Construction management system and construction management method

Abstract

A work management system (1) comprises a control unit (111) that generates a process schedule for construction, and a display unit (221) that displays the process schedule. The control unit (111) determines a load state indicating a task load of a plurality of managers on the basis of information including schedule information for the plurality of managers. The display unit (221) displays schedule information and the load state in association with the processes that each of the plurality of managers is responsible for.

Description

Construction management system and construction management method

[0001] The present disclosure relates to a construction management system and a construction management method.

[0002] At construction sites such as those for buildings, multiple workers (persons in charge) often work on multiple tasks in parallel. The construction manager (site agent) who manages the construction needs to understand the workload of each worker and the progress of multiple tasks being carried out simultaneously. Furthermore, the construction manager needs to take measures to ensure that all work is completed safely by the deadline while maintaining the quality of the work.

[0003] The construction manager grasps the workload of the workers by holding morning meetings, interviewing the work supervisor, and checking the daily work reports, and also grasps the progress of the work by patrolling the work site to make visual inspections and checking on-site photographs, etc. Based on the situation grasped in this way, the construction manager takes measures such as increasing or replacing the number of workers and reviewing the work process.

[0004] Japanese Patent Application Laid-Open No. 2020-135098

[0005] A certain amount of on-site experience is required for construction managers to be able to make appropriate decisions based on the situation they have grasped, such as whether it is necessary to increase or replace workers, or to review work processes.In the current situation where the construction industry's labor force is declining, it is necessary to train construction managers who can make such appropriate decisions, but there is also a demand for construction management systems that allow even inexperienced construction managers to efficiently manage construction.

[0006] The present disclosure has been made to solve such problems, and the purpose of the present disclosure is to provide a construction management system and a construction management method that allow even inexperienced construction managers to efficiently manage construction.

[0007] The construction management system disclosed herein is a system for managing construction work consisting of multiple processes that are handled by multiple workers. The construction management system includes a control unit that generates a construction schedule and a display unit that displays the schedule. The control unit determines a load status indicating the workload of multiple workers based on information including the schedule information of the multiple workers. The display unit displays the schedule information and the load status in association with the process that each of the multiple workers is responsible for.

[0008] The construction management method disclosed herein is a method for managing construction work consisting of multiple processes that are handled by multiple workers. The construction management method includes a step of generating a construction schedule and a step of displaying the schedule. The generating step includes a step of determining a load status indicating the workload of multiple workers based on information including schedule information for the multiple workers. The displaying step includes a step of displaying the schedule information and the load status in association with the process that each of the multiple workers is responsible for.

[0009] According to the present disclosure, even construction managers with little experience can efficiently manage construction.

[0010] FIG. 1 is a diagram showing the hardware configuration of a construction management system according to a first embodiment. FIG. 2 is a diagram showing a functional block diagram in the construction management system. FIG. 3 is a flowchart of main processing. FIG. 4 is a flowchart of display information generation processing. FIG. 5 is a diagram showing an example of a schedule. FIG. 6 is a diagram showing an example of a schedule. FIG. 7 is a diagram showing a functional block diagram in a construction management system according to a second embodiment. FIG. 8 is a flowchart of main processing. FIG. 9 is a diagram showing an example of a schedule. FIG. 10 is a diagram for explaining a learning dataset. FIG. 11 is a diagram for explaining a learning dataset. FIG. 12 is a flowchart of route display processing according to a modified example. FIG. 13 is a diagram showing an example of a schedule according to a modified example.

[0011] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings. While several embodiments will be described below, it was originally intended that the configurations described in each embodiment be combined as appropriate. Note that identical or corresponding parts in the drawings will be designated by the same reference numerals, and their description will not be repeated.

[0012] 1 is a diagram showing the hardware configuration of a construction management system 1 according to a first embodiment. The construction management system 1 is a system for managing construction work consisting of multiple processes. These multiple processes are handled by multiple people in charge (also referred to as "workers").

[0013] For example, the example in Figure 5, which will be described later, shows a schedule showing multiple processes in the construction of new air conditioning equipment in Building X. In this example, companies A to F (each person in charge at companies A to F) are in charge of the work in each process.

[0014] Here, from April 1st to 2nd, a person in charge (worker) from Company A is in charge of the work (process) of bringing in the equipment on the second floor. Then, finally, from April 23rd to 24th, a person in charge from Company F is in charge of the trial operation on the second and third floors, and once these tasks are completed, the construction of the main work will be completed.

[0015] Here, the "person in charge" may refer to each person in charge of the contractors (Company A to Company F) who undertake each process as described above, or may refer to the contractors (Company A to Company F) themselves who undertake each process. There may be one or more people in charge (workers) in charge of one process.

[0016] Returning to Figure 1, the construction management system 1 comprises a server 100 and a plurality of terminals 200. The server 100 performs processes such as generating a schedule. The plurality of terminals 200 comprises a terminal used by a construction manager (also simply referred to as "manager") who manages the construction, and a plurality of terminals used by a plurality of people in charge of a plurality of construction processes.

[0017] The construction manager manages the site so that the construction is completed by the deadline while understanding the workload of each person in charge of each process of the construction work and the progress of each task. In the example of Figure 1, a terminal 200 used by the construction manager, a terminal 200 used by worker K1, and a terminal 200 used by worker K2 are shown.

[0018] The construction management system 1 is configured to be connectable to a schedule management system 400. The schedule management system 400 is a system that manages the schedules of multiple people in charge.

[0019] Each person in charge accesses the schedule management system 400 from their own terminal 200 and manages their own schedule using schedule management software provided by the schedule management system 400. The schedule management software may be used by launching software that has been installed in the terminal 200 in advance, or may be used via a browser that is launched on the terminal 200.

