JP7612030B2 - Surveillance system and method - Google Patents

Description

The present disclosure relates to a monitoring system and a monitoring method for monitoring the provision of elevator maintenance services.

In recent years, service level agreements (SLAs) have been established for the provision of elevator maintenance services. For example, in the facility remote monitoring system described in JP 2019-191662 A (Patent Document 1), evaluation items are defined for each maintenance service provided, and the target value of the quality level to be provided for each evaluation item is predefined as an SLA contract.

JP 2019-191662 A

In the remote equipment monitoring system described in Patent Document 1, the scope of the SLA covers the period from the start to the completion of the maintenance service. However, no consideration was given to improving the quality level for the entire maintenance service (time management from the start of the maintenance service to the delivery of the maintenance results to the customer).

This disclosure has been made to solve such problems, and its purpose is to provide technology that can improve the compliance rate with the SLA covering the entire maintenance service in a maintenance service that specifies an SLA, thereby increasing customer satisfaction.

The monitoring system according to the present disclosure monitors the provision of elevator maintenance services. The monitoring system includes a memory unit, a prediction unit, and a display unit. The memory unit stores a prediction model that predicts a delay state of a work period required for multiple tasks from the start of the maintenance service to the completion of reporting the results of the maintenance service to the customer, and the actual period for each of the multiple tasks relative to a reference period. The prediction unit predicts the delay state based on the prediction model and the actual period for each of the multiple tasks. The display unit displays the delay state predicted by the prediction unit. The prediction model is a model that predicts the delay state using the difference between the reference period for each of the multiple tasks and the actual period for the reference period.

The monitoring method according to the present disclosure is a method for monitoring the provision of elevator maintenance services. The monitoring method includes a step of storing a prediction model for predicting a delay state of a work period required for a plurality of tasks from the start of the maintenance service to the completion of reporting the results of the maintenance service to the customer, a step of storing an actual period for each of the plurality of tasks relative to a reference period, a step of predicting the delay state based on the prediction model and the actual period for each of the plurality of tasks, and a step of displaying the delay state predicted by the predicting step. The prediction model is a model for predicting the delay state using the difference between the reference period for each of the plurality of tasks and the actual period for the reference period.

According to the present disclosure, in a maintenance service that specifies an SLA, it is possible to improve the compliance rate with the SLA covering the entire maintenance service, thereby increasing customer satisfaction.

1 is a diagram illustrating an overview of a monitoring system according to a first embodiment. FIG. 2 is a diagram illustrating an example of a hardware configuration of the monitoring system according to the first embodiment. FIG. 1 is a diagram illustrating an example of a functional block diagram of a monitoring system according to a first embodiment. 4 is a flowchart of a process executed by the monitoring system according to the first embodiment. FIG. 11 is a diagram for explaining a delay state of a work period according to the first embodiment. FIG. 11 is a diagram for explaining a delay state of a work period according to the first embodiment. FIG. 11 is a diagram for explaining a delay state of a work period according to the first embodiment. FIG. 11 is a diagram for explaining a delay state of a work period according to the first embodiment. FIG. 11 is a diagram for explaining a delay state of a work period according to the first embodiment. FIG. 13 is a diagram illustrating an example of a functional block diagram of a monitoring system according to a second embodiment. 10 is a flowchart of a process executed by a monitoring system according to a second embodiment.

Hereinafter, the embodiment will be described with reference to the drawings. In the following description, the same parts are given the same symbols. Their names and functions are also the same. Therefore, detailed descriptions thereof will not be repeated.

[First embodiment]
First, a monitoring system 1 according to the first embodiment will be described. Fig. 1 is a diagram illustrating an overview of the monitoring system 1 according to the first embodiment. Fig. 2 is a diagram illustrating an example of a hardware configuration of the monitoring system according to the first embodiment.

The monitoring system 1 is a system that monitors the provision of elevator maintenance services. Below, an elevator is used as an example of an elevator, and elevator maintenance services are explained. The monitoring system 1 includes a monitoring device 200, a maintenance company terminal 400, and an elevator control device 100.

In this embodiment, the contract between the customer and the maintenance company specifies the entire elevator maintenance service as an evaluation item of the SLA. The monitoring device 200 manages the delay status for the entire elevator maintenance service. Here, "the entire maintenance service" refers to the period from the start of the maintenance service (also referred to as "maintenance work") to the completion of reporting the results of the maintenance service to the customer (also referred to as "maintenance result report").

As part of the SLA contract, a target value for the period from the start of the maintenance service to the completion of the maintenance result report ("target number of days X" described below) is set. In this embodiment, the monitoring device 200 manages the delay status of the entire maintenance service, alerting users to work delays and encouraging them to take action to meet the deadline for the maintenance result report. This can improve the compliance rate with the SLA.

Here, it is assumed that the maintenance company has a contract (maintenance service contract) with the customer, who is the owner of building 10, to periodically inspect the elevators of building 10. For example, if an SLA contract is made with a target number of days X = 28 days, the monitoring device 200 manages the delay status of work so that the inspection results can be reported within 28 days. The specific method for this is described below.

