Method and apparatus for predicting construction site complaints

By integrating GIS, on-site data, and weather information with AI, the method predicts construction site complaints and accidents, improving accuracy and enabling proactive measures to reduce delays and resident inconvenience.

KR103047011B1Active Publication Date: 2026-11-11SOL IT CO LTD
View PDF 4 Cites 0 Cited by

Patent Information

Application Number
KR1020250177129
Authority / Receiving Office
KR · KR
Patent Type
Patents
Current Assignee / Owner
Filing Date
2025-11-20
Publication Date
2026-11-11
Estimated Expiration
2045-11-20

AI Technical Summary

Technical Problem

Current construction site management systems lack the ability to accurately predict environmental complaints such as noise and dust, leading to reactive measures that cause construction delays and resident inconvenience, particularly at small-scale sites, due to the non-integration of multimodal data and lack of proactive decision-making support.

Method used

A method and apparatus that integrates GIS-based spatial information, on-site shooting information, process management data, and weather information using AI models to predict complaints and provide preemptive response measures, supporting local governments and construction companies in minimizing delays.

Benefits of technology

Enhances complaint prediction accuracy by 35% and enables proactive response measures within 6 to 24 hours, reducing decision-making time and predicting safety accidents, thus minimizing construction delays and resident inconvenience.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 112025130262465-PAT00003_ABST
    Figure 112025130262465-PAT00003_ABST
Patent Text Reader

Abstract

A method for predicting construction site complaints according to exemplary embodiments of the present invention may include: a method executed by at least one processor, comprising the steps of: acquiring geographic information system-based information of a construction site, site photography information, process management information, environmental measurement information, and external weather information; extracting geographic information system-based feature information based on the geographic information system-based information using a rule-based model or a trained artificial intelligence model; extracting site work feature information based on the site photography information using a trained artificial intelligence model; and predicting the risk of construction site complaints at a predetermined time point based on the geographic information system-based feature information, the site work feature information, the process management information, the environmental measurement information, and the external weather information using a rule-based model or a trained artificial intelligence model.
Need to check novelty before this filing date? Find Prior Art

Description

Technology Field

[0001] The present invention relates to a method, apparatus, and system for predicting and managing construction site complaints in advance by utilizing GIS (Geographic Information System)-based spatial information, multimodal sensor data, and AI. In particular, it relates to a technology for minimizing resident inconvenience and preventing construction delays by early detection of environmental complaints, such as noise and dust, occurring at small-scale and medium-scale construction sites. Background Technology

[0002] Environmental hazards such as noise and dust generated at construction sites can cause significant inconvenience to residents living near the site. For example, noise stress and air pollution can lower the quality of life for residents. In such cases, residents may file complaints regarding the noise and dust generated at the construction site to delay or halt construction.

[0003] According to the results of a big data analysis of civil complaints operated by the government, the number of complaints regarding construction sites has been rapidly increasing over the past few years. In other words, residents' awareness of environmental hazards associated with construction sites is rising. As of 2023, 156,927 noise-related complaints were received by local governments in Korea, and among these, complaints caused by construction site noise accounted for 75.3% (118,904 cases). In particular, complaints arising from small- and medium-sized construction sites (gross floor area of ​​5,000㎡ or less) account for more than 60% of all construction-related complaints; however, systematic management is currently lacking due to an environmental management system centered on large-scale construction sites.

[0004] Currently, the management of environmental hazards at construction sites relies primarily on a reactive approach. Since environmental hazards are measured and necessary measures are taken only after a complaint arises, the occurrence of complaints cannot be prevented in advance. This leads to problems such as construction delays (an average of 5–10 days), additional costs, and intensified conflicts with residents. In particular, environmental management capabilities are severely limited at small-scale construction sites due to the difficulty of installing IoT sensors and deploying dedicated management personnel. Environmental measurement solutions currently on the market also focus solely on real-time data analysis, making it difficult to predict potential issues and respond proactively.

[0005] Korean Published Patent Application No. 10-2025-0125047 discloses a method for managing construction complaints based on an integrated environmental measurement module. The method according to the document has a limitation in that it generates guide information based only on environmental measurement information and surrounding facility information. Specifically, it has the following problems.

[0006] 1) Non-integration of Multimodal Data: Accuracy is low because on-site imaging information and process management data are not utilized, and judgments are made solely based on environmental measurements regardless of actual work content.

[0007] 2) Lack of prior prediction: Since response is only made after environmental measurements have already increased, preventing complaints is impossible.

[0008] 3) Lack of decision-making support: Providing only simple guide information does not assist local governments in determining inspection priorities or construction companies in establishing preemptive response plans.

[0009] 4) Unpredictable future changes: Ultra-short-term (6-24 hours) forecasting is difficult as it does not reflect future weather changes or changes in process schedules.

[0010] Therefore, technology is required to accurately predict the likelihood of complaints arising at construction sites and to provide necessary information to stakeholders, such as construction companies and local governments, in a timely manner. Prior art literature

[0011] Korean Published Patent Application No. 10-2025-0125047 The problem to be solved

[0012] One objective of the present invention is to provide a method and apparatus capable of accurately predicting complaints at construction sites in advance by integratively analyzing GIS-based spatial information, on-site shooting information, process management information, environmental measurement information, and weather information, and presenting preemptive response measures to minimize inconvenience to residents based on the prediction results. In addition, it is also an objective of the present invention to support local governments managing multiple construction sites in efficiently determining inspection priorities and to support construction companies in establishing preemptive response plans to prevent construction delays. means of solving the problem

[0013] A method for predicting construction site complaints according to exemplary embodiments of the present invention may include: a method executed by at least one processor, comprising the steps of: acquiring geographic information system-based information of a construction site, site photography information, process management information, environmental measurement information, and external weather information; extracting geographic information system-based feature information based on the geographic information system-based information using a rule-based model or a trained artificial intelligence model; extracting site work feature information based on the site photography information using a trained artificial intelligence model; and predicting the risk of construction site complaints at a predetermined time point based on the geographic information system-based feature information, the site work feature information, the process management information, the environmental measurement information, and the external weather information using a rule-based model or a trained artificial intelligence model.

