermore, a dispatch pattern in which a combination of dispatch targets and a dispatch order are defined for each group, which is indicated by the optimal dispatch plan, may be output in addition to the estimated completion time. For example, the output control unit 15 outputs to the office equipment 2 an optimal dispatch plan for the office where the office equipment 2 is installed, that is, the above-mentioned dispatch pattern for the office and the estimated completion time for the office. may be output.

The disaster information acquisition unit 16 acquires information indicating the degree of disaster in each region. For example, when a disaster occurs, the disaster information acquisition unit 16 may acquire information indicating the extent of the disaster for each region from the damage information provision system 200 or other disaster information provision system (not shown). . Information indicating the degree of disaster in each region is not limited to information directly indicating the degree of disaster in each region, but may also be indirect information that allows the degree of disaster to be specified or estimated based on the information. It's okay.

If the disaster is an earthquake, for example, the extent of the disaster is calculated from (1) the seismic intensity at the location of the elevator equipment, or (2) the magnitude of the epicenter, the distance from the epicenter, and the depth of the epicenter. , the magnitude of shaking at the location of the elevator equipment, etc. Alternatively, the degree of disaster may be (3) characteristics of shaking at the location, such as pitching and rolling. When an earthquake occurs, the disaster information acquisition unit 16 may acquire at least any of the information (1) to (3) above from the damage information providing system 200 or other information providing system.

According to the above-described configuration, the recovery target specifying unit 14 can specify elevator equipment located in an area where a specified degree of disaster has occurred as a recovery target.

For example, the restoration target identifying unit 14 can identify elevator equipment located in an area where an earthquake of a predetermined seismic intensity or higher has occurred as a restoration target. For example, the restoration target identification unit 14 selects elevator equipment located in an area where the magnitude of shaking calculated from the magnitude of the epicenter, the distance from the epicenter, and the depth of the epicenter is greater than a predetermined value as a restoration target. can be specified. Furthermore, for example, the restoration target specifying unit 14 can specify, as a restoration target, elevator equipment located in an area where a characteristic shaking that seriously affects the operation of the elevator equipment has occurred.

As a result, even if the elevator equipment is a non-monitored facility that is not subject to remote monitoring, the secondment planning system 10 determines that the non-monitored facility is subject to recovery or restored, as it may have been affected by a disaster. It can be treated as a target candidate. As a result, it is possible to formulate an accurate dispatch plan that takes into account the operational status of equipment not subject to monitoring at the time of a disaster. In addition, the non-monitored facilities considered in planning the dispatch plan are narrowed down from the extremely large number of non-monitored facilities to elevator facilities located in areas where disasters of a specified degree have occurred. It will be done. Therefore, it is possible to prevent an inefficient dispatch plan from being created in which elevator equipment with low necessity for dispatch is mixed as a dispatch target.

The confinement information acquisition unit 17 acquires confinement information indicating whether or not a user is confined in the restoration target. For example, the confinement information acquisition unit 17 may acquire confinement information for each monitoring target facility from the damage information providing system 200. In other examples, confinement information may be included in driving situation information. Specifically, in the operating status information, confinement information may be associated with each monitored facility. In this case, the confinement information acquisition unit 17 may extract the confinement information of each monitored equipment from the operational status list by business establishment acquired by the operational status acquisition unit 11.

According to the above-described configuration, it is possible to formulate a dispatch plan that takes into account whether or not the user is confined. Specifically, the search unit 12 can generate a dispatch plan in which the dispatch order of the dispatch targets for which user confinement has occurred is earlier than the dispatch order for the dispatch targets for which no confinement has occurred. . As a result, a dispatch plan is drawn up that allows recovery work to be carried out, giving top priority to recovery targets where users are trapped.

<S4: Search for optimal dispatch plan>
(About input information)
In the following, input information that is input to the search unit 12 in order to search for an optimal dispatch plan for one office (for example, office A) will be explained. For example, the search unit 12 receives as input information the number of groups belonging to office A, the position information of the group leader of each group belonging to office A, and the list of recovery targets for office A. . The recovery target list for office A is generated by the recovery target specifying unit 14 in the recovery target specifying step S4-1.

The number of groups may be registered in advance in the storage device of the dispatch planning system 10, or may be specified by the operator. As the group leader's position information, for example, information provided to the dispatch planning system 10 from the group leader's terminal device 3 immediately before the search process may be employed.

FIG. 25 is a diagram illustrating an example of the data structure of the recovery target list. As an example, first, the restoration target identification unit 14 selects the elevators and escalators whose operation status is "stopped" from the operation status list by business establishment (FIG. 13) of elevator equipment located in the area under the jurisdiction of business establishment A. The equipment is identified as a restoration target and added to the restoration target list shown in FIG. 25. Here, among the non-monitored equipment, the recovery target identification unit 14 adds non-monitored equipment that is known to be stopped to the recovery target list based on a callback from the owner or user. Good too.

Next, the restoration target specifying unit 14 identifies elevator equipment that satisfies predetermined conditions as a restoration target from the estimated operating status (FIG. 21) output by the estimating unit 52 of the operating status estimating device 5, and creates the above-mentioned restoration target list. Add to. For example, the estimation unit 52 estimates the probability of the operating status of non-monitored equipment, and the recovery target identification unit 14 specifies non-monitored equipment with a probability exceeding a specified value as a restoration target. good.

According to the above-mentioned configuration, the operating status at the time of disaster occurrence is estimated as a probability for the non-monitored equipment whose operating status cannot be grasped until it is contacted because it is not subject to remote monitoring. As a result, it is possible to specify not only equipment to be monitored whose operation status can be grasped in real time through remote monitoring, but also equipment not to be monitored to be restored. In particular, non-monitored equipment that is highly likely to require restoration work can be added to the restoration targets. As a result, it is possible to formulate a more accurate dispatch plan that takes into consideration the operating status of equipment not to be monitored and its probability in the event of a disaster.