In this example, building 10 is a four-story building owned by a customer. An elevator is installed in building 10. Elevator car 341 moves in hoistway 351. Cage 341 can stop at each floor, and passengers can board car 341 from landing 331 on the first floor to landing 334 on the fourth floor. An elevator control device 100 is installed in the machine room of building 10. Elevator control device 100 controls car 341.

In this embodiment, the elevator control device 100 can be connected to the monitoring device 200 via a network. The monitoring device 200 is a device that monitors the provision of elevator maintenance services. The monitoring device 200 is installed, for example, in an information center 20 of a maintenance company.

The maintenance company terminal 400 can be connected to the monitoring device 200 via a network. The maintenance company terminal 400 is installed, for example, at a branch office 40 of the maintenance company. The maintenance company terminal 400 is a terminal for inputting information related to elevator maintenance services.

In this example, the customer is in building 10 and has access to a customer terminal 300 owned by the customer. The customer terminal 300 can be connected to the monitoring device 200 via a network.

2, the monitoring device 200 has a CPU (Central Processing Unit) 211, a ROM (Read Only Memory) 212, a RAM (Random Access Memory) 213, a storage unit 214, a communication interface 215, and an I/O interface 216. These are connected to each other via a bus so that they can communicate with each other.

The CPU 211 provides overall control of the entire monitoring device 200. The CPU 211 deploys a program stored in the ROM 212 into the RAM 213 and executes it. The ROM 212 stores a program that describes the processing procedures for the processing performed by the monitoring device 200.

The RAM 213 serves as a working area when the CPU 211 executes a program, and temporarily stores the program and data required for executing the program. The storage unit 214 is a non-volatile storage device, such as a hard disk drive (HDD) or a solid state drive (SSD).

The communication interface 215 controls communication with the elevator control device 100, the maintenance company terminal 400, and the customer terminal 300. The monitoring device 200 can communicate with these devices or systems via the communication interface 115.

The I/O interface 216 is an interface for connecting the CPU 211 to an external device. For example, the CPU 211 can be connected to a display device such as a display, or an input device such as a keyboard or a mouse via the I/O interface 216.

Although not shown, the customer terminal 300 and the maintenance company terminal 400, like the monitoring device, also have a CPU, ROM, RAM, a memory unit, a communication interface, and an I/O interface.

The elevator control device 100 has a CPU 111, a ROM 112, a RAM 113, and a communication interface 115. These are connected to each other so that they can communicate with each other via a bus.

The CPU 111 provides overall control of the entire elevator control device 100. The CPU 111 deploys a program stored in the ROM 112 into the RAM 113 and executes it. The ROM 112 stores a program that describes the processing procedures for the processes performed by the elevator control device 100.

RAM 113 serves as a working area when CPU 111 executes programs, and temporarily stores programs and data required for executing programs.

The communication interface 115 controls communication with the car 341, devices installed at the landings on each floor (landing devices including call buttons), and the monitoring device 200. The elevator control device 100 can communicate with these devices etc. via the communication interface 115.

The elevator control device 100 can be connected to a maintenance computer 30 carried by a maintenance worker 21. The maintenance computer 30 is a terminal (e.g., a tablet terminal) used during maintenance work and can be connected to the elevator control device 100.

By connecting to the elevator control device 100, the maintenance computer 30 can acquire various information (signals) held by the elevator control device 100, change the settings of the elevator control device 100, etc.

Returning to Figure 1, when starting maintenance work, the maintenance worker 21 connects the maintenance computer 30 to the elevator control device 100.

The elevator control device 100, by connecting to the maintenance computer 30, issues alert data indicating that maintenance work has started. The monitoring device 200 receives and records this alert data via the network (details will be described later). After that, in this example, the maintenance worker 21 performs inspection work at the landing 333 on the third floor.

A number of tasks must be performed from the start of the maintenance service to the completion of the maintenance result report. In this embodiment, N tasks (e.g., 10 tasks) must be performed as the number of tasks. The names of the N tasks are Flow 1, Flow 2, Flow 3, ..., Flow N.

More details will be given later using Figures 4 to 6, but flow 1 is the maintenance work performed by a maintenance technician. Flow 2 is the clerical work performed by an office worker. Flow 3 is the confirmation work performed by a checker. And the final flow N is the reporting work performed by a reporter. Once the reporter has performed the reporting work, the customer can view the maintenance work report on the customer terminal 300 in building 10.

The input operations for flows 1 to N are performed using the maintenance company terminal 400. The maintenance company terminal 400 may be multiple terminals owned by multiple workers. The data input at the maintenance company terminal 400 is stored in the memory unit 214 of the monitoring device 200 via the network.

The monitoring device 200 predicts the delay status for flows 1 to N. The predicted delay status can be confirmed on a display device connected to the maintenance company terminal 400 or a display device connected to the monitoring device 200. The monitoring device 200 and the maintenance company terminal 400 may be configured as a single device.

Next, we will explain the various processes executed by the monitoring system 1. Figure 3 is a diagram showing an example of a functional block diagram of the monitoring system 1 according to the first embodiment.