[0014] In one embodiment, the geographic information system-based information includes at least one of surrounding population density, surrounding facility location information, past civil complaint occurrence information, surrounding traffic information, and topographic information, and the field work characteristic information may include at least one of field work type characteristic information, field work location characteristic information, field equipment characteristic information, and field personnel characteristic information.

[0015] In one embodiment, the process management information includes at least one of work type information, process progress rate, scheduled work time, scheduled work location, scheduled equipment information, and scheduled personnel information, the environmental measurement information includes at least one of noise data, dust data, and weather data of the construction site, and the external weather information may include at least one of wind direction, wind speed, humidity, atmospheric pressure, temperature, solar radiation, and precipitation.

[0016] In one embodiment, the above-described method may further include the step of determining a guide for response measures based on at least one of the construction site complaint risk level, geographic information system-based feature information, site work feature information, process management information, environmental measurement information, and external weather information, using a rule-based model or a trained artificial intelligence model.

[0017] In one embodiment, the above-described method may further include the step of providing a user with a guide on the construction site complaint risk level and the response measures when the construction site complaint risk level is greater than a predetermined threshold.

[0018] In one embodiment, the above-described method may further include the step of determining whether a construction site is a priority inspection site based on at least one of the construction site complaint risk level, geographic information system-based feature information, site work feature information, and process management information, using a rule-based model or a trained artificial intelligence model.

[0019] In one embodiment, the above-described method may further include the step of predicting the risk of a construction site safety accident at a predetermined time point based on the geographic information system-based feature information, the site work feature information, the process management information, the environmental measurement information, and the external weather information, using a rule-based model or a trained artificial intelligence model.

[0020] In one embodiment, the above-described method may further include the step of determining a guide for response measures based on at least one of the construction site safety accident risk level, geographic information system-based feature information, site work feature information, process management information, environmental measurement information, and external weather information, using a rule-based model or a trained artificial intelligence model.

[0021] An apparatus for predicting construction site complaints according to exemplary embodiments of the present invention may include at least one memory; and at least one processor that executes instructions stored in the at least one memory.

[0022] The above-mentioned at least one processor acquires geographic information system-based information, site photography information, process management information, environmental measurement information, and external weather information of a construction site, and uses a rule-based model or a trained artificial intelligence model to extract geographic information system-based feature information based on the above-mentioned geographic information system-based information, and uses a trained artificial intelligence model to extract site work feature information based on the above-mentioned site photography information, and uses a rule-based model or a trained artificial intelligence model to predict the risk of construction site complaints at a predetermined point in time based on the geographic information system-based feature information, the above-mentioned site work feature information, the above-mentioned process management information, the above-mentioned environmental measurement information, and the above-mentioned external weather information.

[0023] According to exemplary embodiments of the present invention, an application program stored on a recording medium may be provided to execute the method for predicting the aforementioned construction site complaints when operated by at least one processor. Effects of the invention

[0024] According to exemplary embodiments of the present invention, the following technical effects are provided.

[0025] 1) Multimodal Data Integration: By comprehensively analyzing GIS-based spatial information, on-site footage, process management data, environmental measurement data, and weather information, the accuracy of complaint prediction is improved by more than 35% compared to the existing environmental measurement-based method.

[0026] 2) Predictability: Preemptive response measures can be implemented by predicting the risk of complaints within the ultra-short term (6 to 24 hours) in advance.

[0027] 3) Intelligent Decision Support: Reduces decision-making time for management personnel through predictive response guides and automatic determination of inspection priorities.

[0028] In addition, the present invention can be utilized as an overall risk management system for construction sites by predicting not only construction site complaints but also the risk of safety accidents. Brief explanation of the drawing

[0029] Figure 1 shows a block diagram of a construction site complaint prediction system according to one embodiment of the present invention. FIG. 2 shows a block diagram of a device according to one embodiment of the present invention. FIG. 3 shows a flowchart of operations for predicting the risk of civil complaints at a construction site according to one embodiment of the present invention. FIG. 4 shows a flowchart of operations for predicting the risk of safety accidents at a construction site according to one embodiment of the present invention. Specific details for implementing the invention

[0030] The advantages and features of the present invention and the methods for achieving them will become clear by referring to the embodiments described below in detail together with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below but can be implemented in various different forms.

[0031] To clearly explain the embodiments of the present invention, parts unrelated to the description may be omitted. Additionally, in describing the embodiments of the present invention, if it is determined that a detailed description of related known components or functions could obscure the essence or description of the present invention, such detailed description may be omitted.

[0032] In this specification, the terms “…part,” “…unit,” and “…module” may refer to a unit that processes at least one function or operation. The “…part,” “…unit,” and “…module” may be implemented in hardware, software, or a combination of hardware and software.

[0033] The classification of components in this specification is merely based on the primary function each component is responsible for. That is, two or more components may be combined into a single component, or a single component may be divided into two or more components based on more subdivided functions. Additionally, each component may perform some or all of the functions of other components in addition to its primary function, and some of the primary functions of each component may be exclusively performed by other components.

[0034] In describing the components in this specification, terms such as "first," "second," etc., may be used. These terms are intended for convenience of explanation to distinguish one component from another, and unless otherwise specifically stated, the nature, order, etc., of the components are not limited by these terms.

[0035] In each step mentioned in this specification, the steps may proceed differently from the specified order unless the context clearly indicates a specific order. That is, the steps may be performed in the same order as specified, substantially simultaneously, or in the reverse order.

[0036] In this specification, "and / or" may mean each of the listed components and a combination of two or more of the components. For example, "A, B and / or C" may be used with the same meaning as "at least one of A, B and C."

[0037] In this specification, the meaning of "may include a, b, c, etc." may be a concept that includes the meaning of "may include at least one of a, b, and c."

[0038] In this specification, "predetermined" may be a concept encompassing "pre-configured." For example, "pre-configured" may be performed by a provider, manager, operator, etc. of a device, system, or service.

[0039] In this specification, unless otherwise specifically noted, the weights in "weighted sum" and "weighted average" may be pre-set. For example, the weights may be pre-set by the provider, manager, operator, etc. of the device, system, or service.