Specifically, the probability of the operating status estimated by the estimation unit 52 may be the probability that the non-monitored equipment is stopped. Hereinafter, this probability will be referred to as a stop probability. The recovery target specifying unit 14 may specify non-monitored equipment whose outage probability exceeds a predetermined value as a recovery target. Assuming that the predetermined value is, for example, 75%, in the estimated operating situation shown in FIG. 21, the equipment whose failure probability exceeds 75% is the non-monitored equipment with the elevator equipment ID “EV001”. Therefore, the recovery target specifying unit 14 can specify the non-monitored equipment with the elevator equipment ID "EV001" as a recovery target, and add the non-monitored equipment to the recovery target list.

For example, as shown in FIG. 25, the recovery target specifying unit 14 generates a recovery target list for each business establishment, which is a list of recovery targets assigned to the business establishment. Assume that the recovery target list shown in FIG. 25 is a recovery target list for office A, as an example. The generated recovery target list is delivered to the search unit 12 of the dispatch planning system 10. Further, the recovery target list may be provided to the front end system 1. Further, the recovery target list for office A may be provided to office device 2 of office A.

The recovery target list may include, for example, the following items: target ID, location, presence or absence of confinement, recovery priority, number of work locations, and outage probability.

The target ID is identification information for uniquely identifying the restoration target. Unique identification information is given to one elevator facility. In this embodiment, one elevator facility is treated as one restoration target (secondment target) regardless of how many elevators the one elevator facility is composed of. As the target ID, the elevator equipment ID used in the first to third embodiments may be used as is.

The location is information indicating the location of the restoration target, and includes a place name, street address, building name, etc.

The presence or absence of confinement is information indicating whether or not a user is trapped in the recovery target due to a failure or stoppage of the recovery target. In this embodiment, whether or not a user is trapped indicates whether or not a user is trapped in any of the elevators belonging to the elevator equipment to be restored. In the illustrated example, "Yes" indicates that there is a user who is trapped inside the elevator without being able to get out, and "No" indicates that there is no such user. The recovery target specifying unit 14 can reflect the contents of the confinement information acquired by the confinement information acquisition unit 17 in the item of presence or absence of confinement.

Note that whether or not there is a confinement in the elevator equipment is information that is known only in the monitored equipment or the non-monitored equipment that has received a callback indicating that the entrapment has occurred. Therefore, no confinement information is associated with non-monitored equipment whose operating status is unknown.

The recovery priority is information indicating the recovery priority set for the recovery target. In this embodiment, as an example, the recovery priority of the recovery target in which the user is trapped is set to the highest (for example, priority S). For recovery targets for which no entrapment has occurred, the predetermined recovery priority may be adopted as is. The restoration priorities set in advance for restoration targets may have three levels of priorities A to C, excluding priority S, for example. For each restoration target, one of A, B, and C is set in advance in descending order of restoration priority.

For example, the restoration priority may be determined according to the functional properties of the building in which the elevator equipment to be restored is installed. For example, a high recovery priority (priority A) may be set for elevator equipment operating in a facility that plays an important role in society, especially when a disaster occurs. Examples of facilities that play an important role include medical institutions such as general hospitals, news organizations such as television stations, and administrative institutions such as city and ward offices. Alternatively, the recovery priority may be set according to the rank of the maintenance contract concluded between the manager of the building where the elevator equipment is installed and the maintenance company.

The number of locations where work is required is information indicating the number of locations where restoration work by a worker is required in the elevator equipment to be restored. The number of required work areas may be, for example, the number of floors (hereinafter referred to as the number of boarding and alighting floors) in which an elevator stops for users to get on and off in the elevator facility. Alternatively, the number of required work areas may be the number of elevators included in the elevator equipment multiplied by the number of floors. For example, in the case of elevator equipment installed in a five-story building and consisting of four elevators that stop at each floor, the number of required work locations may be 4 elevators x 5 floors = 20 locations.

The outage probability is the outage probability estimated by the estimation unit 52 for the non-monitored equipment specified as a restoration target. The recovery target specifying unit 14 can reflect the outage probability estimated by the estimating unit 52 in the item of outage probability.

Note that the stop probability of elevator equipment is information estimated for non-monitored equipment whose operating status is unknown. Therefore, a stop probability is not associated with elevator equipment whose operating status is known through remote monitoring or callbacks.

(About exploration)
FIGS. 26 and 27 are diagrams showing an example of a procedure for searching for an optimal dispatch plan, which is executed by the search unit 12 in the search step S4-2. Note that the maps M1 to M8 shown in FIGS. 26 and 27 are merely information that is visualized to explain the search procedure in an easy-to-understand manner, which is internally held by the search unit 12 when it searches. Therefore, the search unit 12 does not actually have to generate visualization information as shown in maps M1 to M8.

As shown in the map M1, the search unit 12 is given the locations of all recovery targets included in the recovery target list, the presence or absence of confinement of these recovery targets, and the recovery priority. Further, the search unit 12 is given the number of groups in the office A and the position information of the group leader of each group. Since the search unit 12 generates a dispatch pattern with the recovery target as the dispatch target, in the following description, the term "dispatch target" will be used for the recovery target.

As shown in map M2, first, the search unit 12 narrows down to the secondment target with the highest recovery priority, so that the estimated completion time of all the secondment targets becomes early, preferably, the estimated completion time becomes the earliest. Assign each seconded object to a group. Hereinafter, the seconded object with the highest recovery priority, that is, the seconded object with priority S in which confinement has occurred, will be referred to as seconded object S. For example, the search unit 12 may generate several dispatch patterns targeting only the dispatch target S, and search for a dispatch pattern that is an optimal solution or a semi-optimal solution from among them. The search unit 12 may execute a genetic algorithm, which will be described later, to search for a dispatch pattern that is an optimal solution or a semi-optimal solution.