[0020] Each person in charge specifies a date and time and uses schedule management software to register their own schedule in schedule management system 400. Users who are permitted to view the schedule (for example, users within the company) can access schedule management system 400 and view the registered schedule.

[0021] In this embodiment, the server 100 is configured to be able to access the schedule management system 400 and acquire the registered schedules. The schedule information includes schedules that are not related to the construction work managed by the construction manager.

[0022] 5, it is assumed that the companies (Company A to Company F) that are carrying out construction work to install new air conditioning equipment in Building X manage their schedules using a schedule management system 400. The server 100 can access the schedule management system 400 and obtain the schedules registered by the workers (persons in charge) of Companies A to F.

[0023] This schedule includes schedules unrelated to the construction of new air conditioning equipment in Building X. For example, it may be construction schedule information for Building Y, which is different from Building X, or a schedule for an internal meeting, or any other schedule.

[0024] Furthermore, companies A to F are not limited to using the same schedule management system (schedule management system 400), and each company may use a different schedule management system. In this case, server 100 is assumed to be able to access these multiple schedule management systems and acquire schedule information.

[0025] 1, the server 100 includes a control unit 111, a storage unit 112, and a communication unit 113. These are connected to each other via a bus so that they can communicate with each other.

[0026] The control unit 111 is, for example, a central processing unit (CPU). The storage unit 112 includes a read-only memory (ROM), a random access memory (RAM), and a non-volatile storage device (for example, a hard disk drive (HDD) or a solid state drive (SSD)). The storage unit 112 stores a training dataset 160, a trained model 163, a training dataset 170, and a trained model 173, which will be described later.

[0027] The control unit 111 loads programs stored in the ROM into the RAM and executes them to realize the various functions of the server 100. The control unit 111 performs processes such as generating a construction schedule. The ROM stores programs that describe the processing procedures of the server 100. The RAM serves as a working area when the control unit 111 executes the programs, and temporarily stores programs, data used when executing the programs, and the like. The server 100 can be connected to the terminal 200 and the schedule management system 400 wirelessly or via a wired connection via the communication unit 113.

[0028] The terminal 200 may be, for example, a mobile terminal such as a smartphone or a tablet, or a personal computer such as a notebook computer or a desktop computer.

[0029] The terminal 200 includes a control unit (CPU) 211, a storage unit 212, a communication unit 213, an input unit 220, and a display unit 221. These are connected to each other via a bus so that they can communicate with each other. Similarly, the storage unit 212 may be configured to include a ROM, a RAM, and a non-volatile storage device such as an HDD or SSD.

[0030] The control unit 211 loads programs stored in the ROM into the RAM and executes them to realize various functions of the terminal 200. The ROM stores programs that describe the processing procedures of the terminal 200. The terminal 200 can be connected to the server 100 and the schedule management system 400 via the communication unit 213.

[0031] The input unit 220 accepts input from a user. The input unit 220 is, for example, a touch panel, but may also be a keyboard or a mouse. The display unit 221 displays various information. The display unit 221 is, for example, a liquid crystal display or a display. The display unit 221 can display the process chart generated by the control unit 111. The process chart may be configured to be viewable only on the terminal 200 used by the manager, or may be configured to be viewable on the terminals 200 used by each worker.

[0032] The construction management system 1 may be a device that integrates the server 100 and the terminal 200 used by the manager. In this case, the device is configured so that the display unit (display) of the device displays the schedule created by the control unit (CPU) of the device. The construction management system 1 may also be configured to include the schedule management system 400, or may be configured only with the server 100 and the terminal 200 used by the manager.

[0033] 2 is a functional block diagram of the construction management system 1. The control unit 111 of the server 100 is capable of executing the processes executed by the second estimating unit 121 and the generating unit 123.

[0034] The second estimating unit 121 acquires schedule information of a plurality of people in charge of construction work from the schedule management system 400. The second estimating unit 121 acquires second input information, which will be described later, from the storage unit 112.

[0035] The second estimation unit 121 determines a load state indicating the workload of multiple personnel based on the schedule information of the multiple personnel and the second input information. The load state includes a high load state. A "high load state" is a state in which a delay in work is predicted in the target process. The second input information is information necessary to determine the load state. The second input information and the process of determining the load state will be described in detail using FIGS. 10 and 12.

[0036] The generation unit 123 acquires process information (described later) from the storage unit 112. The generation unit 123 generates display information including a process chart based on the process information, the load state, etc. The display unit 221 of the terminal 200 displays the display information including the generated process chart.

[0037] The main processing will be specifically described below with reference to a flowchart. Fig. 3 is a flowchart of the main processing. The processing shown in this flowchart may be started based on a display request from the terminal 200, for example.

[0038] When the main processing starts, in S101, the server 100 acquires schedule information from the schedule management system 400. In S102, the server 100 acquires second input information from the storage unit 112. In S103, the server 100 inputs the schedule information and the second input information into the trained model 163 ( FIG. 10 ) and acquires the load state output from the trained model 163.

[0039] In S104, the server 100 acquires process information from the storage unit 112. The process information is various information set regarding multiple processes in the target construction work. For example, in the example of FIG. 5 , the information is information necessary to generate a process schedule, such as the construction work name (construction of new air conditioning equipment in Building X), the item name of each process (e.g., equipment delivery to the second floor), the company name (Company A), and the planned construction date (April 1st to 2nd, 2022).

[0040] In S105, the server 100 executes the display information generation process (FIG. 4) to generate display information (e.g., a schedule). In S106, the server 100 outputs the display information (e.g., a schedule) to the display unit 221, and ends the main process.

[0041] Specifically, the main process is executed in response to a display request from the terminal 200, and the display information generated as a result is acquired by the terminal 200 via the communication unit 213 and displayed on the display unit 221. Examples of the process chart displayed on the display unit 221 are shown in Figures 5 and 6. Below, the process charts shown in Figures 5 and 6 will be described while explaining the flowchart of the display information generation process using Figure 4.