[0040] In this specification, "time series data" may mean data observed in chronological order. For example, "time series data" may be data observed in chronological order according to the passage of time, such as hourly, daily, weekly, quarterly, semi-annually, monthly, or yearly (i.e., unit of time).

[0041] In this specification, "third device" may refer to a separate device other than the user device and the server device. In this specification, "third device" may be mentioned multiple times, provided that each of the third devices does not necessarily refer to the same device.

[0042] In this specification, "construction site complaints" may refer to complaints raised by residents due to noise and dust, etc., generated at a construction site.

[0043] In this specification, "Geographic Information System (GIS)" may mean a system that collects, stores, analyzes, manages, and visualizes geographic locations (e.g., addresses, GPS coordinates, GIS coordinates) and attribute information (e.g., spatial information) associated with said locations.

[0044] In this specification, "Process Management System (Enterprise Resource Planning, ERP)" may refer to a system that collects, stores, analyzes, and manages information for the integrated management of resources, schedules, work processes, and quality at a construction site.

[0045] According to exemplary embodiments of the present invention, a method, apparatus, and system for predicting construction site complaints may be provided.

[0046] FIG. 1 shows a block diagram of a construction site complaint prediction system according to one embodiment of the present invention. Hereinafter, for convenience of explanation, a service for predicting construction site complaints may be abbreviated as 'construction site complaint prediction service', and a system for predicting construction site complaints may be abbreviated as 'construction site complaint prediction system'.

[0047] Referring to FIG. 1, the construction site complaint prediction system (100) may include a user device (101) and a server device (102).

[0048] The user device (101) may be a device of a user who wishes to receive a construction site complaint prediction service. In one example, the user device (101) may be a device of a construction site manager, a construction company headquarters manager, a local government official (e.g., a public official in charge of complaints) who wishes to check the risk level of complaints at the construction site. The following description may be an example in which the user device (101) is a device of a construction site manager.

[0049] A user device (101) can connect to a server device (102) via a network and interact with the server device (102), such as transmitting or receiving certain data. For example, the user device (101) can provide the server device (102) with information based on a geographic information system of a construction site, on-site filming information, process management information, environmental measurement information, external weather information, etc. The user device (101) can receive from the server device (102) information such as the risk level of complaints at the construction site at a certain point in time, a guide on measures to prevent complaints, whether the construction site is a priority site for inspection, the risk level of safety accidents at the construction site at a certain point in time, a guide on measures to prevent safety accidents, and the priority of construction site inspections.

[0050] The user device (101) may be a device of various forms, for example, an electronic device such as a smartphone, laptop, desktop computer, or tablet PC. Although FIG. 1 illustrates only one user device (101), multiple devices including the user device (101) may connect to the server device (102) to use the construction site complaint prediction service. However, the range of accessible information may vary depending on the authority granted to the user device (101).

[0051] The server device (102) may be a device of a service provider that provides a construction site complaint prediction service. The server device (102) may be a computing device that can be implemented in various forms, such as a standard server, a server group, or a rack server system. The server device (102) may interact with a user device (101), etc., through a network, such as transmitting or receiving certain data.

[0052] The server device (102) can receive geographic information system-based information, site shooting information, process management information, environmental measurement information, and external weather information of the construction site via a network from the user device (101) and a third device (not shown; for example, a geographic information system data server, a process management system data server, an Internet of Things (IoT) sensor, a noise sensor, a dust sensor, a weather sensor, a weather forecast data server, a meteorological agency data server, etc.). The server device (102) can provide the user device (101) with the risk of civil complaints at the construction site at a predetermined point in time according to the methods of exemplary embodiments of the present invention, a guide for measures to prevent civil complaints, whether the construction site is a priority inspection site, the risk of safety accidents at the construction site at a predetermined point in time, a guide for measures to prevent safety accidents, and the priority of construction site inspections.

[0053] For example, the server device (102) can predict the risk of construction site complaints at a predetermined point in time based on geographic information system-based information, site shooting information, process management information, environmental measurement information, and external weather information of the construction site provided by the user device (101), a third device, etc. The server device (102) can determine a guide for preventive response measures for complaints based on the risk of construction site complaints and at least one of geographic information system-based feature information, site work feature information, process management information, environmental measurement information, and external weather information. The server device (102) can determine whether a construction site is a priority inspection site or the priority of construction site inspection based on the risk of construction site complaints and at least one of geographic information system-based feature information, site work feature information, and process management information. The server device (102) can predict the risk of safety accidents at the construction site at a predetermined point in time based on geographic information system-based feature information, site work feature information, process management information, environmental measurement information, and external weather information. The server device (102) can determine a guide for safety accident prevention response measures based on at least one of the construction site safety accident risk level, geographic information system-based feature information, site work feature information, process management information, environmental measurement information, and external weather information. The server device (102) can determine whether a construction site is a priority inspection site or the priority of construction site inspection based on at least one of the construction site safety accident risk level, geographic information system-based feature information, site work feature information, and process management information.

[0054] The server device (102) may store a rule-based model and a trained artificial intelligence model for extracting geographic information system-based feature information; a trained artificial intelligence model for extracting field work feature information; a rule-based model and a trained artificial intelligence model for predicting the risk of construction site complaints at a predetermined point in time; a rule-based model and a trained artificial intelligence model for determining a guide for complaint prevention response measures; a rule-based model and a trained artificial intelligence model for determining whether a construction site is a priority inspection site; a rule-based model and a trained artificial intelligence model for predicting the risk of safety accidents at a predetermined point in time; a rule-based model and a trained artificial intelligence model for determining a guide for safety accident prevention response measures; a rule-based model and a trained artificial intelligence model for determining the priority of construction site inspections; etc. The server device (102) can provide the user device (101) with the risk of construction site complaints at a predetermined time, a guide on measures to prevent complaints, whether the construction site is a priority inspection site, the risk of safety accidents at a predetermined time, a guide on measures to prevent safety accidents, and the priority of construction site inspections, in the form of a platform, upon the request of the user device (101).