As an example, the map M2 shows that the search unit 12 has searched for a dispatch pattern in which the secondment target OS1 is first assigned to the group W and the secondment target OS2 is first assigned to the group T, as an optimal solution or a semi-optimal solution. ing. The seconded target OS1 is the first seconded target assigned to the group W, and the seconded target OS2 is the first seconded target assigned to the group T. In this way, in the dispatch plan output by the search unit 12, the dispatch order of the dispatch target S in which user confinement has occurred is earlier than the dispatch order of the dispatch targets A to C in which no confinement has occurred. . As a result, a dispatch plan is generated that allows recovery work to be carried out, giving top priority to the recovery target where the user is trapped.

As shown in map M3, next, the search unit 12 virtually shifts the status of all secondment targets S to a recovery work completed state. Furthermore, for each group, the search unit 12 sets the location of the seconded target S who is scheduled to be seconded last as the latest current location.

As shown in the map M4, the search unit 12 then narrows down the secondment targets with priority A next to priority S (hereinafter referred to as secondment targets A), so that the estimated completion time of all secondment targets A will be earlier. Preferably, each of the secondment targets A is assigned to a group so that the estimated completion time is the earliest. Similar to the secondment target S, the search unit 12 generates several dispatch patterns targeting only the secondment target A, and executes a genetic algorithm to be described later from among them to obtain an optimal solution or a semi-optimal solution. You may also search for patterns of secondment.

According to map M4, as an example, two secondment targets A are assigned to group T in the secondment order of secondment target OA2 and secondment target OA3. The seconded target OA2 is the second seconded target assigned to the group T, and the seconded target OA3 is the third seconded target assigned to the group T. Since no secondment target S is assigned to group U, secondment target OA4 is the first seconded target assigned to group U in the secondment order.

As shown in map M5, similarly to the case of priority S, the search unit 12 virtually shifts the status of all secondment targets A to the recovery work completed state. Furthermore, for each group, the search unit 12 sets the location of seconded target A, who is scheduled to be seconded last, as the latest current location.

As shown in map M6, the search unit 12 then narrows down to the secondment targets with priority B next to priority A (hereinafter referred to as secondment targets B), so that the estimated completion time of all secondment targets B will be earlier. Preferably, each of the secondment targets B is assigned to a group so that the estimated completion time is the earliest. The search unit 12 generates several dispatch patterns targeting only the dispatch target B, as well as the dispatch targets S and the dispatch target A, and executes a genetic algorithm to be described later to find the optimal or semi-optimal solution. You may also search for a dispatch pattern that is the optimal solution.

According to the map M6, as an example, three secondment targets B are assigned to the group W in the secondment order of secondment target OB1, secondment target OB2, and secondment target OB3. At this point, secondment target OB1, secondment target OB2, and secondment target OB3 are respectively assigned to group W from third to fifth in the secondment order, following secondment target OS1 and secondment target OA1.

As shown in map M7, similarly to the case of priority S and priority A, the search unit 12 virtually shifts the status of all secondment targets B to the recovery work completed state. Furthermore, for each group, the search unit 12 sets the location of seconded target B, who is scheduled to be seconded last, as the latest current location.

As shown in map M8, the search unit 12 then narrows down to the secondment targets with priority C next to priority B (hereinafter referred to as secondment targets C), so that the estimated completion time of all secondment targets C will be earlier. Preferably, each of the secondment targets C is assigned to a group so that the estimated completion time is the earliest. The search unit 12 generates several secondment patterns targeting only secondment target C, as well as secondment target S, secondment target A, and secondment target B, and executes a genetic algorithm, which will be described later, from among the secondment patterns. A dispatch pattern that provides an optimal solution or a suboptimal solution may be searched.

According to map M8, at this point, group W has secondment targets OS1, secondment targets OA1, secondment targets OB1, secondment targets OB2, secondment targets OB3, and the second to eighth targets in the order of secondment. OC1, secondment target OC2, and secondment target OC3 are respectively assigned. In this way, in the dispatch plan output by the search unit 12, the dispatch targets A to C in which no user is trapped are arranged in the order of dispatch as the recovery priority is higher. As a result, for the remaining recovery targets excluding the recovery target in which the user is trapped, a dispatch plan is generated in which the recovery work can be carried out in descending order of predetermined recovery priority.

As described above, the search unit 12 may search for an optimal dispatch plan for dispatching only restoration targets associated with a specified recovery priority. As a result, a dispatch pattern in which the estimated completion time is relatively early is searched among recovery targets with the same recovery priority, and by concatenating these in order of recovery priority, the recovery targets are sorted in order of predetermined recovery priority. , a dispatch plan that can be used for recovery work is efficiently generated.

Finally, the search unit 12 can obtain the optimal dispatch plan for office A by combining the dispatch patterns searched for each recovery priority from priority S to priority C in descending order of recovery priority.

(About output information)
Below, the output information output from the search unit 12 that searched for the optimal dispatch plan for office A will be explained. For example, the search unit 12 outputs, as output information, the optimal dispatch plan for office A obtained according to the above-described search procedure and the estimated completion time for office A based on the optimal dispatch plan.

- Output information: About the optimal dispatch plan FIG. 28 is a diagram showing an example of the data structure of the optimal dispatch plan. The optimal secondment plan shown in FIG. 28 is the optimal secondment plan for office A, which includes group W, group T, and group U. The optimal dispatch plan includes a group dispatch pattern 71 that defines, for each group, the combination of dispatch targets to which the group performing recovery work is scheduled to dispatch and the order of dispatch. Since office A has three groups, group W, group T, and group U, the optimal dispatch plan for office A includes three group-based dispatch patterns 71. The group-by-group secondment pattern 71 includes information for uniquely identifying the secondment targets assigned to the group, and the secondment order of each assigned secondment target.