3, the monitoring system 1 includes a prediction unit 130, a display unit 131, and a generation unit 132. The memory unit 214 stores information such as a first prediction model 191, implementation history information 192, alarm data 193, and past history information 194. Some or all of the information stored in the memory unit 214 may be stored in a storage device provided in the maintenance company terminal 400, or in a storage device provided in another external device (such as a database server).

The various processes executed by the monitoring system 1 may be executed, for example, by a user operating an input device connected to the maintenance company terminal 400 or the monitoring device 200. The various processes executed by the monitoring system 1 may be executed by either the monitoring device 200 or the maintenance company terminal 400, or some of the processes may be executed by the monitoring device 200 and the rest by the maintenance company terminal 400.

The monitoring device 200 may be a device equipped with a web server that can be connected from the maintenance company terminal 400 by accessing a specific URL. By accessing the specific URL, the input screen for flows 1 to N and the delay status predicted by the monitoring device 200 may be displayed on a browser launched by the maintenance company terminal 400. Alternatively, dedicated software may be installed in the maintenance company terminal 400, and the input screen for flows 1 to N and the delay status may be displayed on the display device of the maintenance company terminal 400.

In the first embodiment, the prediction unit 130 predicts the delay state of the work period using the first prediction model 191 as a prediction model. Here, the "work period" refers to the period required for multiple tasks (flows 1 to N) from the start of the maintenance service to the completion of the maintenance result report, that is, the period until the entire maintenance service is completed.

The prediction model (first prediction model 191) is a model that predicts a delay state using the difference between the reference period for each of multiple tasks (flows 1 to N) and the actual period for that reference period. The "reference period" (also called "reference number of days") specifies the number of days expected to be required for each task (flow). The "actual period" (also called "actual number of days") is the number of days actually required for each task (flow).

The performance period refers to the period from when a task is assigned to a worker (the start of the flow) to when the worker completes the task (the end of the flow). For example, in flow 1, the start of the flow is not necessarily when a maintenance worker begins inspection work on-site. The performance period also includes the period from when a task is assigned to a maintenance worker to when the maintenance worker actually goes to the site. The base period may be calculated from the average value of past performance periods, but the base period may also be adjusted taking into account the worker's work efficiency or monthly working hours.

Hereinafter, the reference period for flow 1 will be referred to as "reference number of days 1", and the actual period for flow 1 will be referred to as "actual number of days 1". The reference period for flow 2 will be referred to as "reference number of days 2", and the actual period for flow 2 will be referred to as "actual number of days 2". Similarly below, the reference period for flow N will be referred to as "reference number of days N", and the actual period for flow N will be referred to as "actual number of days N". For example, the prediction model can be expressed by the following formula.

Forecast model = -1 x [X - (base number of days 1 + base number of days 2 ... + base number of days N) + {(base number of days 1 - actual number of days 1) x m1 + (base number of days 2 - actual number of days 2) x m2 ... + (base number of days N - actual number of days N) x mN}] + Y
When using this prediction model, it is shown that the larger the prediction result output, the greater the delay that will occur until the maintenance work results are reported.

Here, X is the target number of days from the start of maintenance work until the maintenance result report is completed. m1, m2...mN are set to "1" if each flow has been performed, and "0" if each flow has not been performed. For flows that have not been performed, "0" is set in the actual number of days. For flows that have been performed, data is set in the actual number of days.

Y is set to 100 when no report data has been issued (this state is also referred to as the "unreported state"), as described below, and is set to "0" when report data has been issued (this state is also referred to as the "reported state"). Note that it is not necessary to use report data in the prediction model. In this case, Y is always 0.

The generation unit 132 generates a first prediction model 191 using the reference periods for each of the multiple tasks (flows 1 to N). Specifically, the generation unit 132 determines the reference numbers of days 1 to N, respectively, based on the past history information 194. The generation unit 132 then generates the first prediction model 191 by setting the determined reference numbers of days 1 to N.

The past history information 194 is the history information of the entire maintenance service for all buildings managed by the maintenance company. The generation unit 132 can obtain a set of performance days 1 to N from the past history information 194 for each building, each elevator in the building, or each model of elevator.

For example, the average value of the multiple acquired numbers of actual days 1 is determined as the reference number of days 1. Similarly, the average value of the multiple acquired numbers of actual days 2 is determined as the reference number of days 2, and the average value of the multiple acquired numbers of actual days N is determined as the reference number of days N.

When obtaining all the actual days 1 to N for building 10 and calculating the reference days 1 to N, it is possible to construct a prediction model according to the unique circumstances of building 10. It is possible to take into account circumstances such as the fact that work times will be longer if the building is large.

When all the actual days 1-N for elevator unit A installed in building 10 are obtained and the reference days 1-N are calculated, a prediction model can be constructed according to the circumstances specific to unit A. When all the actual days 1-N for elevator model X are obtained and the reference days 1-N are calculated, a prediction model can be constructed according to the circumstances specific to model X. In this way, a prediction model can be constructed according to the situation, such as for each elevator or each site.