[0042] 4 is a flowchart of the display information generation process. When the display information generation process starts, the server 100 acquires schedule information, load information, process information, etc. in S201. A process chart is generated based on this information.

[0043] 5 is a diagram showing an example of a schedule. As shown in FIG. 5, the name of the work is displayed at the top of a schedule 91 (hereinafter also simply referred to as a "schedule"). In this example, it is shown that the work is the installation of a new air conditioning system in Building X. In this embodiment, the schedule is a bar chart schedule that displays multiple bar charts corresponding to multiple processes.

[0044] In this example, a bar chart shows a schedule for the period from April 1 to April 24, 2022. A vertical line indicating that today is April 9, 2022 is displayed on the schedule.

[0045] The schedule shows the item name, company name, progress rate, and bar chart for each process. The item name column shows the task name (type of task) for each process. The company name column shows the name of the company in charge of the task. The progress rate column shows the progress rate of the task as a percentage.

[0046] The construction of new air conditioning equipment for Building X shows multiple processes, including equipment delivery to the second floor by Company A, equipment installation on the second floor by Company B, piping work on the second floor by Company C, wiring work on the second floor by Company D, insulation work on the second floor by Company E, and trial operation on the second floor by Company F.

[0047] Furthermore, for the construction of new air conditioning equipment in Building X, the following multiple processes are displayed: equipment delivery work on the third floor, which will be handled by Company A; equipment installation work on the third floor, which will be handled by Company B; piping work on the third floor, which will be handled by Company C; wiring work on the third floor, which will be handled by Company D; insulation work on the third floor, which will be handled by Company E; and trial operation on the third floor, which will be handled by Company F.

[0048] According to this schedule, equipment delivery to the second floor (Company A), equipment installation on the second floor (Company B), and equipment delivery to the third floor (Company A) are scheduled for April 1st and April 2nd. Next, piping work on the second floor (Company C) is scheduled for April 3rd to April 8th (in this embodiment, work is not performed on Saturdays and Sundays), and equipment installation on the third floor (Company B) is scheduled for April 3rd to April 4th. Today is April 9th, and all of the above processes have been completed as planned, so the bar chart is displayed in "blue" to indicate completed work. Furthermore, the progress rate for these tasks is displayed as "100%."

[0049] 4 , in S202, the server 100 extracts a first process and a second process from the process information. In this embodiment, the multiple processes include a first process and a second process that follows the first process. The second process is a process that will be affected by a delay in the work of the first process. This relationship is set in advance as process information.

[0050] In S203, the server 100 generates an image connecting a bar chart corresponding to the first process with a bar chart corresponding to the second process. As a result, when the process schedule is displayed on the display unit 221, an image connecting the first bar chart corresponding to the first process with the second bar chart corresponding to the second process is displayed. In this way, when a delay occurs in a certain process (the first process), it is possible to intuitively grasp the process (the second process) that is affected by the delay.

[0051] In the example of Figure 5, Company B is responsible for the installation of equipment on the second floor from April 1st to April 2nd, and the installation of equipment on the third floor from April 3rd to April 4th, and the installation of equipment on the third floor needs to be completed the day after the installation of equipment on the second floor is completed.

[0052] If the installation of the equipment on the second floor (first process) is delayed, the installation of the equipment on the third floor (second process) cannot be started, which may result in a delay to the installation of the equipment on the third floor. In this way, if a delay in one process (first process) could affect the next process (second process), an image connecting the two processes with a line (here, an image of an arrow pointing from the first process to the second process) is displayed as a "related task."

[0053] Furthermore, Company C is scheduled to carry out plumbing work on the third floor from April 9 to April 12 after the second floor plumbing work is completed from April 3 to April 8. In this case, if the second floor plumbing work is delayed, there is a possibility that the third floor plumbing work will not be able to begin. In this case, these two processes are also shown to be "related work."

[0054] On the other hand, even if the installation of equipment on the second floor (Company B) is delayed, the piping work on the second floor (Company C) can begin as scheduled on April 3. For this reason, these two processes are not set as "related work."

[0055] Next, wiring work on the second floor (Company D) is scheduled for April 9th ​​to April 10th, insulation work on the second floor (Company E) for April 11th to April 16th, and insulation work on the third floor (Company E) for April 17th to April 22nd. For this project, insulation work on the second floor cannot begin until the wiring work on the second floor is completed. For this reason, these processes are also shown to be "related work." Furthermore, because Company E needs to begin insulation work on the third floor the day after the insulation work on the second floor is completed, these processes are also shown to be "related work."

[0056] Furthermore, Company D will carry out wiring work on the second floor from April 9th ​​to April 10th, followed by wiring work on the third floor from April 15th to April 16th. Although there is a two-day gap between these two work processes, in this example, these work processes are also shown as "related work" because they are judged to have the potential to affect the next work process. Finally, Company F is scheduled to carry out a trial run on the second and third floors from April 23rd to April 24th.

[0057] 4 , in S204, the server 100 sets the bar chart image corresponding to the process in a high-load state to red. At this time, the entire bar chart image may be displayed in red, or only the period (days) in the high-load state may be displayed in red. In S205, when the load state of the first process is high, the server 100 sets the bar chart corresponding to the second process to pink.

[0058] When the load state of the first process is high, the display unit 221 displays the first bar chart in a first manner (red) that is different from when the load state of the first process is not high. This allows the construction manager to intuitively understand the load state of the person in charge.