Note that the group-by-group dispatch pattern 71 shown in FIG. 28 is merely information visualized so that the combinations of dispatch targets assigned to the groups and the dispatch order can be easily recognized. Therefore, the search unit 12 does not actually need to generate visualization information of the group-based dispatch pattern 71 as shown in FIG. 28.

- Output information: Estimated completion time Furthermore, the search unit 12 may output the estimated completion time of office A in the searched optimal secondment plan together with the optimal secondment plan. In the process of the above-described search, the search unit 12 can estimate the estimated completion time for each dispatch plan based on the map information provided from the map information providing system 300.

For example, the search unit 12 may estimate, for each group, the time at which the restoration work of the dispatch target whose dispatch order is the last will be completed for each of the dispatch plans of a plurality of dispatch patterns. Then, the search unit 12 sets the latest time among the estimated times for each group as the estimated completion time in the above-mentioned dispatch plan. This makes it possible to grasp the time of complete restoration at the business office. Hereinafter, the estimated time for each group at which the recovery work for the last dispatched object will be completed will be referred to as the estimated completion time for each group.

For example, in the optimal dispatch plan shown in FIG. 28, it is assumed that the estimated completion time for each group of group W, that is, the estimated time when the recovery work of dispatch target OC3, which is the last dispatched person in the order of dispatch, will be completed is 10:48. Assume that the estimated completion time for each group of group T, that is, the time when group T is estimated to complete the recovery work of the second dispatch target OC6, which is the last dispatched target OC6, is 10:26. Assume that the estimated completion time for each group of group U, that is, the time when group U is estimated to complete the recovery work for the second dispatch target OC8, which is the last dispatched target OC8, is 10:12. In this case, the search unit 12 determines "10:48", which is the latest time of the three estimated completion times for each group, as the estimated completion time for the entire office A in the optimal dispatch plan shown in FIG. do. The search unit 12 may output the estimated completion time "10:48" together with the optimal dispatch plan shown in FIG.

As an example, the search unit 12 calculates the estimated completion time for each group to find the estimated completion time for the business office based on the total travel time for the group to go to the target, and the total travel time required for the group to perform the recovery work for the target. Estimate at least taking into account the working time. The total travel time is, for example, the total travel time required for a group to move to a secondment target assigned to the group in accordance with the secondment order. The total work time is, for example, the sum total of work time required for recovery work for each target to which the group is dispatched. For example, the search unit 12 may calculate the estimated completion time for each group by adding the above-mentioned total travel time and the above-mentioned total work time to the current time.

The procedure in which the search unit 12 calculates the total travel time of a group will be explained using an example in which the search unit 12 calculates the total travel time of the group W based on the optimal dispatch plan shown in FIG. 28.

First, the search unit 12 calculates the travel distance from the current location of the group leader of group W (map M1 in FIG. 26) to the location of the first seconded target assigned to group W (seconded target OS1 in map M2). Obtain the corresponding travel time MT1. Note that the travel time acquired by the search unit 12 may be acquired from the map information providing system 300 by the travel time acquisition unit 13. Moreover, the above-mentioned travel time may be an average travel time based on the travel distance, or may be a travel time that takes traffic conditions into consideration. The same holds true for various travel times acquired by the search unit 12 later.

Thereafter, the search unit 12 repeats the process of acquiring the travel time MTn from the location of the n-1st seconded target to the location of the nth seconded target (n is an integer of 2 or more). In the example of group W, the search unit 12 repeats the above-described process until it obtains the travel time MT8 from the seconded object OC2 to the seconded object OC3, which is the last movement route.

Then, the search unit 12 calculates the sum of the travel times MT1, MT2, . . . , MTn as the total travel time of the group W.

A procedure in which the search unit 12 calculates the total working time of a group will be explained. The search unit 12 specifies the working time WT for each seconded target according to the number of work required locations determined for each seconded target. In this way, by accurately estimating the work time to be restored according to the situation at the site, it is possible to further reduce the error in the estimated completion time.

The search unit 12 can refer to the recovery target list (FIG. 25) generated by the recovery target identification unit 14 and identify the number of locations where work is required for the seconded target (recovery target). Note that the seconded object to be restored can be uniquely identified by, for example, the object ID.

The search unit 12 can calculate the working time WT of one seconded employee based on, for example, the following formula.
Formula: Working time WT = Number of places where work is required x Unit time + Fixed time Unit time is the average work time required for recovery for one work place, and is, for example, 3 minutes. The fixed time is a fixed work time required for restoration of one restoration target, that is, one elevator facility, regardless of the scale of the number of places where work is required, and is, for example, 5 minutes.

For example, if the number of locations where work is required for the first OS1 to be seconded in group W is 20, the search unit 12 calculates the work time WT1 for the OS1 to be seconded as 20 locations x 3 minutes + 5 minutes = 65 minutes. do. In this way, the search unit 12 calculates the work time WT for each of all the secondment targets up to the nth assigned to the group W (work time WT1, WT2, . . . , WTn).

Then, the search unit 12 calculates the sum of the work times WT1, WT2, .

Finally, the search unit 12 calculates the estimated completion time for each group of group W by adding the total travel time calculated as described above and the total work time to the current time. The search unit 12 similarly calculates the estimated completion time for each group for the other groups T and U of the office A. Then, the latest time of the three estimated completion times for each group is set as the estimated completion time for office A in the optimal dispatch plan shown in FIG. 28.

The search unit 12 stores the optimal secondment plan for office A shown in FIG. 28 and the estimated completion time of the optimal secondment plan for office A estimated as described above in the latest secondment plan DB 31. The search unit 12 searches for the optimal secondment plan, calculates the estimated completion time of the optimal secondment plan, and stores these in the latest secondment plan DB 31 for each office corresponding to the disaster that has occurred. to be executed.