The memory unit 214 of the monitoring device 200 stores the alert data 193 that is issued by the elevator when maintenance service is performed. As described above, when the elevator control device 100 detects a connection with the maintenance computer 30, it issues alert data indicating that maintenance service has started.

The monitoring device 200 receives the issued alert data via the network. The received alert data is stored in the memory unit 214. The prediction unit 130 acquires the alert data from the memory unit 214 when predicting a delay state.

The prediction unit 130 predicts the delay state so that the unalarmed state is evaluated as being more delayed than the alarmed state. Specifically, Y = 100 is set in the unalarmed state, and Y = 0 is set in the alarmed state. In this way, it is possible to reliably grasp that maintenance work has been performed from the alarm data, and to call attention to the implementation of maintenance work (promotion of the implementation of maintenance work).

The prediction model may also have weights W1 and W2 added, as follows: W1 is set based on the number of past trouble occurrences in the target case, or the elevator's own failure history (for example, the value of W1 is set large for cases where trouble is likely to occur). W2 is set based on the speed of customer response (for example, W2 is set small (negative) for cases where customer response is fast).

Prediction model = -1 x [X - (base number of days 1 + base number of days 2 ... + base number of days N) + {(base number of days 1 - actual number of days 1) x m1 + (base number of days 2 - actual number of days 2) x m2 ... + (base number of days N - actual number of days N) x mN}] + Y + W1 + W2
Fig. 4 is a flowchart of a process executed by the monitoring system 1 according to the first embodiment. The flowchart in Fig. 4 shows a process executed by the prediction unit 130 and the display unit 131. Hereinafter, a "step" may also be simply referred to as "S".

As shown in Figure 4, when the above process starts, in S1, the prediction unit 130 of the monitoring system 1 acquires the prediction model stored in the memory unit 214 and the actual period for each of the multiple tasks (flows 1 to N) relative to the reference period, and proceeds to S2. Specifically, the prediction unit 130 acquires the first prediction model 191 and the actual number of days 1 to N from the implementation history information 192.

In S2, the prediction unit 130 of the monitoring system 1 predicts the delay state based on the first prediction model 191 and the performance periods of each of the multiple tasks (flows 1 to N), and proceeds to S2. Specifically, the prediction unit 130 predicts the delay state based on the first prediction model 191 and the performance days 1 to N.

In S3, the display unit 131 of the monitoring system 1 displays the delay state predicted by the prediction unit 130, and the process ends. Specific examples are described below with reference to Figures 5 to 9. Figures 5 to 9 are diagrams for explaining the delay state of the work period in the first embodiment.

First, we will explain the calculation image of the delay state using Figures 5 and 6. For simplicity, we will assume that the flow consists of flows 1 to 3. Also, in this example, we will not use the notification data to predict the delay state (Y = 0 at all times).

First, using Figure 5, we will show an example where maintenance result reporting was completed without delay. Here, it is assumed that the target number of days X is set to 28 days. "Flow 1 base" indicates base number of days 1 for flow 1, where base number of days 1 = 7 days. "Flow 2 base" indicates base number of days 2 for flow 2, where base number of days 2 = 7 days. "Flow 3 base" indicates base number of days 3 for flow 3, where base number of days 3 = 7 days.

If all goes according to plan, the maintenance result report is scheduled to be completed in 21 days (= 7 + 7 + 7). This number of days is denoted as "estimated days to completion." In the diagram, "reserve" (also called "reserve days") = target days X - estimated days to completion. In this case, reserve = 7 days (= 28 - 21). In other words, in this case, the number of days delayed from target days X = -7 days. The number of days delayed is the result of the delay predicted by the prediction unit 130.

Now, let's say that flow 1 is implemented and the actual number of days for flow 1 (flow 1 actual) = 4, meaning that the work for flow 1 is completed 3 days (= 7 - 4) earlier than planned. This increases the number of reserve days by 3, making it 10 days (= 7 + 3) (number of days delayed = -10 days).

Furthermore, suppose that flow 2 is implemented and the actual number of days for flow 2 (flow 2 actual) = 4, meaning that the work for flow 2 is completed 3 days (= 7 - 4) earlier than planned. This increases the number of reserve days by 3, making it 13 days (= 10 + 3) (number of days delayed = -13 days).

Finally, flow 3 is implemented, and the actual number of days for flow 3 (flow 3 actual) = 4, meaning that the work for flow 3 is completed 3 days (= 7 - 4) earlier than planned. This increases the number of reserve days by 3, making it 16 days (= 16 + 3) (number of days delayed = -16 days).

The prediction unit 130 calculates the number of days of delay as the delay state. The delay state may include the actual period (actual days 1 to 3). The display unit 131 displays the number of days of delay. In addition, the display unit 131 may display the actual period, or may display a diagram such as that shown in Figure 5 that makes it easier to grasp the overall delay situation. In this way, the delay state (expected delay) before the flow starts, after flow 1 is performed, and after flow 2 is performed can be grasped.