[0059] When the load state of the first process is high, the display unit 221 displays the first bar chart in a first mode (red) and the second bar chart in a second mode (pink). This allows the construction manager to intuitively grasp which processes are likely to be delayed due to a high load situation and which processes may be affected by this high load situation and thus may be delayed. This allows the manager to quickly review the schedules of these processes and consider load adjustments, etc.

[0060] In the example of Figure 5, it is assumed that the load state for the wiring work (first process) on the second floor on April 10th is determined to be a high load state. In this case, the bar chart for the wiring work (first process) on the second floor on April 10th is displayed in red. Note that the entire bar chart for this process may also be displayed in red. Then, the bar chart for the insulation work (second process) on the second floor and the bar chart for the wiring work (second process) on the third floor, which are set as "work related to" this process, are each displayed in pink.

[0061] Also, assume that the load state for the insulation work on the second floor on April 15th is determined to be high. In this case, the bar chart for the insulation work on the second floor on April 15th is displayed in red. Note that the entire bar chart for the insulation work on the second floor may be displayed in red. Then, the bar chart for the insulation work on the third floor, which is set as a "related task" to this process, is displayed in pink.

[0062] As a general rule, processes that are in progress or scheduled to be carried out today (April 9th) are displayed as "scheduled work" in light blue on the bar chart. The wiring work on the second floor is currently underway (50% progress), so it is displayed in light blue. However, if red or pink is displayed as above, these colors take priority. Furthermore, the piping work on the third floor is also currently underway (25% progress), so it is displayed in light blue. Processes scheduled for April 10th and after have not yet started, so this is indicated in the progress rate column. The test runs on the second and third floors scheduled for April 23rd and 24th are displayed in light blue on the bar chart.

[0063] 4, in S206, the server 100 associates each bar chart with the schedule information of the person in charge. In S207, the server 100 generates display information (such as a schedule), and ends the display information generation process. In the example of FIG. 5, a schedule 91 is displayed on the display unit 221 as display information.

[0064] Next, an example of displaying schedule information (see S206) of a person in charge associated with each bar chart will be described with reference to Fig. 6. The schedule information can be switched between display and non-display depending on the user's preference, and Fig. 5 shows an example in which the schedule information is set to non-display, while Fig. 6 shows an example in which the schedule information is set to display.

[0065] 6 is a diagram showing an example of a schedule. As described above, Company D will carry out wiring work on the second floor from April 9th ​​to April 10th. For example, the schedule for April 10th for the person in charge of wiring work on the second floor at Company D (in this example, the person will be working alone) includes the following registered times: "8:00-17:00 Building X: wiring work," "18:00-20:00 Building Y: estimate creation," and "20:00-22:00 Building Z: drawing creation."

[0066] The display unit 221 displays schedule information in association with a bar chart corresponding to the process for which each of multiple workers is in charge. In this example, the schedule information is displayed in association with the bar chart corresponding to the wiring work on the second floor. At this time, the schedule information is displayed in association with the April 10th area of ​​the bar chart so that it is clear that it is the schedule for April 10th. At this time, the name of the person in charge may also be displayed.

[0067] This allows the construction manager to grasp the load status of each worker at a glance. After the worker from Company D performs wiring work on the second floor of this building (Building X) from 8:00 to 17:00 on April 10th, he returns to the office and then needs to work on creating estimates and drawings for other buildings (Building Y and Building Z) from 18:00 to 22:00.

[0068] In this way, in addition to the work in Building X, the person in charge at Company D also needs to work until 10 pm on April 10. If the work in Building X is delayed, work in Buildings Y and Z is scheduled to start at 6 pm, so the end time of work in Building X cannot be extended, and there is a possibility that the work will be carried over to the next day (the work will be delayed).

[0069] In this case, the load state of the wiring work on the second floor is determined to be high on April 10th, and the bar chart is displayed in red. Furthermore, as explained in FIG. 5, the bar chart of the process that may be affected is displayed in pink. Note that although not shown in the example of FIG. 6, other processes are also displayed with associated schedule information. Details of the determination of whether a high load state will occur will be described later using FIGS. 10 to 12.

[0070] In this way, the display unit 221 displays the load status in association with the process for which each of the multiple personnel is in charge. In addition, the display unit 221 can also display schedule information in association with the process for which each of the multiple personnel is in charge.

[0071] This allows the construction manager to accurately and intuitively grasp the workload status of each worker without having to interview them, review daily work reports that record daily work content, or check the on-site situation. Furthermore, the schedule information includes schedules unrelated to the target construction project. In this example, the construction manager can grasp the workload status, including schedules for Buildings Y and Z, in addition to Building X, which he manages. Information about other construction projects is often discussed in casual conversation, but is usually difficult to grasp. In this embodiment, even such information is displayed on the schedule and can be used to determine the workload status. The construction manager can easily grasp such workload status and schedule information on a single screen, without having to create separate documents, reconstruct the information in his mind, or switch screens or systems. Furthermore, based on this information, the construction manager can estimate the impact on the entire process from information that was previously unrecognizable, such as whether work hours should be extended and the impact of a high-load process on subsequent processes. As a result, the construction manager can take measures to ensure that all work is completed safely and by the deadline while maintaining the quality of the work. This allows even inexperienced construction managers to efficiently manage construction.

[0072] <Second embodiment> Fig. 7 is a diagram showing the hardware configuration of a construction management system 1 according to a second embodiment. In the first embodiment, the control unit 111 is configured to be able to execute the processes executed by the second estimating unit 121 and the generating unit 123. In the second embodiment, the control unit 111 is further configured to be able to execute the processes executed by the first estimating unit 122 and the notifying unit 124. Hereinafter, descriptions of parts common to the first embodiment may be omitted.

[0073] The second estimator 121 acquires schedule information from the schedule management system 400. The second estimator 121 acquires second input information from the storage unit 112. The second estimator 121 determines the load state based on the second input information and the schedule information.