(Example of search: genetic algorithm)
The search unit 12 may generate each of the dispatch plans for a plurality of dispatch patterns by generational change according to a genetic algorithm.

For example, the search unit 12 executes the following steps: (Step 1) initialization, (Step 2) evaluation, (Step 3) selection, (Step 4) crossover, and (Step 5) mutation. The search unit 12 repeats each step from step 2 to step 5 until a pre-specified number of generations has passed. The number of generations may be specified in advance by the operator.

In the initial setting of step 1, the search unit 12 randomly generates individuals so as to satisfy the first condition that at least one secondment target is assigned to all groups within the office, for example. The above-mentioned first condition may be specified in advance by the operator, for example.

In this embodiment, identification of dispatch patterns according to a genetic algorithm is performed for each recovery priority. Therefore, an individual is generated as information indicating to which group of a business establishment a predetermined number of secondment targets with a certain recovery priority are assigned.

When allocating Y groups to X secondment targets, the search unit 12 randomly generates individuals having the data structure shown below (X and Y are integers of 2 or more).
[G ₁ , G ₂ ,..., G _x ]
Here, G _i is a value that identifies the group to be assigned to the i-th seconded target, and is an integer greater than or equal to 0 and less than Y (i is an integer greater than or equal to 1 and less than or equal to X).

For example, when assigning three groups (first to third groups) to ten secondment targets (first to tenth secondment targets), an example of the individuals generated by the search unit 12 is [1, 0, 2, 2, 1, 0, 0, 2, 0, 1]. The three numbers "0", "1", and "2" indicate the first group, the second group, and the third group, respectively. In other words, this individual assigns the second group to the first, fifth, and tenth seconded targets, assigns the first group to the second, sixth, seventh, and ninth seconded targets, and assigns the third, This indicates that the third group is assigned to the fourth and eighth seconded targets.

The search unit 12 generates a plurality of different individuals (for example, 300 individuals) and sets these as an initial population. Note that the number of initial populations (second condition) may be specified in advance by the operator.

In the evaluation of step 2, the search unit 12 calculates, for each generated individual, the estimated completion time at which each group will complete the recovery work for all the dispatched targets, and calculates the estimated completion time of the latest group. It is associated as the estimated completion time of the individual. As described above, the estimated completion time for each group is obtained by adding the sum of the travel time MT and the sum of work time WT when visiting all the assigned targets to the current time.

In the selection step 3, the search unit 12 uses a tournament method to compare the estimated completion times of each individual, and selects the top few (for example, 3) individuals with the earliest estimated completion times as the parent individuals of the next generation. do.

In the intersection in step 4, the search unit 12 randomly executes intersection on the selected parent individuals so that the first condition is satisfied.

In the mutation in step 5, the search unit 12 causes each individual to undergo mutation with a predetermined probability (for example, 5%). The probability of causing a mutation may be specified in advance by the operator.

The search unit 12 repeats the processes from step 2 to step 5 for the number of generations specified in advance, and selects the remaining individuals after the repetition as the optimal or semi-optimal group for the dispatch target group with a certain restoration priority. Output as a secondment pattern.

The search unit 12 executes a search according to the genetic algorithm consisting of steps 1 to 5 above for each recovery priority, and outputs an optimal or semi-optimal dispatch pattern for each recovery priority.

<Screen example of optimal dispatch plan>
The output control unit 15 provides the front end system 1 or the office device 2 with a dispatch plan screen for presenting to the operator the optimal dispatch plan for each office found by the search unit 12 in step S4. Good too. FIG. 29 is a diagram showing an example of a secondment plan screen showing details of an optimal secondment plan for one office. The output control unit 15 generates the dispatch plan screen 700 based on the optimal dispatch plan stored in the latest dispatch plan DB 31 and the dispatch confirmation target managed by the dispatch arrangement system 20. The output control unit 15 transmits the generated secondment plan screen 700 to an output unit such as the front end system 1 or the office device 2, for example. The secondment plan screen 700 is a screen for presenting the optimal secondment plan searched for office A, as an example.

The secondment plan screen 700 may include basic information 701 about the office. The basic information 701 includes, for example, the number of groups belonging to the business office, the number of all workers assigned to the business office, elevators and escalators for which restoration work has not been completed among the recovery targets assigned to the business office. It may also include the number of equipment, etc. In the present embodiment, the basic information 701 may include an estimated number of outages as the number of non-monitored facilities estimated to be out of service by the driving status estimating device 5.

Further, the basic information 701 may include an estimated completion time. The estimated completion time is estimated by the search unit 12 based on the optimal dispatch plan searched by the search unit 12. The estimated completion time refers to the time at which the entire business establishment is expected to complete the restoration work for all restoration targets if the business establishment carries out the restoration work with its current personnel according to the optimal dispatch plan.

The basic information 701 may be designed to represent the latest information regarding basic information regarding the system and progress of recovery work at the business office. For example, the secondment plan screen 700 may include a latest button 702. When the latest button 702 is operated by the operator, an update instruction is sent from the front end system 1 or office device 2 to the dispatch support system 100. Upon receiving the update instruction, the secondment planning system 10 of the secondment support system 100 reads the latest optimal secondment plan for office A from the latest secondment plan DB 31, and calculates the number of secondment targets whose work has not been completed. Count the number of incomplete items. The output control unit 15 provides the front end system 1 or the office device 2 with a secondment plan screen 700 on which the latest number of incomplete items is reflected.

When the operating status list by business establishment or the estimated operating status is updated in accordance with a change in the situation, the recovery target specifying unit 14 updates the recovery target list. In response to the operation of the latest button 702, the recovery target identification unit 14 updates the number of recovery targets or the estimated number of stops displayed in the basic information 701 to the number counted based on the updated list of recovery targets. It's okay.