Next, Figure 6 will be used to show an example where a delay ultimately occurred. As in Figure 5, target days X = 28 days, base days 1 = 7 days, base days 2 = 7 days, base days 3 = 7 days. Estimated days for completion = 21 days, and reserve (reserve days) = 7 days (number of days delayed = -7 days).

Now, let's say that flow 1 is carried out and the actual number of days for flow 1 (flow 1 actual) = 10, meaning that the work for flow 1 is delayed by 3 days (= 10 - 7) compared to the plan. This reduces the number of reserve days by 3 days, making it 4 days (= 7 - 3) (number of days delayed = -4 days).

Furthermore, suppose that flow 2 is implemented and the actual number of days for flow 2 (flow 1 actual) = 12, meaning that the work for flow 1 is delayed by 5 days (= 12 - 7) compared to the plan. As a result, the number of reserve days is reduced by 5 days to -1 day (= 4 - 5) (number of days delayed = 1 day). In this way, if the number of days delayed is > 0, it is expected that a delay will occur from the target number of days X. In other words, there is a possibility that the SLA contract will not be met.

Finally, let's say that flow 3 is implemented and the actual number of days for flow 3 (flow 3 actual) = 9, meaning that the work for flow 3 is delayed by 2 days (= 9 - 7) compared to the plan. This reduces the number of reserve days by 2 days, making it -3 days (= -1 - 2). In other words, the number of days delayed from the target number of days X = 3 days.

In this example, when Flow 2 is performed before the final work process, Flow 3, is performed, it is expected that there will be a delay of one day from the target number of days X. Therefore, in order to complete the work within the target number of days X, it is necessary to create a work plan that shortens the work of Flow 3 by one day, so that the actual number of days 3 = 6 days (= 7 - 1).

In this way, in this embodiment, it is possible to constantly grasp the delay status of work related to the entire maintenance service. This allows for warnings about delays at an intermediate stage, and makes it possible to prevent work delays.

Next, a specific example will be described with reference to Figures 7 to 9. Here, it is assumed that the flow is composed of flows 1 to N. In addition, in this example, it is assumed that notification data is used to predict the delay state.

Figure 7 shows a state in which none of the flows have been carried out. Each flow is set with information such as "Work content," "Person in charge," "Base number of days," and "Actual number of days." In this state, "Work content," "Person in charge," and "Base number of days" are set in advance, but "Actual number of days" is not set.

Flow 1 is elevator maintenance work performed by a maintenance worker. The "Work content" of Flow 1 is set to "Elevator maintenance," the "Person in charge" is set to "Maintenance worker," and the "Base number of days (Base number of days 1)" is set to "3 days." However, since all flows have not yet been carried out, the "Actual number of days (Actual number of days 1)" has not yet been set.

Flow 2 is an administrative task performed by an office worker. Specifically, the results of maintenance work performed by a maintenance worker are entered. "Maintenance result input" is set as the "Work content" of Flow 2, "Office worker" is set as the person in charge, and "1 day" is set as the "Base number of days (Base number of days 2)". "Actual number of days (Actual number of days 2)" is not set.

Flow 3 is the confirmation work performed by the confirmer. Specifically, it is the work of the maintenance staff and the superior of the office staff confirming and verifying the contents of the maintenance results that have been entered. "Confirmation work" is set as the "Work content" of Flow 3, "Confirmer (superior)" is set as the person in charge, and "3 days" is set as the "Base number of days (Base number of days 3)". "Actual number of days (Actual number of days 3)" is not set.

The final flow, Flow N, is the reporting work performed by the reporter. Specifically, it is the work of making the report viewable from the customer terminal 300. The "reporting work" is set as the "work content" of Flow N, the "reporter" is set as the person in charge, and "3 days" is set as the base number of days. The "actual number of days" is not set.

The above example is one example of a flow, but is not limited to this. As a reviewer, multiple people may verify using multiple flows, or the work process may be further subdivided. The reporting work may be divided into a flow for uploading the report to the server, approving the upload, and publishing the uploaded report (at this point it becomes viewable from the customer terminal 300).

In the state shown in Figure 5, base number of days 1 = 3, base number of days 2 = 1 ... base number of days N = 3, m1 = 0, m2 = 0 ..., mN = 0, Y = 100 (no report issued yet). Therefore, the prediction model = -1 x [28 - (3 + 1 ... + 3) + {(3 - number of days of performance 1) x 0 + (1 - number of days of performance 2) x 0 ... + (3 - number of days of performance N) x 0}] + 100.

Here, the sum of the reference days 1 to N is assumed to be 28 (i.e., equal to X). In this case, the prediction model (first prediction model 191) = 100. In other words, the estimation result obtained by the first prediction model 191 is a delay = 100.

As described above, in the unalarmed state, Y = 100 is set, and in the alarmed state, Y = 0 is set. In this way, by setting a large value in the unalarmed state, it is possible to promote the implementation of maintenance work (flow 1).

Now, let us assume that flow 1 (maintenance work) was carried out and its actual number of days was 2 days. As a result, as shown in Figure 8, "Actual number of days (Actual number of days 1)" for flow 1 is set to "2 days." Furthermore, as a result of the maintenance work being carried out, alert data is issued. As a result, the delay evaluation due to the alert status changes from Y = -100 to Y = 0.