[0074] When the load state of the first process is high, the notification unit 124 notifies the manager managing the construction, the person in charge of the first process, and the person in charge of the second process of information including the load state. Such notification can be performed, for example, by sending emails to the email addresses of the manager and the person in charge that are registered in advance. This makes it possible to quickly grasp the delay status of the construction and smoothly prepare the necessary adjustments between the parties involved to resolve the delay.

[0075] The first estimation unit 122 acquires first input information (described later) from the storage unit 112. The first estimation unit 122 determines countermeasure information (described later) based on the load state and the first input information. The countermeasure information is information on a plurality of countermeasure proposals that indicate how to deal with a high-load state (extension of work period, increase in personnel, etc.).

[0076] The generating unit 123 acquires process information (described later) from the storage unit 112. The generating unit 123 generates a process chart based on the process information, countermeasure information, and load status. The display unit 221 of the terminal 200 displays the generated process chart.

[0077] 8 is a flowchart showing the main processing. The processing shown in this flowchart may be started in response to a request from the terminal 200.

[0078] When the main processing starts, in S301, the server 100 acquires schedule information from the schedule management system 400. In S302, the server 100 acquires second input information from the storage unit 112. In S303, the server 100 inputs the schedule information and the second input information to the trained model 163 and acquires the load state output from the trained model 163.

[0079] In S304, the server 100 acquires first input information from the storage unit 112. In S305, the server 100 inputs the load state and the first input information to the trained model 173 and acquires countermeasure information output from the trained model 173.

[0080] In S306, the server 100 acquires the process information from the storage unit 112. In S307, the server 100 executes a display information generation process to generate display information (such as a process chart).

[0081] In S308, the server 100 associates the countermeasure information with the display information. Fig. 9 is a diagram showing an example of a process chart. The process chart 93 in Fig. 9 displays the associated countermeasure information.

[0082] 9, the wiring work on the second floor (first process) is in a high load state. The insulation work on the second floor and the insulation work on the third floor are set as the second process.

[0083] As described above, the schedule for April 10th for the person in charge of wiring work on the second floor at Company D is displayed as "8:00-17:00 X Building wiring work," "18:00-20:00 Y Building estimate creation," and "20:00-22:00 Z Building drawing creation."

[0084] In this example, countermeasure information associated with the wiring work on the second floor that has become overloaded is displayed. Specifically, the countermeasure information "Countermeasure 1" is "extending the schedule of Company D" and "increasing the number of employees at Company E," and "Countermeasure 2" is "increasing the number of employees at Company D."

[0085] Countermeasure 1 assumes that Company D's staff are under pressure and will not be able to complete the wiring work on the second floor by April 10th as scheduled. Furthermore, Company D does not have the manpower to spare at this time. Therefore, it is recommended that Company D's schedule be extended beyond April 11th, which would delay the start of Company E's insulation work on the second floor. Furthermore, Company E has the manpower to spare at this time. Therefore, Countermeasure 1 further increases the number of workers working on the insulation work on the second floor at Company E, thereby allowing the work on the second floor to be completed as scheduled.

[0086] Countermeasure 2 assumes that Company E is unable to increase the number of workers working on the insulation work on the second floor. In this case, Company D must complete the wiring work on the second floor by April 10th as scheduled. Therefore, Countermeasure 2 recommends increasing the number of workers at Company D as the second option.

[0087] Based on the above countermeasure information, the construction manager coordinates the work with Company D and Company E. Then, as a result of the coordination, he decides whether to adopt Countermeasure Plan 1 or Countermeasure Plan 2.

[0088] 8, in S309, the server 100 outputs display information (such as a process chart) to the display unit 221. As a result, the display unit 221 displays a process chart 93 such as that shown in FIG.

[0089] In S310, when the load state of the first process is in a high load state, the notification unit 124 of the server 100 notifies the administrator and the person in charge of the process in a high load state (first process) and the related process (second process) of the load state, and then terminates the main processing.

[0090] In the above example, the manager, the person in charge of the wiring work on the second floor (Company D), which is under heavy load, and the people in charge of the related processes of the insulation work on the second floor (Company E) and the wiring work on the third floor (Company D), are notified of this. This allows them to quickly grasp the delay status of the work and smoothly prepare for the necessary adjustments between the parties involved to resolve the delay.

[0091] In this embodiment, the determination of whether or not a high-load state exists and the determination of countermeasure information (plurality of countermeasure proposals) when a high-load state exists are performed using a trained model that has been subjected to machine learning. Note that these determinations or judgments are not limited to those using a trained model, and may be determined or judged using other estimation models or based on some rules. The learning process performed in this embodiment will be described below with reference to FIGS. 10 to 12.

[0092] 10 is a diagram illustrating the learning dataset 160. There is a certain correlation between the schedule information and second input information (input data) and the load state (output data).

[0093] For this reason, in this embodiment, the "load state" is determined using AI (Artificial Intelligence). For example, a trained model that infers output Y from input X is generated based on training data consisting of a combination of input data and output data. For example, training can be performed using supervised learning using a neural network. As a training algorithm used for model generation, deep learning, which learns to extract feature quantities themselves, can also be used.

[0094] In this embodiment, the storage unit 112 stores the training dataset 160 and the trained model 163. The control unit 111 inputs the schedule information and the second input information to the second trained model (trained model 163), and outputs the load state and the like as an estimation result from the trained model 163.

[0095] The trained model 163 is a model that has undergone machine learning processing using training data so as to output the load state, etc. as an estimated result when schedule information and second input information are input. The second input information includes information indicating the schedule priority of each of the multiple personnel, information on the company to which each of the multiple personnel belongs, and information on each of the multiple processes.