[0055] The relationship between the server device (102) and the user device (101) as illustrated in FIG. 1 can be established when the construction site complaint prediction service is provided in the form of a platform. According to another embodiment of the present invention, the construction site complaint prediction service may be provided in the form of a non-network-based program rather than a platform. In this case, unlike as illustrated in FIG. 1, the construction site complaint prediction service may be provided by a program or application installed on the user device (101) without a server device (102). When the construction site complaint prediction service is provided in the form of a non-network-based program, the server device (102) may be the entity providing the program or application.

[0056] FIG. 2 illustrates a block diagram of a device according to an embodiment of the present invention. FIG. 2 illustrates an example of the structure of a user device (101) or a server device (102) of FIG. 1. In this specification, the user device (101) and the server device (102) may be collectively referred to as 'devices'.

[0057] Referring to FIG. 2, the device may include a control unit (201), a communication unit (202), and a storage unit (203).

[0058] The control unit (201) can control the overall functions and operations of the device. The control unit (201) can control the components of the device, for example, the communication unit (202), the storage unit (203), and at least one other component. That is, the control unit (201) can provide information or data necessary for the operation of the components of the device and can perform operations based on the information or data generated or managed by the components. For example, the control unit (201) may include at least one processor, at least one circuit, etc. The at least one processor may include at least one of a central processing unit (CPU), a graphics processing unit (GPU), and a neural network processing unit (NPU). The control unit (201) can perform the necessary controls to enable the device to operate according to various embodiments described in this specification. For example, the control unit (201) can control the operation of the device by executing software, programs, or instructions stored in the storage unit (203).

[0059] The communication unit (202) can perform the function of transmitting and receiving signals with other devices. The communication unit (202) performs wired communication or wireless communication and can process signals according to the control of the control unit (201). The communication unit (202) may include at least one network interface. For example, the communication unit (202) may include an RF circuit, an antenna, etc. for wireless communication, or a connection terminal, a modem, a driver module, etc. for wired communication. For example, the communication unit (202) may support at least one of various communication protocols, such as cellular communication like LTE or 5G, short-range wireless communication like WiFi or Bluetooth, and short-range wired communication like Ethernet.

[0060] The storage unit (203) can store data used in the device, software, programs, and instructions for the operation of the device. Additionally, the storage unit (203) can provide stored data under the control of the control unit (201). Additionally, the storage unit (203) can store applications, drivers, etc. to be driven by the control unit (201). The storage unit (203) may include at least one memory. For example, the storage unit (203) may include RAM such as DRAM, SRAM, etc.; ROM; EEPROM; HDD; SSD; flash storage means, etc.

[0061] Although not shown in FIG. 2, the device may further include at least one of a power supply (e.g., battery, power supply, etc.), a display device (e.g., display, etc.), an input device (e.g., touch pad, etc.), and an output device, depending on the type of device.

[0062] FIG. 3 illustrates a flowchart of operations for performing a prediction of the risk of construction site complaints according to an embodiment of the present invention. FIG. 3 illustrates an example where a construction site complaint prediction service is provided in the form of a platform. That is, FIG. 3 may show operations performed by a server device (102). However, as previously described, the construction site complaint prediction service may be provided by a program or application installed on a user device (101). In this case, FIG. 3 may show the operation of the user device (101), and a person skilled in the art will clearly understand that some of the descriptions below may be implemented in a modified form without departing from the essential characteristics of the present invention.

[0063] Referring to FIG. 3, the server device (102) can acquire geographic information system-based information of the construction site, site shooting information, process management information, environmental measurement information, and external weather information (e.g., S301). That is, the server device (102) can receive multimodal data necessary to predict the risk of on-site complaints. As described below, the server device (102) can integrate and analyze heterogeneous data.

[0064] For example, the server device (102) can transmit a request for information provision to a device that provides geographic information system-based information, site shooting information, process management information, environmental measurement information, and external weather information of the construction site, and can receive said information (however, the transmission of the request for information provision may be omitted). The geographic information system-based information, site shooting information, process management information, environmental measurement information, and external weather information may include information from the same point in time or substantially the same point in time (e.g., the same date or daytime of the same date) (e.g., the server device (102) receives site shooting information and environmental measurement information for the same point in time). The server device (102) can receive geographic information system-based information, site shooting information, process management information, environmental measurement information, and external weather information from a user device (101), a third device (not shown; e.g., geographic information system data server, process management system data server, Internet of Things (IoT) sensor, noise sensor, dust sensor, weather sensor, weather forecast data server, meteorological agency data server, etc.), etc.

[0065] In one example, a dedicated platform or website designed for the exchange of information may be used. The server device (102) provides an interface for uploading geographic information system-based information, site photography information, process management information, environmental measurement information, and external weather information of the construction site online, and can receive said information through said interface. In another example, the server device (102) may receive some of the geographic information system-based information, site photography information, process management information, environmental measurement information, and external weather information of the construction site from a third device. For example, the server device (102) may send a request for information provision to a device providing geographic information system-based information of the construction site (e.g., geographic information system data server) and receive geographic information system-based information. The server device (102) may send a request for information provision to a device providing process management information of the construction site (e.g., process management system data server) and receive process management information. The server device (102) can send a request for information provision to a device that provides environmental measurement information for a construction site (e.g., Internet of Things (IoT) sensor, noise sensor, dust sensor, weather sensor) and receive environmental measurement information. The server device (102) can send a request for information provision to a device that provides external weather information for a construction site (e.g., weather forecast data server, weather agency data server) and receive environmental measurement information.

[0066] Specific examples of data collection methods are as follows.

[0067] (1) For geographic information system-based information, the collection sources are the National Spatial Information Portal, local government GIS DBs, and construction companies, the collection cycle is static data (updated at least once a year) and dynamic data (monthly), and the processing format may be standard formats such as WMS (Web Map Service) and GeoJSON.

[0068] (2) For on-site filming information, the collection method is ① fixed CCTV (installed on access roads and major work sections, 10fps or higher), ② drone filming (once a week), ③ submission by construction manager's smartphone (at least once a day), the filming resolution is 1080p or higher, night filming in low-light environments is possible, and the transmission cycle can be real-time or batch processing at 1-hour intervals.