Prediction model = -1 x [0 + {(3 - actual number of days 1) x 1 + (1 - actual number of days 2) x 0 ... + (3 - actual number of days N) x 0}] + 0 = -(3 - actual number of days 1). Actual number of days 1 = 2. Therefore, the estimated result from the first prediction model 191 is delay = -(3 - 2) = -1.

Next, flow 2 (maintenance result input) is performed, and the number of days of performance is three days. As a result, "three days" is set as the "number of days of performance (number of days of performance 2)" of flow 2, as shown in Figure 7.

The prediction model = -(3 - actual number of days 1) - (1 - actual number of days 2). Actual number of days 1 = 2, actual number of days 2 = 3. Therefore, the estimation result obtained by the first prediction model 191 is delay = -(3 - 2) - (1 - 3) = 1. Although not shown, when flows 3 to N are implemented, the estimation result is updated in the same manner as above.

The display unit 131 displays the delay (number of days of delay). Additionally, the display unit 131 may display the actual period, or may display a diagram that makes it easier to grasp the overall delay situation, such as those shown in Figures 7 to 9. In this way, the delay state (expected delay) before the start of the flow and after each flow is performed can be grasped.

As explained above, by displaying the delay status of work for the entire maintenance service, it is possible to call attention to delays in work for the entire maintenance service and prevent work delays. In addition, by displaying the base number of days, actual number of days, and number of delay days for each flow as the delay status, it is possible to visualize and monitor the flow from when maintenance is performed to when the maintenance results are reported to the customer. With the above configuration, in a maintenance service that specifies an SLA, it is possible to improve the compliance rate with the SLA for the entire maintenance service and increase customer satisfaction.

[Second embodiment]
The second embodiment will be described below. In the description of the second embodiment, differences from the first embodiment will be described, and descriptions of common parts will be omitted.

Figure 10 is a diagram showing an example of a functional block diagram of a monitoring system 1 according to the second embodiment. In the first embodiment, the monitoring system 1 includes a prediction unit 130, a display unit 131, and a generation unit 132, and the memory unit 214 is configured to store a first prediction model 191, implementation history information 192, alert data 193, and past history information 194.

In contrast, in the second embodiment, as shown in FIG. 10, the monitoring system 1 further includes a first update unit 134 and a second update unit 135. The memory unit 214 further stores a second prediction model 197, inspection information 195, and company-wide inspection information 196.

In the second embodiment, the prediction model further includes a second prediction model 197. Here, the first prediction model 191 is a model that predicts a delay state when no trouble occurs during maintenance service. The second prediction model 197 is a model that predicts a delay state when a trouble occurs during maintenance service. When a trouble occurs, trouble information is recorded in the inspection information 195.

In addition, the first prediction model 191 in the first embodiment may be a model that predicts a delay state when no trouble occurs during maintenance service, or a model that predicts a delay state both when no trouble occurs during the work period and when a trouble occurs.

The process flow will be explained below based on the flowchart. FIG. 11 is a flowchart of the process executed by the monitoring system 1 according to the second embodiment. As shown in FIG. 11, when the process executed by the elevator control device 100 starts, the process of S21 is executed.

In S21, the prediction unit 130 of the monitoring system 1 determines whether or not a problem has occurred during the maintenance service. Whether or not a problem has occurred is determined based on the problem information recorded in the inspection information 195. The inspection information 195 stores the results of regular inspections, including the problem information.

If the prediction unit 130 determines that a problem has occurred during the maintenance service (YES in S21), the process proceeds to S23. If the prediction unit 130 does not determine that a problem has occurred during the maintenance service (NO in S21), the process proceeds to S22.

In S22, the prediction unit 130 selects the first prediction model 191 as the prediction model and proceeds to S23. In S23, the first update unit 134 of the monitoring system 1 determines whether an update of the first prediction model 191 has been requested.

The update of the first prediction model 191 may be requested by a user through an input device, or may be requested every predetermined period (for example, four months). Alternatively, an update request may be made constantly. In this case, the first prediction model 191 will always be updated.

If the first update unit 134 determines that an update of the first prediction model 191 has been requested (YES in S23), the process proceeds to S24. If the first update unit 134 does not determine that an update of the first prediction model 191 has been requested (NO in S23), the process proceeds to S28.

In S24, the first update unit 134 updates the first prediction model 191 based on the results of past maintenance services, and proceeds to S28. The processing content of the first update unit 134 is the same as the processing content of the generation unit 132, and determines each of the reference days 1 to N based on the past history information 194. The first update unit 134 then updates the first prediction model 191 by setting the determined reference days 1 to N. In this way, the prediction model can be updated according to the implementation status of past maintenance services.

In S25, the prediction unit 130 selects the second prediction model 197 as the prediction model and proceeds to S26. In S26, the second update unit 135 of the monitoring system 1 determines whether an update of the second prediction model 197 has been requested.