[0096] Specifically, a learning process is performed using the above-described set of input and output data (learning dataset 160). Then, using the obtained trained model, an estimation process is performed in which the schedule information and the second input information are used as input data and the load state and the like are used as output data (estimation results). The set of input and output data is accumulated daily in the storage unit 112 of the server 100 as historical information on past construction work.

[0097] The above-mentioned "input data" and "output data" are as shown in Fig. 10. As shown in Fig. 10, input data 161 includes schedule information and second input information. Output data 162 includes a load state and an adjustment result.

[0098] A combination of input data 161 and output data 162 is prepared as the learning data set 160. Here, the output data 162 is the correct answer data. An example of a combination of the input data 161 and the output data 162 will be described below.

[0099] For example, input data includes schedule priority, company information (responsible company, etc.), process information (task content, implementation period, schedule information of the person in charge during the implementation period, implementation time, etc.), and output data includes adjustment results (additional personnel, number of extended work days), load status, etc. Schedule priority is the priority (high, medium, or low can be set) of each schedule registered by the person in charge, and this priority can be set in advance in the schedule management system 400.

[0100] For example, the schedule priority is "Task A: high priority," the company in charge is "Company D," the work content is "wiring work," the implementation period is "March 1st," the schedule information of the person in charge during the implementation period is "March 1st: 8:00-17:00 wiring work, 18:00-22:00 Task A," and the implementation time is "March," and the output data is additional personnel "0 people," number of extended work days "1 day," and load status "high load status." If there are multiple people in charge of the work, schedule information for all people is obtained.

[0101] In the above example, during the wiring work period, Company D's staff are scheduled to work on Task A, which has a higher priority, after the wiring work from 8:00 to 17:00. In this case, even if the wiring work is unexpectedly delayed, the scheduled work cannot be completed on the same day because the staff must complete Task A, which has a higher priority, which is scheduled after the wiring work. As a result, the wiring work is extended until the next day, and the extended work period is "1 day." Furthermore, because March is Company D's busy season, additional personnel could not be assigned.

[0102] In this embodiment, the data manager checks the input and output data and then sets the load status to "high load" or "normal." For example, if a high-priority task other than the target construction is scheduled for a predetermined time (predetermined ratio) or more, the load status may be set to "high load." Furthermore, information other than the schedule may also be taken into consideration. For example, if there is one or more additional personnel, or if the work is extended by one or more days, the load status may be set to "high load" regardless of the schedule registration status. Furthermore, the load status may be set automatically using the above rules.

[0103] As another example, suppose the schedule priority is "Work B: low priority," the work content is "insulation work," the implementation period is "February 3rd," the responsible company is "Company E," the schedule information of the person in charge during the implementation period is "February 3rd: 8:00-17:00 insulation work, 18:00-22:00 Work B," and the implementation time is "February," and the output data is additional personnel "0 people," number of extended work days "0 days," and load status "normal status."

[0104] In the above example, during the insulation work period, Company E's staff are scheduled to work on Work B, which has a lower priority, after the insulation work from 8:00 to 17:00. In this case, even if the insulation work is unexpectedly delayed, the insulation work can be completed on the same day by postponing Work B, which has a lower priority, to the next day. Also, because February is not Company E's busy season, additional personnel can be added if work is likely to be delayed.

[0105] As explained above, if a person's schedule contains many high-priority tasks, the workload will be high, increasing the likelihood of work delays or additional personnel being required. In this way, there is a certain correlation between the schedule of the person in charge and the workload. Furthermore, depending on the specific company (e.g., a small company), the specific type of process (a process prone to equipment problems), and the time of execution (e.g., a busy period), there is a possibility that work delays are more likely to occur (the workload is more likely to be high). Therefore, it can be said that there is a certain correlation between this information and the workload.

[0106] From the above, when a learning process is performed using the above-mentioned set of input and output data, the resulting trained model can be used to accurately estimate the load state. For processes with high loads, the construction manager can make adjustments such as adjusting the schedule or adding personnel, thereby completing the work by the deadline. In this way, even an inexperienced construction manager can efficiently manage construction.

[0107] 11 is a diagram for explaining the learning dataset. There is a certain correlation between the load state and the first input information (input data) and the countermeasure information (output data).

[0108] In this embodiment, the storage unit 112 stores a training dataset 170 and a trained model 173. The control unit 111 inputs the load state and the first input information to the first trained model (trained model 173), and outputs countermeasure information as an estimation result from the trained model 173. The trained model 173 is a model that has been subjected to machine learning processing using training data so as to output countermeasure information as an estimation result when the load state and the first input information are input.

[0109] The first input information includes information on the company to which each of the plurality of personnel belongs and information on each of the plurality of processes. The countermeasure information is information indicating countermeasure methods for the first process and the second process when a high-load state occurs.

[0110] Countermeasures include extending the work period of either or both of the first and second steps, and increasing the number of workers in either or both of the first and second steps.

[0111] Specifically, a learning process is performed using the above-described set of input and output data. Then, using the obtained trained model, an estimation process is performed in which the load state and the first input information are used as input data and the countermeasure information is used as output data (estimation result).

[0112] In the example shown in Figure 9 above, as countermeasures for the wiring work (first process) on the second floor, which is in a high-load state, "Countermeasure 1" is to "extend the schedule of Company D (first process)" and "increase the number of personnel at Company E (second process)," and "Countermeasure 2" is to "increase the number of personnel at Company D (first process)."

[0113] The above-mentioned "input data" and "output data" are as shown in Fig. 11. As shown in Fig. 11, input data 171 includes a load state and first input information. Output data 172 includes countermeasure information.

[0114] A combination of input data 171 and output data 172 is prepared as a training data set 170. Here, the output data 172 is correct answer data. An example of a combination of the input data 171 and the output data 172 will be described below.