When the search unit 12 updates the optimal dispatch plan stored in the latest dispatch plan DB 31, in response to the operation of the latest button 702, the search unit 12 changes the estimated completion time displayed in the basic information 701 to the updated one. It may be updated to the estimated completion time in the optimal dispatch plan.

The secondment plan screen 700 may include a list box 703 that shows the secondment pattern by group, that is, the combination of secondment targets assigned to the group and the secondment order. The list box 703 may include the order of secondment, the name of the property to be seconded, and the like. Further, the list box 703 may reflect the work progress of each seconded object. In the illustrated example, the background color of the dispatch order is color-coded depending on the work progress. The work progress of each secondment target is managed by the secondment arrangement system 20.

The secondment plan screen 700 may include a map image 704 in which information about the secondment target is reflected on a map. As an example, the map image 704 may reflect information on secondment targets assigned to the specific group selected on the secondment plan screen 700. In the illustrated example, list box 703 for group U is selected. Therefore, on the map image 704, icons of secondment targets assigned to group U may be plotted together with a boundary line representing the area in charge of group U. A number indicating the order of dispatch may be given to the icon of the dispatch target.

By displaying the dispatch plan screen 700 as described above on the front end system 1 or office device 2, the operator can grasp the progress of recovery work for each office in real time.

<Modified example>
(Narrowing down recovery targets)
The recovery target specifying unit 14 may specify elevator equipment located within the designated area as a recovery target. As a result, the non-monitored facilities to be considered in planning the dispatch plan are narrowed down from a huge number of non-monitored facilities to the elevator facilities located within the designated area. Therefore, it is possible to prevent elevator equipment in areas where there is little need for dispatch from being mixed as targets for dispatch, thereby preventing an inefficient dispatch plan from being drawn up. Operators can determine areas where there is a high need for dispatch and specify those areas, making it easy to make efficient dispatch plans that take into account the operational status of equipment that is not subject to monitoring only in that area. It becomes possible to make a plan.

The elevator equipment is associated with a restoration priority, and the restoration target specifying unit 14 may identify the elevator equipment associated with the designated restoration priority as a restoration target. As a result, the non-monitored facilities considered in planning the dispatch plan are narrowed down from the extremely large number of non-monitored facilities to the elevator facilities with the specified recovery priority. Therefore, it is possible to prevent an inefficient dispatch plan from being created in which elevator equipment in areas with low recovery priority and low need for dispatch are mixed as dispatch targets. On the other hand, since the operating status is taken into consideration only for non-monitored facilities that have a high restoration priority, it becomes possible to easily formulate an efficient dispatch plan.

(Estimation of driving situation)
The estimation unit 52 uses training data including location information indicating the location of the elevator equipment, disaster information including the degree of disaster at the location, and operating status information indicating the operating status of the elevator equipment when a disaster occurs. Using a learning model generated by performing machine learning, the operating status of the non-monitored equipment may be estimated using the location information of the non-monitored equipment and disaster information as input data.

For example, if the disaster is an earthquake, the degree of disaster included in the disaster information may include (1) the seismic intensity at the location of the elevator equipment, or (2) the magnitude of the epicenter, the distance from the epicenter, and the depth of the epicenter. It may also be the magnitude of shaking at the location, which is calculated from the location. Alternatively, the degree of disaster may be (3) characteristics of shaking at the location, such as pitching and rolling.

The teacher data further includes peripheral operating status information indicating the operating status of peripheral equipment that is monitored equipment located within a predetermined range from the elevator equipment when a disaster occurs, and the input data is The information may further include operating conditions obtained by remotely monitoring peripheral equipment located within the range.

The teacher data may further include equipment information indicating at least one of the type and specification of the elevator equipment, and the input data may further include equipment information of equipment not to be monitored.

Even if the teacher data further includes building information indicating at least one of the completion date and seismic specifications of the building in which the elevator equipment is installed, and the input data further includes building information about the attention elevator equipment (facilities not subject to monitoring), good.

The building information indicates, for example, the year of construction (completion), the type of structure, earthquake resistance specifications, etc. The structure type may be information indicating whether the building in which the elevator equipment is installed falls under a reinforced concrete structure (RC structure), a steel frame structure (S structure), or a steel reinforced concrete structure (SRC structure). . The seismic specifications may be information indicating whether the building in which the elevator equipment is installed is seismically isolated, earthquake resistant (earthquake resistance grade for houses only), specifications compliant with each local government, and the like.

For example, it can be determined whether the old earthquake resistance standards or the new earthquake resistance standards are applied based on the completion date. In addition, it is possible to determine the seismic isolation structure and earthquake-resistant structure of a building from the seismic specifications. For example, it is possible to determine the seismic resistance class 1 to 3 of a house, or to determine the history of seismic reinforcement work carried out to meet the new seismic standards in buildings that meet the old earthquake resistance standards. In this way, by considering the disaster resistance of the building in which the elevator equipment is installed, it becomes possible to estimate with even greater precision the operating status of the elevator equipment installed in the building at the time of a disaster.

The learning model may output the probability of the operating status of the non-monitored equipment. The probability may be an outage probability that indicates the probability that the non-monitored equipment is out of service.

<Effect>
According to the dispatch planning system 10 of the present invention, even if a large number of events such as outages or failures occur simultaneously in a wide area due to a disaster such as an earthquake, the time when all of the recovery work by a plurality of work groups will be completed can be determined. It is possible to formulate a plan for early dispatch.

Moreover, according to the dispatch planning system 10 of the present invention, not only the monitored equipment whose operating status can be immediately grasped, but also non-monitored equipment whose operating status cannot be grasped are restored based on the estimated operating status. It is possible to judge whether work is necessary or not. Therefore, it becomes possible to generate an accurate dispatch plan in accordance with the disaster situation. As a result, it is possible to improve the accuracy of the work group's dispatch plan. In this way, it becomes possible to accurately instruct each group or each worker to carry out recovery work in accordance with an accurate dispatch plan that takes into account the operating status of the non-monitored equipment.