The second prediction model 197 may be configured so that an update is requested by a user via an input device, or may be requested every predetermined period (for example, four months). Alternatively, an update request may be made constantly. In this case, the second prediction model 197 will always be updated when a problem occurs.

If the second update unit 135 determines that an update of the second prediction model 197 has been requested (YES in S26), the process proceeds to S27. If the second update unit 135 does not determine that an update of the second prediction model 197 has been requested (NO in S26), the process proceeds to S28.

In S27, the second update unit updates the second prediction model 197 based on the results of past maintenance services in the event that a problem occurs during maintenance services, and proceeds to S28.

Specifically, the second update unit updates the second prediction model 197 based on company-wide inspection information 196. Company-wide inspection information 196 is information that accumulates the results of past maintenance services when trouble occurs for elevators in each building. In this way, it is possible to update the prediction model according to the implementation status of past maintenance services when a trouble occurs.

If a problem occurs, not only will the maintenance work for flow 1 take time, but it is highly likely that the work in the subsequent processes will also take time. As mentioned above, by switching the prediction model, it is possible to grasp the overall extent of delays that will occur, making it easier to deal with delays when a problem occurs.

In S28, the prediction unit 130 predicts the delay state using the selected prediction model, and proceeds to S29. Specifically, if the first prediction model 191 is selected as the prediction model, the delay state is predicted based on the first prediction model 191 and the number of actual days 1 to N. On the other hand, if the second prediction model 197 is selected as the prediction model, the delay state is predicted based on the second prediction model 197 and the number of actual days 1 to N. In S29, the display unit 131 displays the delay state predicted by the prediction unit 130, and the process ends.

As described above, if no trouble occurs during the maintenance service, the prediction unit 130 predicts the delay state using the first prediction model 191. If a trouble occurs during the maintenance service, the prediction unit 130 predicts the delay state using the second prediction model 197. In this way, it is possible to grasp the delay state appropriate to the trouble occurrence situation, and it is possible to suppress delays in work under the trouble occurrence situation. By configuring as described above, in a maintenance service that specifies an SLA, it is possible to improve the compliance rate with the SLA covering the entire maintenance service, and to further increase customer satisfaction.

[Main components and effects]
The main configurations and effects of the above-described embodiment will be described below.

(1) The monitoring system 1 monitors the provision of elevator maintenance services. The monitoring system 1 includes a memory unit 214, a prediction unit 130, and a display unit 131. The memory unit 214 stores a first prediction model 191 (second prediction model 197) that predicts a delay state of a work period required for multiple tasks from the start of the maintenance service to the completion of a report of the implementation result of the maintenance service to the customer, and an actual period for each of the multiple tasks with respect to a reference period. The prediction unit 130 predicts the delay state based on the first prediction model 191 and the actual period for each of the multiple tasks. The display unit 131 displays the delay state predicted by the prediction unit 130. The first prediction model 191 is a model that predicts the delay state using the difference between the reference period for each of the multiple tasks and the actual period with respect to the reference period. In this way, by displaying the delay state of the tasks related to the entire maintenance service, attention can be drawn to delays in the tasks related to the entire maintenance service, and delays in the tasks can be suppressed. This makes it possible to improve the compliance rate with the SLA covering the entire maintenance service in a maintenance service that specifies an SLA, thereby increasing customer satisfaction.

(2) The monitoring system 1 further includes a generation unit 132. The generation unit 132 generates a first prediction model 191 using a reference period for each of a plurality of tasks. This makes it possible to build a prediction model according to the situation, such as for each elevator or each work site.

(3) The memory unit 214 further stores the notification data issued by the elevator when the maintenance service is performed. The prediction unit 130 predicts the delay state so that when the notification data is not issued, the delay is evaluated to be greater than when the notification data is issued. In this manner, it is possible to reliably grasp that maintenance work has been performed from the notification data, and to call attention to the performance of the maintenance work. As a result, in a maintenance service that specifies an SLA, it is possible to improve the compliance rate with the SLA covering the entire maintenance service, thereby increasing customer satisfaction.

(4) The memory unit 214 further stores the results of past maintenance services. The monitoring system 1 further includes a first update unit that updates the first prediction model 191 based on the results of past maintenance services. In this manner, the prediction model can be updated according to the implementation status of past maintenance services. This makes it possible to improve the compliance rate with the SLA covering the entire maintenance service in a maintenance service that specifies an SLA, thereby increasing customer satisfaction.

(5) The prediction model includes a first prediction model 191 that predicts a delay state when no trouble occurs during the maintenance service, and a second prediction model 197 that predicts a delay state when a trouble occurs during the maintenance service. When no trouble occurs during the maintenance service, the prediction unit 130 predicts a delay state using the first prediction model 191. When a trouble occurs during the maintenance service, the prediction unit 130 predicts a delay state using the second prediction model 197. In this way, it is possible to grasp a delay state that matches the trouble occurrence situation, and it is possible to suppress delays in work under the trouble occurrence situation. As a result, in a maintenance service that specifies an SLA, it is possible to improve the compliance rate with the SLA for the entire maintenance service, thereby increasing customer satisfaction.