[0115] For example, the input data is the company in charge of the first process, the load state, the work content, and the implementation time, and the company in charge of the second process, the work content, and the implementation time, and the output data is countermeasure information.

[0116] For example, for the first process, the company in charge is "Company D," the load condition is "high load," the work content is "wiring work," and the implementation time is "March," and for the second process, the company in charge is "Company E," the work content is "insulation work," and the implementation time is "March," and the output data includes countermeasure information such as extension of the first process by "1 day," increase in personnel for the first process by "0 people," extension of the second process by "0 days," and increase in personnel for the second process by "1 person."

[0117] In the above example, assume that Company D was unable to increase staff because March was their busy season, while Company E was able to increase staff because March was not their busy season. As a result, Company D's work, which was heavy, was extended by one day, while Company E, which had manpower to spare, increased staff and completed the insulation work one day shorter than planned. Alternatively, if Company D had a small number of employees, it would be difficult to increase staff, but if Company E had a large number of employees, it would be easier to increase staff. In this way, there is a certain correlation between the company in charge, the implementation timing, and the countermeasures taken. Furthermore, because there are processes that can only be performed by one person and therefore cannot increase staff, it can be said that there is also a certain correlation between the type of process and the countermeasures taken.

[0118] In such a situation where the load is high and adjustments are necessary, there is a certain correlation between the input data and the output data regarding the determination of whether to increase personnel or extend or shorten the period for the first or second process. When outputting a countermeasure using the trained model 173, one countermeasure may be output, or multiple countermeasures may be output in descending order of likelihood, such as countermeasure plan 1, countermeasure plan 2, and so on.

[0119] From the above, when learning processing is performed using the above-mentioned set of input and output data, the obtained trained model can be used to determine the optimal countermeasure plan. Construction managers can make adjustments such as adjusting schedules and adding personnel for processes that are under high load, so that construction can be completed by the deadline. In this way, even construction managers with little experience can efficiently manage construction.

[0120] In particular, the more input and output data sets are accumulated, the more AI learns, allowing it to present more appropriate countermeasures and more accurately predict work delays. Furthermore, given the ongoing decline in the workforce in the construction industry, there is a need to quickly train construction managers who can make decisions such as increasing or replacing workers and reviewing processes. By utilizing the method according to this embodiment, it is also possible to quickly train construction managers of a certain level.

[0121] FIG. 12 is a flowchart of the learning process. Hereinafter, "step" may be simply referred to as "S." Here, the process of learning the trained model 163 will be described. As shown in FIG. 12, when the learning process starts, the server 100 selects training data from the training dataset 160 in S401, and proceeds to S402.

[0122] In S402, the server 100 inputs the selected training data into the estimation model, and proceeds to S403. In S403, the server 100 executes the estimation process, outputs the estimation result, and proceeds to S404. In S404, the server 100 updates the parameters of the estimation model based on the error between the estimation result and the ground truth data corresponding to the training data, and proceeds to S405.

[0123] In S405, the server 100 determines whether learning has been completed based on all of the learning data. If the server 100 determines that learning has been completed based on all of the learning data (YES in S405), the process proceeds to S406. If the server 100 does not determine that learning has been completed based on all of the learning data (NO in S405), the process returns to S401. In S406, the server 100 stores the trained estimation model as the trained model 163 and ends the learning process.

[0124] The trained model 173 is also trained using the training dataset 170 in the same manner as above.

[0125] [Modification] In the first embodiment, the process schedule shown in Figures 5 and 6 is displayed, and in the second embodiment, the process schedule shown in Figure 9 is displayed. However, the present invention is not limited to this, and a process schedule as described below may be displayed.

[0126] Fig. 13 is a flowchart of a route display process according to a modified example. Fig. 14 is a diagram showing an example of a process schedule according to a modified example. The process schedule in Fig. 14 is illustrated in a simpler form than the process schedules described using Figs. 5 and 6.

[0127] As shown in FIG. 13, when the route display process starts, the server 100 classifies the route from the start to the end of construction into a plurality of routes in S501.

[0128] The example in Fig. 14 shows a bar chart of multiple companies (multiple staff members) in charge of the construction work and the processes for which each staff member is responsible. In Fig. 14, when a process can be started only after another process is completed, the former process and the latter process are connected by a line.

[0129] In this example, after the process of Company A is completed, the process of Company C can start. After the process of Company B is completed, the process of Company C can start. After the process of Company C is completed, the process of Company F can start. After the process of Company F is completed, the process of Company G can start.

[0130] Furthermore, after the process of company D is completed, the process of company E can start. After the process of company E is completed, the process of company F can start.

[0131] In this way, the order of the process is classified into a route that proceeds in the order of Company A, Company C, Company F, and Company G (referred to as "route 1"), a route that proceeds in the order of Company B, Company C, Company F, and Company G (referred to as "route 2"), and a route that proceeds in the order of Company D, Company E, Company F, and Company G (referred to as "route 3").

[0132] Returning to FIG. 13, in S502, the server 100 selects the route with the longest number of required days from the number of required days for each route.

[0133] Route 1 is the route that takes the longest number of days from start to finish of construction. For example, if Company A's work on Route 1 is delayed, Company C's work, which starts immediately afterwards, will also be delayed. On the other hand, in the case of Route 2, even if Company B's work is delayed, there is ample time before Company C's work starts next, so delays are unlikely to occur.

[0134] Furthermore, if Company C's process on Route 1 is delayed, Company F's process, which starts immediately thereafter, will also be delayed. On the other hand, in the case of Route 3, even if Company E's process is delayed, there is still time before Company F's process, which starts next, so delays are unlikely to occur.

[0135] In this way, it can be said that delays are likely to occur on Route 1, which has the longest number of days required from the start to the end of construction work.