[0069] (3) For process management information, the collection source is automatic integration with an ERP system (Primavera, MS Project, etc.), data synchronization is automatic integration once a day, and manual input options are supported, and the supported formats may be XML, CSV, or API integration.

[0070] (4) For environmental measurement information, noise is measured every 10 minutes by IoT sensors or professional measuring instruments (±2dB accuracy), dust is measured every 10 minutes by PM10 and PM2.5 measuring sensors (±15% error range), and weather is measured every 10 minutes by temperature, humidity, wind direction / wind speed sensors, and the sensor installation locations may be at least 4 locations at the site boundary and in the direction of the residential boundary.

[0071] (5) For external weather information, the source of collection is the Korea Meteorological Administration API and weather forecast service providers, the update cycle is real-time (updated every 3 hours), and the forecast range is a long-term forecast for 7 days from the current time.

[0072] The server device (102) synchronizes the above-described information based on timestamps, and can process missing values ​​using linear interpolation or forward fill. If the missing value rate is 20% or higher, the time zone prediction can be provided along with a reliability indicator.

[0073] Geographic Information System-based information may refer to spatial information regarding a construction site provided by a Geographic Information System (GIS). Geographic Information System-based information may include spatial information linked to the location of the construction site. Geographic Information System-based information may include at least one, two, or three of surrounding population density, location information of surrounding facilities, information on past complaints, surrounding traffic information, and topographic information. Geographic Information System-based information includes location information of surrounding facilities and information on past complaints, and may further include at least one of surrounding population density, surrounding traffic information, and topographic information. Geographic Information System-based information may be visualized by being superimposed on a Geographic Information System map (i.e., in the form of a layer). Geographic Information System-based information may include text data, continuous data, categorical data, image data, etc.

[0074] For example, surrounding population density may include the population density (or population density) within a specified radius (e.g., 500 meters) of the construction site. Nearby facility location information may include facility location information within a specified radius (e.g., 500 meters) of the construction site. In this case, facilities may include educational facilities (e.g., schools), residential facilities (e.g., housing complexes), medical facilities (e.g., hospitals), etc. Nearby facility location information may further include the distance between said facilities and the construction site. Information on past complaints may refer to information regarding construction site complaints previously raised by residents in the area where the construction site is located. Information on past complaints may include the total number of complaints and types of complaints during a specified period (e.g., the past 6 months). Nearby traffic information may refer to traffic information within a specified radius (e.g., 500 meters) of the construction site. In this case, surrounding traffic information may include road locations, road traffic volume, public transportation routes (e.g., bus routes, railway routes), etc. Topographic information may refer to the physical and natural form of an area within a predetermined radius (e.g., 500 meters) centered on the construction site. Topographic information may include mountains, plains, rivers, valleys, coasts, etc.

[0075] On-site footage information may refer to information obtained by photographing the work performed at a construction site. For example, on-site footage information may include images captured by a user (e.g., a construction site manager) of the actual work site using a camera or similar device. On-site footage information may be images captured by a user observing the overall view of the work site, work equipment, personnel, etc. On-site footage information may include multiple on-site footage images. On-site footage information may include image data.

[0076] Project management information may refer to information provided by an Enterprise Resource Planning (ERP) system for managing work plans and progress status. For example, project management information may be automatically collected from an ERP system operated by a construction company or a construction management company. Project management information may include at least one, two, three, or four of work type information, project progress rate, scheduled work time, scheduled work location, scheduled equipment information, and scheduled personnel information. Project management information includes work type information, scheduled equipment information, and scheduled personnel information, and may further include at least one of project progress rate, scheduled work time, and scheduled work location. Project management information may include text data, continuous data, categorical data, etc.

[0077] For example, work type information may refer to information regarding the types of work being carried out at a construction site. Work type information may include at least one of excavation work, reinforced concrete work, finishing work, temporary works, steel frame work, masonry work, plastering work, stonework, interior finishing work, and window and door work. Work type information may be classified into major categories (e.g., civil engineering works, building construction), medium categories (e.g., excavation work, reinforced concrete work, steel frame work, masonry work, finishing work, installation work, demolition work, etc.), and detailed categories (10 to 20 specific work types per medium category). The progress rate may refer to the ratio of actual completed work to the total process. The progress rate may include the daily progress rate (%). Scheduled work time may refer to the time when work is planned to be carried out (e.g., from 9:00 AM to 3:00 PM, 3 hours, etc.). Scheduled work time may include time intervals, duration, etc. Scheduled work time may be expressed in units of hours, minutes, etc. Scheduled work time may reflect work interruptions caused by holidays, inclement weather, etc. For example, the process progress rate and scheduled work time can be used to calculate work intensity variables (daily completion rate / planned rate ratio). Scheduled work location may refer to the location where work is planned to proceed. Scheduled work location may include spatial coordinates (e.g., GPS coordinates, GIS coordinates, etc.). Scheduled equipment information may refer to information regarding work equipment (e.g., heavy equipment) planned to be deployed for the work. Scheduled equipment information may include the type of equipment (e.g., crane (by tonnage), excavator, backhoe, dump truck, concrete pump truck, cutter, welder, etc.), the number of pieces of equipment, equipment operating hours (e.g., scheduled operating time slot, scheduled deployment date), equipment deployment location, equipment operating range (e.g., work radius), noise contribution (e.g., a database of standard noise values ​​by equipment type; e.g., crane 95 dB, backhoe 88 dB), dust contribution, etc. Scheduled manpower information may refer to information regarding the manpower planned to be deployed for the work.Planned workforce information may include the number of personnel (e.g., average daily workforce), deployment locations, safety equipment, and working conditions (e.g., day shift, night shift, weekly rest day plans, etc.).

[0078] Environmental measurement information may refer to environmental information (e.g., noise, dust, weather) of a construction site measured by sensors installed at or near the construction site (e.g., IoT sensors). Weather data measured by IoT sensors or weather sensors may be distinguished from external weather information. Environmental measurement information may include at least one of noise data, dust data, and weather data of the construction site. Environmental measurement information may include text data, continuous data, categorical data, etc. Environmental measurement information may include time-series data.