Further, by utilizing the estimated completion time determined for each dispatch plan, it is possible to present the outlook for recovery to the relevant parties. For example, it is possible to provide the estimated completion time obtained from an accurate dispatch plan that takes into account the operating status of non-monitored equipment to users, owners, etc. who are looking forward to the restoration of the equipment to be restored. By presenting the outlook for recovery in this way, it is possible to give a sense of security to the users or owners who are subject to recovery.

As described above, the dispatch planning system 10 of the present invention is a system that brings about an excellent effect on early recovery from a disaster for facilities such as elevators that are the basis of daily life. Therefore, the dispatch planning system 10 of the present invention can contribute to the development of high-quality infrastructure with excellent disaster mitigation or prevention performance, and can further contribute to the Sustainable Development Goals (SDGs), such as Goal 11 ( We can contribute to achieving the goal of creating a city where people can continue to live.

[Example of implementation using software]
FIG. 30 is a block diagram showing an example of the hardware configuration of the learning model generation device 4, the learning model generation device 4a, the driving situation estimation device 5, and each device or system included in the maintenance system 1000.

The functions of one or more information processing devices that realize the learning model generation device 4, the learning model generation device 4a, the driving situation estimation device 5, the dispatch planning system 10, and the dispatch arrangement system 20 cause a computer to function as the information processing device. This can be achieved using a program for This program can be realized by a program for causing a computer to function as each control block of the above-mentioned information processing device. The above-mentioned control blocks include, in particular, each part of the control unit 40, each part of the control unit 40a, each part of the control unit 50, the driving status acquisition unit 11, the search unit 12, the travel time acquisition unit 13, the recovery target identification unit 14, They are an output control section 15, a disaster information acquisition section 16, and a confinement information acquisition section 17.

In this case, the information processing device includes a computer 910 having at least one control device (for example, a processor 911) and at least one storage device (for example, a memory 912) as hardware for executing the above-described program. . The memory 912 stores, for example, a program 921 as the above-mentioned program. By executing the program using the above-described control device and storage device, each function described in each of the above-described embodiments is realized.

The above-mentioned program may be recorded on one or more computer-readable recording media (for example, recording medium 931) instead of being temporary. This recording medium may or may not be included in the above-described information processing apparatus. In the latter case, the above program may be supplied to the information processing device via any wired or wireless transmission medium.

Moreover, part or all of the functions of each control block described above can also be realized by a logic circuit. For example, an integrated circuit in which a logic circuit functioning as each of the control blocks described above is formed is also included in the scope of the present invention. In addition to this, it is also possible to realize the functions of each of the above-mentioned control blocks using, for example, a quantum computer.

Moreover, each process described in each of the above-mentioned embodiments may be performed by AI (Artificial Intelligence). In this case, the AI may operate on the control device or may operate on another device (for example, an edge computer or a cloud server).

The present invention is not limited to the embodiments described above, and various modifications can be made within the scope of the claims, and embodiments obtained by appropriately combining technical means disclosed in different embodiments. are also included within the technical scope of the present invention.

1 Front-end system 2 Office equipment 3 Terminal equipment (information terminal)
4, 4a Learning model generation device 5 Operating status estimation device 6a Disaster record database 6b Elevator equipment installation information database 6c Operating status information database 6d Disaster occurrence information distribution device 6e Elevator equipment remote monitoring device 9 Display device 10 Dispatch planning system 11 Operating status acquisition Section 12 Search section 13 Travel time acquisition section 14 Recovery target identification section 15 Output control section 16 Disaster information acquisition section 17 Confinement information acquisition section 20 Dispatch arrangement system 31 Latest dispatch plan database 40, 40a, 50 Control section 41, 41a Teacher data generation Units 42, 42a Learning units 45, 54 Storage unit 51 Input data generation unit 52 Estimation unit 53 Output control unit 100 Dispatch support system 200 Damage information provision system 300 Map information provision system 451, 451a Learning model 910 Computer (information processing device)
911 Processor (control device)
912 Memory (storage device)
921 Program 931 Recording medium

Claims (20)