[0136] 13 , in S503, the server 100 changes the color and size of the bar chart included in the route requiring the longest number of days, and then ends the route display process. In this modified example, of the multiple routes in the multiple processes, the route requiring the longest number of days from the start to the end of construction is displayed in a different manner from the other routes.

[0137] Specifically, the bar chart for each process on route 1, which takes the longest number of days from start to finish, is displayed as a red image with a large vertical width, and the lines connecting these processes are displayed as solid lines. The bar charts for each process on routes 2 and 3 are displayed as white images with a smaller vertical width than route 1, and the lines connecting these processes are displayed as dashed lines. However, even in the bar charts for routes 2 and 3, those included in route 1 are displayed in the same display method as route 1 (red images with a large vertical width). This makes it easier to intuitively grasp routes that are likely to be affected by delays, and it is possible to draw attention to such routes.

[0138] This modification can be applied to the first or second embodiment. In the example shown in Fig. 14, the process of company A in the first route corresponds to the first process, and the process of company C corresponds to the second process. The process of company C in the first route corresponds to the first process, and the process of company F corresponds to the second process. The process of company F in the first route corresponds to the first process, and the process of company G corresponds to the second process.

[0139] In the present embodiment and the modified example, a bar chart schedule has been used as the schedule, but any other schedule may be used. For example, the schedule may be a network schedule. In this case, the display showing each process may be displayed with associated schedule information. Furthermore, if a process is in a high load state, the process (first process) may be displayed in red, and a process (second process) that may be affected by this process and cause a delay may be displayed in pink. Furthermore, path 1 in the modified example corresponds to the critical path.

[0140] The embodiments disclosed herein are intended to be combined as appropriate within the scope of any technical inconsistency. The embodiments disclosed herein should be considered to be illustrative and not restrictive in all respects. The technical scope of the present disclosure is defined by the claims, not the description of the above-mentioned embodiments, and is intended to include all modifications within the meaning and scope of the claims.

[0141] 1 Construction management system, 91 to 94 Process chart, 100 Server, 111, 211 Control unit, 112, 212 Memory unit, 113, 213 Communication unit, 121 Second estimation unit, 122 First estimation unit, 123 Generation unit, 124 Notification unit, 161, 171 Input data, 162, 172 Output data, 163, 173 Trained model, 200 Terminal, 220 Input unit, 221 Display unit, 400 Schedule management system.

Claims

1. A construction management system that manages the construction of a construction project consisting of multiple processes that are handled by multiple personnel, comprising: a control unit that generates a construction schedule for the construction project; and a display unit that displays the schedule, wherein the control unit determines a load state indicating the workload of the multiple personnel based on information including schedule information for the multiple personnel, and the display unit displays the schedule information and the load state in association with the process that each of the multiple personnel is responsible for.

2. The construction management system of claim 1, wherein the construction management system is configured to be connectable to a schedule management system that manages the schedules of the multiple personnel in charge, the control unit acquires the schedule information from the schedule management system, and the schedule information includes schedules unrelated to the construction work.

3. A construction management system as described in claim 1 or claim 2, wherein the schedule is a bar chart schedule displaying a plurality of bar charts corresponding to the plurality of processes, the plurality of processes including a first process and a second process that is a process that follows the first process, the second process being a process that will be affected by a delay in work on the first process, and the display unit displays an image connecting the first bar chart corresponding to the first process and the second bar chart corresponding to the second process.

4. The construction management system described in claim 3, wherein the load state includes a high load state in which delays to work on the first process are predicted, and the display unit displays the schedule information in association with a bar chart corresponding to the process for which each of the multiple personnel is in charge, and when the load state of the first process is the high load state, displays the first bar chart in a first manner different from when the load state of the first process is not the high load state.

5. A construction management system as described in claim 4, wherein the display unit displays the first bar chart in the first manner and the second bar chart in the second manner when the load state of the first process is the high load state.

6. A construction management system as described in claim 5, wherein when the load state of the first process is the high load state, the control unit notifies the manager managing the construction work, the person in charge of the first process, and the person in charge of the second process of the load state.

7. A construction management system as described in any one of claims 4 to 6, wherein, among the multiple routes in the multiple processes, the route that takes the longest number of days from the start to the end of the construction work is displayed in a manner different from the other routes.

8. A construction management system as described in any one of claims 4 to 7, wherein the control unit inputs the load condition and first input information into a first trained model, and outputs countermeasure information from the first trained model as an estimated result; the first trained model is a model that has been subjected to machine learning processing using training data so as to output the countermeasure information as the estimated result when the load condition and the first input information are input; the first input information includes information on the company to which each of the plurality of personnel belongs and information on each of the plurality of processes; the countermeasure information is information indicating countermeasure methods for the first process and the second process when the high-load condition occurs; and the countermeasure methods include a method of extending the work period of either or both of the first process and the second process, and a method of increasing the number of workers for either or both of the first process and the second process.

9. A construction management system as described in any one of claims 1 to 8, wherein the control unit inputs the schedule information and second input information into a second trained model, and outputs the load state as an estimated result from the second trained model; the second trained model is a model that has been subjected to machine learning processing using training data so as to output the load state as the estimated result when the schedule information and the second input information are input; and the second input information includes information indicating the schedule priority of each of the multiple personnel, information about the company to which each of the multiple personnel belongs, and information about each of the multiple processes.

10. A construction management method for managing the construction of a construction project consisting of multiple processes for which multiple personnel are responsible, comprising the steps of: generating a construction schedule for the construction project; and displaying the schedule, wherein the generating step includes a step of determining a load state indicating the workload of the multiple personnel based on information including schedule information for the multiple personnel; and the displaying step includes a step of displaying the schedule information and the load state in association with the process for which each of the multiple personnel is responsible.

Images