(6) The monitoring system 1 further includes a second update unit. The second update unit updates the second prediction model 197 based on the results of past maintenance services when a problem occurs during maintenance services. In this way, the prediction model can be updated according to the implementation status of past maintenance services when a problem occurs. This makes it possible to improve the compliance rate with the SLA covering the entire maintenance service in a maintenance service that specifies an SLA, thereby increasing customer satisfaction.

(7) The monitoring method is a method for monitoring the provision of an elevator maintenance service. The monitoring method includes a step of storing a first prediction model 191 (second prediction model 197) that predicts a delay state of a work period required for a plurality of tasks from the start of the maintenance service to the completion of a report of the implementation result of the maintenance service to the customer, a step of storing an actual period for each of the plurality of tasks with respect to a reference period, a step of predicting a delay state based on the first prediction model 191 and each of the actual periods of the plurality of tasks, and a step of displaying the delay state predicted by the predicting step. The first prediction model 191 is a model that predicts a delay state using a difference between a reference period for each of the plurality of tasks and an actual period for the reference period. In this way, by displaying the delay state of the tasks related to the entire maintenance service, it is possible to call attention to delays in the tasks related to the entire maintenance service, and to suppress delays in the tasks. As a result, in a maintenance service in which an SLA is specified, it is possible to improve the compliance rate with the SLA for the entire maintenance service and increase customer satisfaction.

The embodiments disclosed herein should be considered to be illustrative and not restrictive in all respects. The scope of the present disclosure is indicated by the claims, not the above description, and is intended to include all modifications within the meaning and scope of the claims.

1, 1a monitoring system, 10 building, 20 information center (maintenance company), 21 maintenance staff, 30 maintenance computer, 40 branch office (maintenance company), 100 elevator control device, 111, 211 CPU, 112, 212 ROM, 113, 213 RAM, 214 storage unit, 115, 215 communication interface, 116, 216 I/O interface, 130 prediction unit, 131 display unit, 132 generation unit, 134 first update unit, 135 second update unit, 190 prediction model, 191 first prediction model, 192 implementation history information, 193 report data, 194 past history information, 195 inspection information, 196 company-wide inspection information, 197 second prediction model, 200 monitoring device, 300 customer terminal, 341 cage, 351 Elevator shaft, 331-334 landing, 400 maintenance company terminal.

Claims (7)

A monitoring system for monitoring provision of elevator maintenance services, comprising:
a storage unit that stores a prediction model that predicts a delay state of a work period required for a plurality of tasks from the start of the maintenance service to the completion of a report of the implementation result of the maintenance service to a customer, and stores an actual period for each of the plurality of tasks relative to a reference period;
a prediction unit that predicts the delay state based on the prediction model and an actual performance period of each of the plurality of operations;
a display unit that displays the delay state predicted by the prediction unit,
The customer has a contract for the maintenance service with a maintenance company for the elevator,
the plurality of tasks include the maintenance service, an administrative task of inputting the results of the maintenance service, and a reporting task of making a report of the maintenance service available for viewing by the customer;
A monitoring system, wherein the prediction model is a model that predicts the delay state using a difference between a reference period for each of the plurality of tasks and an actual period for that reference period. The monitoring system according to claim 1, further comprising a generating unit that generates the prediction model using a reference period for each of the plurality of tasks. The storage unit further stores alarm data that is issued by the elevator when the maintenance service is performed,
The monitoring system according to claim 1 or claim 2, wherein the prediction unit predicts the delay state so that when the alert data is not issued, the alert data is evaluated as being delayed more than when the alert data is issued. The storage unit further stores results of the past maintenance services,
The monitoring system according to any one of claims 1 to 3, further comprising a first update unit that updates the prediction model based on results of the past maintenance services. the prediction model includes a first prediction model that predicts the delay state when no trouble occurs during the maintenance service, and a second prediction model that predicts the delay state when a trouble occurs during the maintenance service,
The prediction unit is
If no trouble occurs during the maintenance service, predicting the delay state using the first prediction model;
5. The monitoring system according to claim 1, further comprising: a second prediction model that predicts the delay state when a trouble occurs during the maintenance service. The monitoring system according to claim 5, further comprising a second update unit that updates the second prediction model based on the results of the past maintenance service in the event that a problem occurs during the maintenance service. 1. A computer-implemented monitoring method for monitoring provision of elevator maintenance services, comprising:
storing a prediction model for predicting a delay state of a work period required for a plurality of works from the start of the maintenance service to the completion of a report of the implementation result of the maintenance service to a customer;
storing an actual performance period with respect to a reference period for each of the plurality of tasks;
predicting the delay state based on the prediction model and a performance period of each of the plurality of operations;
and displaying the delay state predicted in the predicting step,
The customer has a contract for the maintenance service with a maintenance company for the elevator,
the plurality of tasks include the maintenance service, an administrative task of inputting the results of the maintenance service, and a reporting task of making a report of the maintenance service available for viewing by the customer;
A monitoring method, wherein the prediction model is a model that predicts the delay state using a difference between a reference period for each of the plurality of tasks and an actual period for that reference period.

Images