an operating status acquisition unit that acquires operating status from monitored equipment, which is the elevator equipment whose operating status can be remotely monitored among a plurality of elevator equipment located at different locations;
an operating status estimating unit that estimates the operating status of non-monitored equipment, which is the elevator equipment whose operating status cannot be remotely monitored among the elevator equipment;
a restoration target identification unit that identifies restoration targets from among the elevator equipment;
An optimal solution or a sub-optimal solution of a dispatch plan that determines for each group the combination of dispatch targets that are the recovery targets to which a group consisting of one or more workers and who performs recovery work is scheduled to dispatch, and the dispatch order. Equipped with a search department that searches for dispatch plans,
The recovery target specifying unit specifies the recovery target from among the monitoring target equipment according to the operating status obtained when the disaster occurs, and also specifies the recovery target from the monitoring target equipment when the disaster occurs, based on the degree of the disaster. identifying the restoration target from among the non-monitored equipment according to the estimated operating status;
The search unit is
In accordance with the number of required work locations determined for each seconded target, specify the work time required for recovery work for each seconded target,
The total work time for each group calculated based on the work time required for recovery work for each seconded target, and the total movement for each group, which is the sum of the travel time required to move the seconded targets in the order of secondment. Calculate the estimated completion time for each group by adding the time and to the current time,
searching for a dispatch plan whose estimated completion time is relatively early as the optimal dispatch plan;
Secondment planning system. The number of required work areas is the number of floors where elevators stop for users to get on and off in the elevator equipment to be restored, or the product of the number of elevators included in the elevator equipment and the number of floors. and
The search unit is
Calculate the work time of the seconded person by adding a fixed time to the product of the number of required work locations and the unit time,
The unit time is the work time required to restore one work required location,
The dispatch planning system according to claim 1, wherein the fixed time is a fixed working time required for restoration of one of the elevator equipment. The operating status estimating unit estimates the probability of the operating status of the non-monitored equipment,
The dispatch planning system according to claim 1 or 2, wherein the recovery target specifying unit specifies the non-monitored equipment whose probability exceeds a specified value as the recovery target. It is further equipped with a disaster information acquisition unit that acquires information indicating the degree of disaster in each region.
The dispatch plan according to any one of claims 1 to 3, wherein the recovery target specifying unit specifies the elevator equipment located in a specified area where a disaster of the degree has occurred as the recovery target. system. The dispatch planning system according to any one of claims 1 to 4, wherein the restoration target specifying unit identifies the elevator equipment located within a designated area as the restoration target. The elevator equipment is associated with a recovery priority,
The dispatch planning system according to any one of claims 1 to 5, wherein the recovery target specifying unit specifies the elevator equipment associated with the specified recovery priority as the recovery target. The operating status estimating unit includes location information indicating the location of the elevator equipment, disaster information including the degree of disaster at the location, and operating status information indicating the operating status of the elevator equipment when the disaster occurred. The operating status of the non-monitored equipment is determined using the location information of the non-monitored equipment and the disaster information as input data using a learning model generated by performing machine learning using training data including the The dispatch planning system according to any one of claims 1 to 6, which estimates. The teacher data further includes peripheral operating status information indicating the operating status of peripheral equipment that is the monitoring target equipment located within a predetermined range from the elevator equipment when the disaster occurs,
8. The dispatch planning system according to claim 7, wherein the input data further includes operating conditions obtained by remotely monitoring the peripheral equipment located within the predetermined range from the non-monitored equipment. The teacher data further includes equipment information indicating at least one of a model and a specification of the elevator equipment,
The dispatch planning system according to claim 7 or 8, wherein the input data further includes the equipment information of the non-monitored equipment. The teacher data further includes building information indicating at least one of the completion date and seismic specifications of the building in which the elevator equipment is installed,
The dispatch planning system according to any one of claims 7 to 9, wherein the input data further includes the building information of the non-monitored equipment. The dispatch planning system according to any one of claims 7 to 10, wherein the learning model outputs a probability of the operating status of the non-monitored equipment. The search unit estimates the completion of all restoration of the secondment target, which is estimated by taking into account at least the total travel time for the secondment target and the total work time required for the recovery work of the secondment target. The dispatch planning system according to any one of claims 1 to 9, wherein the dispatch plan having a relatively early time among the plurality of different patterns of dispatch plans is searched as the optimal dispatch plan. For each of the plurality of patterns of dispatch plans, the searching unit estimates, for each group, the time when the recovery work of the dispatch target whose dispatch order is the last is completed, and selects the earliest of the estimated times. The dispatch planning system according to claim 12, wherein the estimated completion time is a late time. The secondment planning system according to claim 12 or 13, wherein the search unit generates each of the plurality of patterns of secondment plans by generational change according to a genetic algorithm. further comprising a confinement information acquisition unit that acquires information indicating whether or not a user is confined in the recovery target,
Any one of claims 1 to 14, wherein the dispatch plan is such that the dispatch order of the dispatch targets in which user confinement has occurred is earlier than the dispatch order of the dispatch targets in which no confinement has occurred. The secondment planning system described in Section 1. further comprising a confinement information acquisition unit that acquires information indicating whether or not a user is confined in the recovery target,
A recovery priority is associated with the recovery target,
The dispatch plan system according to any one of claims 1 to 15, wherein in the dispatch plan, the dispatch targets for which no user is trapped are arranged in the order in which the dispatch order is earlier as the recovery priority is higher. . The dispatch planning system according to any one of claims 1 to 16, wherein the operating status includes at least normal operation and stoppage. The disaster is an earthquake;
The degree of the disaster is (1) the seismic intensity at the location, or (2) the magnitude of the shaking at the location, calculated from the magnitude of the epicenter, the distance from the epicenter, and the depth of the epicenter. The dispatch planning system according to any one of claims 1 to 17. A control method executed by one or more information processing devices, the method comprising:
an operation status acquisition step of acquiring an operation status from a monitoring target facility, which is the elevator equipment whose operation status can be remotely monitored among a plurality of elevator equipment located at different locations;
an operating status estimation step of estimating the operating status of non-monitored equipment, which is the elevator equipment whose operating status cannot be remotely monitored among the elevator equipment;
a restoration target identification step of identifying restoration targets from the elevator equipment;
An optimal solution or a sub-optimal solution of a dispatch plan that determines for each group the combination of dispatch targets that are the recovery targets to which a group consisting of one or more workers and who performs recovery work is scheduled to dispatch, and the dispatch order. a search step of searching for a secondment plan;
In the recovery target identification step, the recovery target is identified from among the monitoring target equipment according to the operating status obtained when the disaster occurs, and the recovery target is identified based on the estimated operating status when the disaster occurs. Accordingly, identifying the restoration target from among the non-monitored equipment,
In the search step,
In accordance with the number of required work locations determined for each seconded target, specify the work time required for recovery work for each seconded target,
The total work time for each group calculated based on the work time required for recovery work for each seconded target, and the total movement for each group, which is the sum of the travel time required to move the seconded targets in the order of secondment. Calculate the estimated completion time for each group by adding the time and to the current time,
searching for a dispatch plan whose estimated completion time is relatively early as the optimal dispatch plan;
Control method. 2. A program for causing a computer to function as the dispatch planning system according to claim 1, wherein the program causes the computer to function as the driving situation acquisition section, the driving situation estimation section, the recovery target identification section, and the search section. program.

Images