[0079] For example, noise data may refer to noise values ​​measured by sensors such as Internet of Things (IoT) sensors and noise sensors. Noise values ​​can be expressed in decibels (dB). Dust data may refer to dust values ​​measured by sensors such as IoT sensors and dust sensors. Dust values ​​may refer to a numerical value representing the amount of dust contained in a certain volume of air. Dust values ​​may be measured for PM10, PM2.5, total suspended particulate matter, etc. Dust values ​​are expressed in μg / m³ 3 It can be expressed in units. Weather data may refer to data measured from sensors such as Internet of Things sensors and weather sensors. Weather data may include at least one of wind direction, wind speed, humidity, atmospheric pressure, temperature, and precipitation. Weather data may be distinguished from external weather information received from external data servers, such as the Korea Meteorological Administration.

[0080] External weather information may refer to information on the weather conditions of a construction site (or nearby area) observed or predicted by a weather agency, an external weather service, etc. (i.e., an external agency). External weather information may include information on the weather conditions of the construction site observed or predicted at the time when the server device (102) acquires the external weather information or at substantially the same time (e.g., the same date). External weather information may include information on the weather conditions of the construction site predicted at a time after a predetermined period (e.g., 12 hours, etc.) from the said time. External weather information may include actual measurement data or forecast data provided by a weather agency, an external weather service, etc. External weather information may include text data, continuous data, categorical data, etc. For example, external weather information may include at least one, two, three, or four of wind direction, wind speed, humidity, atmospheric pressure, temperature, solar radiation, and precipitation. External weather information may include at least one, two, or three of wind direction, wind speed, humidity, atmospheric pressure, and precipitation.

[0081] The server device (102) can extract geographic information system-based feature information based on geographic information system-based information using a rule-based model or a trained artificial intelligence model (e.g., S302). The geographic information system-based feature information may refer to information in the form of vectors provided as input data to the model when predicting the risk of on-site civil complaints. The geographic information system-based feature information may include vectors corresponding to geographic information system-based information (e.g., surrounding population density, location information of surrounding facilities, information on past civil complaints, surrounding traffic information, topographic information).

[0082] The server device (102) can preprocess the acquired geographic information system-based information. For example, preprocessing may include coordinate system unification (e.g., converting GPS coordinates of a construction site to a Universal Transverse Mercator (UTM) coordinate system), layer-based consistency assurance, vector data structuring, raster data structuring, spatial buffering (e.g., extracting only information on surrounding facilities within a 500m radius centered on the site), standardization, and normalization (e.g., normalizing the scale of each feature to a range of [0,1]).

[0083] For example, a server device (102) can extract geographic information system-based feature information using a rule-based model for extracting geographic information system-based feature information. The rule-based model may be pre-configured and may receive geographic information system-based information as input and output geographic information system-based feature information. In one example, the extraction of geographic information system-based feature information may be performed based on a pre-defined mapping table or algorithm. In another example, geographic information system-based feature information may be extracted based on values ​​calculated by rules including weights and biases for geographic information system-based information. For example, the surrounding population density feature is calculated as 'population within a radius of 500m / area of ​​the corresponding region' and may be expressed as result grades of low (<100 people / ㎢), medium (100-500 people / ㎢), and high (>500 people / ㎢). The surrounding facility distance feature calculates the shortest distance for each type of facility (school, hospital, residential area) and can apply an inverse function weight to the distance (e.g., a weight of 0.9 is applied when a school exists within 300m). The past complaint occurrence density feature is calculated as 'number of complaints in the past 6 months / area' and can be classified into categories by complaint type (e.g., noise, dust, traffic, other). For example, the server device (102) can extract geographic information system-based feature information using a trained artificial intelligence model for extracting geographic information system-based feature information based on geographic information system-based information of the construction site. The trained artificial intelligence model may include a convolutional neural network (CNN), a graph neural network (GNN), a transformer, etc. That is, the trained artificial intelligence model may be composed of a known spatially specialized neural network. For example, the trained artificial intelligence model may be a model trained to extract geographic information system-based feature information based on geographic information system-based information, or it may be built by fine-tuning a known artificial intelligence model as a backbone.For example, the trained artificial intelligence model can generate geographic information system-based feature information using multiple input features (e.g., distance-based features, spatial density features, layer-by-layer integration features, etc.) generated based on geographic information system-based information, hidden layers, activation functions, embedding generation layers, etc. For example, using a pre-trained ResNet50 and custom FC layers (4 layers, 256→128→64→32), a 32-dimensional embedding vector can be output from GIS layer images (each 30x30 pixels, 8 channels) (i.e., input) within a radius of 500m (in this case, the training data is a dataset of 1,200 construction sites from 16 municipalities). The server device (102) can extract field work feature information based on field shooting information using the trained artificial intelligence model (S303). For example, step S303 can be performed after step S301. Step S303 can be performed in parallel with Step S302. For example, in Step S302, the server device (102) can simultaneously extract geographic information system-based feature information and field work feature information in parallel.

[0084] On-site photographic feature information may refer to vector-type information provided as input data to a model when predicting the risk of on-site complaints. On-site photographic feature information may include vectors corresponding to the contents included in the on-site photographic information (e.g., view of the work site, work equipment, personnel, etc.). On-site photographic feature information may include at least one, two, or three of on-site work type feature information, on-site work location feature information, on-site equipment feature information, and on-site personnel feature information. On-site photographic feature information includes on-site work type feature information and on-site equipment feature information, and may further include at least one of on-site work location feature information and on-site personnel feature information.

[0085] For example, on-site work type feature information may refer to a vector corresponding to the type of work performed at the actual work site. In this case, the on-site work types may include scaffolding installation, crane operation, pouring, excavation, welding, painting, rebar placement, concrete pouring, dismantling / demolition, masonry, plastering, finishing work, etc. (i.e., feature information corresponding to the relevant on-site work type is extracted). As an example, the work classification accuracy may be 95% or higher. On-site work location feature information may refer to a vector corresponding to the location of the site where the actual work is performed. On-site equipment feature information may refer to a vector corresponding to information about the work equipment deployed at the actual work site. In this case, the information about the work equipment may include the type of equipment (e.g., 18 major types of heavy equipment such as cranes, backhoes, excavators, dump trucks, and concrete pump trucks), the number of pieces of equipment, the operating time of the equipment, the location where the equipment is deployed, whether the equipment is operating (e.g., operating, idle), and the operating range of the equipment. For example, on-site work location feature information can be extracted through 3D position transformation utilizing camera intrinsic parameters of on-site captured images and on-site reference point GPS coordinates (accuracy: within the range of 0.5m to 1m). On-site personnel feature information may refer to a vector corresponding to information about personnel deployed to the actual work site. In this case, information about personnel may include the number of personnel, the deployment location of personnel, and personnel safety equipment (e.g., wearing rate). For example, the number of personnel has an error range of ±10% and can be corrected using process records during indoor work where real-time detection is not possible. As another example, the safety equipment wearing rate has an accuracy of 85% or higher, and the wearing of helmets, safety clothing, safety shoes, etc., can be detected.

[0086] The server device (102) can preprocess the acquired field shooting information. For example, preprocessing may include image resizing, resolution correction, color normalization, noise removal, brightness adjustment, image alignment, standardization, and normalization (e.g., color normalization (e.g., normalization based on ImageNet statistics)). As an example, the size of the input image may be standardized to 1024x768px, and histogram flattening (CLAHE) may be performed on the night shooting image (i.e., low light correction).

[0087] For example, the server device (102) can extract field work feature information using a trained artificial intelligence model for extracting field work feature information based on field shooting information of a construction site. The trained artificial intelligence model may include a convolutional neural network (CNN), a vision transformer, a backbone network, etc. That is, the trained artificial intelligence model may include a standard computer vision model (ResNet, YOLO v8-based fine-tuning, etc.). For example, the trained artificial intelligence model may be a model trained to extract field work feature information based on field shooting information, or it may be built by fine-tuning a known artificial intelligence model as a backbone. For example, the trained artificial intelligence model may produce field work feature information using a feature map, hidden layer, activation function, fully connected layer, etc. generated based on field shooting information. For example, using a YOLO v8n (nano) model, equipment (18 types), work type (12 types), and personnel can be detected and bounding boxes (coordinates), classes, and confidence scores can be output. Subsequently, embedding vectors are extracted for each detected object (512 dimensions), bounding box positions are normalized (image coordinates → 0~1), and 14 features such as the number of equipment, number of personnel, and work type ratio can be collected. As another example, images captured three times a day (morning, noon, afternoon) are input, temporal patterns are learned through an LSTM encoder (3-day window), and a time-series weighted feature vector (128 dimensions) can be output.

[0088] The server device (102) can predict the risk of construction site complaints at a predetermined time point based on geographic information system-based feature information, field work feature information, process management information, environmental measurement information and external weather information using a rule-based model or a trained artificial intelligence model (e.g., S304).

[0089] A predetermined point in time may include a point in time after the point in time when the information described above in step S301 is obtained (i.e., the point in time when step S301 is performed). For example, a predetermined point in time may refer to a point in time when a predetermined amount of time (e.g., 6 hours, 12 hours, 18 hours, 24 hours, 30 hours, 36 hours, 48 ​​hours, 3 days, 7 days, 10 days, etc.) has elapsed since the information was obtained. The predetermined amount of time may be set in advance based on the type of work, the judgment of the construction site manager, etc. A predetermined point in time may include the current point in time (e.g., the point in time when the information described above in step S301 is obtained).

[0090] The risk of construction site complaints at a given point in time may refer to the degree to which construction site complaints are likely to occur at that time. The risk of construction site complaints may include at least one of a quantitative indicator and a categorized indicator regarding the degree to which construction site complaints are likely to occur. The risk of construction site complaints may be expressed as a score, probability, grade (e.g., safe, low risk, medium risk, moderate risk, high risk, extremely high risk), or a combination thereof. The risk of construction site complaints may be based on criteria predefined by experts.

[0091] The server device (102) can preprocess geographic information system-based feature information, field work feature information, process management information, environmental measurement information, and external weather information. For example, preprocessing may include vectorization, standardization, normalization, etc.

[0092] For example, the server device (102) can predict the risk of construction site complaints at a predetermined point in time using a rule-based model for predicting the risk of construction site complaints. For example, the rule-based model may be pre-configured, and the rules may include weights and biases for geographic information system-based feature information, site work feature information, process management information, environmental measurement information, and external weather information. The risk of construction site complaints at a predetermined point in time may be determined based on the value calculated by the rules. For example, the rule-based model may include a regression model or a classification model depending on the type of risk of construction site complaints at a predetermined point in time, and may receive geographic information system-based feature information, site work feature information, process management information, environmental measurement information, and external weather information as inputs to output the risk of construction site complaints at a predetermined point in time.

[0093] The server device (102) can predict the risk of construction site complaints using a trained artificial intelligence model for predicting the risk of construction site complaints. The trained artificial intelligence model may include a regression model or a classification model depending on the type of risk of construction site complaints. For example, the trained artificial intelligence model may be a model trained to output the risk of construction site complaints based on geographic information system-based feature information, site work feature information, process management information, environmental measurement information, and external weather information, and may be built by fine-tuning a known artificial intelligence model as a backbone. For example, the trained artificial intelligence model may be trained using geographic information system-based feature information, site work feature information, process management information, environmental measurement information, and external weather information labeled with the risk of construction site complaints as a training data set. For example, the trained AI model can output a predicted value for the risk of construction site complaints by utilizing feature information generated based on geographic information system-based feature information, site work feature information, process management information, environmental measurement information, and external weather information, as well as classification layers (e.g., fully connected layers, etc.) and activation functions (e.g., sigmoid functions, etc.). The risk of construction site complaints can be determined based on this predicted value.

[0094] For example, a trained AI model for predicting the risk of construction site complaints may be input with GIS feature vectors (32 dimensions), on-site imaging feature vectors (128 dimensions), process management information (15 dimensions) (e.g., construction type category (one-hot), progress rate), environmental measurement information (5 dimensions) (e.g., noise (current, 1-hour moving average), dust (current, 1-hour moving average), wind direction (category)), and external weather information (5 dimensions) (e.g., wind speed, humidity, temperature, precipitation forecast, time zone). The trained AI model includes a Gradient Boosting Machine algorithm, has a tree depth of 7 to 9, has 500 to 1000 iterations, and can provide interpretability based on feature importance. The prediction confidence score (based on similarity with surrounding training data) may be 0 to 100%. For example, the output construction site complaint risk may include a quantitative score, a qualitative grade, and detailed risk levels by complaint type. For example, regarding quantitative scores, the base score may be 0 to 100 points (higher indicates higher risk), and a confidence interval may be presented as a score range with an 85–95% confidence level (e.g., 73±8 points (90% confidence)). Qualitative grades may include safe (0–20 points) (only on-site management required), low risk (21–40 points) (monitoring recommended), medium risk (41–60 points) (precautionary measures required), medium-high risk (61–80 points) (reinforcement measures required), and high risk (81–100 points) (immediate response or work stoppage recommended). Detailed risk levels by complaint type may include noise complaint risk (%, contribution), dust complaint risk (%, contribution), and other complaint risk (traffic, vibration, %). For example, the specified time points may include 6 hours, 12 hours, 24 hours, and 3 days after the information is acquired (i.e., the prediction range is in the form of a roadmap). The update cycle can be automatically updated when new data is entered (average response time is less than 5 minutes). Trends can be analyzed by visualizing the risk increase / decrease trend.For example, regarding the validation results of the model, AUC-ROC: 0.88 (distinguishing between the occurrence and non-occurrence of complaints), Accuracy: 0.84, and Recall (high-risk detection): 0.91 (provided the training dataset consists of 16 local governments, 3,200 sites, and 2 years of historical data).

[0095] This paper briefly explains the correlation between Geographic Information System-based feature information, field work feature information, process management information, environmental measurement information, and external weather information and the risk of construction site complaints.

[0096] i) Geographic Information System-based feature information may include external environmental factors that are highly likely to generate complaints at construction sites. For example, if a construction site is densely populated and surrounded by sensitive facilities (e.g., schools, hospitals, residential areas), complaints regarding noise and dust damage are likely to be filed. If there have been past instances of complaints, similar cases may occur again. If transportation networks and accessibility are good, the number of people traveling to and from the area increases, leading to more people experiencing inconvenience, and traffic congestion may be exacerbated by construction vehicles. Surrounding topography can alter the propagation and diffusion paths of noise and dust. For instance, noise may refract or resonate in mountainous areas and valleys, while forests can shield or absorb noise and dust. Therefore, Geographic Information System-based feature information can be correlated with the risk of complaints at construction sites.

[0097] ii) Information on site work characteristics may include actual work site factors (e.g., corresponding vectors) that directly cause noise and dust. The amount of noise and dust generated may vary depending on the type of work at the site (e.g., operations such as excavation, cranes, and concrete pouring carry a high risk of noise and dust complaints). The amount of noise and dust may also vary depending on the type, number, operating time, and operating range of work equipment deployed at the site (e.g., cutting machines generate high-frequency noise and localized dust. The more work equipment there is, the longer the operating time, and the wider the operating range, the higher the amount of noise and dust). Since a larger workforce at the site performs more work, the amount of noise and dust may increase. Therefore, information on site work characteristics may be associated with the risk of construction site complaints.

[0098] iii) Process management information may include details regarding work that generates noise and dust. The amount of noise and dust generated may vary depending on the type of construction (e.g., excavation work involves the deployment of large excavators, resulting in low-frequency noise and significant airborne dust; finishing work is primarily indoors, so noise and dust generation are minimal; and steel frame work involves significant noise from cutting, welding, and impact). The number of required equipment and personnel may vary depending on the stage of progress. As mentioned above, the amount of noise and dust may vary depending on the equipment and personnel scheduled to be deployed to the work site. Therefore, process management information may be associated with the risk of complaints regarding the construction site.

[0099] iv) Environmental measurement information may include data on noise and dust that cause complaints. For example, if the amount of noise and dust measured by sensors exceeds standard limits or increases rapidly, the inconvenience to residents may intensify. High noise levels (e.g., 85 dB or higher) and high dust levels (e.g., PM10 < 150 µg / m³) 3There is a high likelihood that complaints regarding construction sites may be raised. Weather conditions can affect the extent to which noise and dust are dispersed or suppressed. For example, strong winds can cause noise and dust to spread further. Dry weather can intensify dust dispersion. If weather conditions are poor, daily inconveniences may be exacerbated, increasing the risk of complaints regarding construction sites. Therefore, environmental measurement data and external weather information may be correlated with the risk of complaints regarding construction sites.

[0100] The server device (102) can provide the risk of construction site complaints for a predetermined time point to the user device (101). For example, the server device (102) can predict the risk of construction site complaints in units of a predetermined time (e.g., 2 hours, 6 hours) and provide the determined risk of construction site complaints to the user device (101). The server device (102) can predict the risk of construction site complaints for a predetermined time point by type of complaint (e.g., noise, dust, etc.) and provide the predicted risk of construction site complaints to the user device (101). The server device (102) can visualize the risk of construction site complaints for a predetermined time point in the form of a heatmap on a Geographic Information System (GIS) map and provide it to the user. For example, the mapping of the risk of construction site complaints can be performed by type of construction work to build a database.

[0101] In some embodiments, the server device (102) may determine a guide for response measures based on at least one of the construction site complaint risk level and geographic information system-based feature information, site work feature information, process management information, environmental measurement information, and external weather information using a rule-based model or a trained artificial intelligence model. In this case, the guide for response measures may include a guide for response measures to prevent complaints.

[0102] Guides on response measures can be used as materials to determine measures to minimize construction site complaints at construction sites. Guides on response measures may include at least one of work suspension, reduction of work intensity (e.g., reduction of work intensity by 50%), noise reduction measures (e.g., installation of sound barriers, reinforcement of sound barriers, addition of sound barriers), dust suppression measures (e.g., watering, spraying of dust suppressants, installation of windbreaks), construction site environmental inspection, and adjustment of work hours.

[0103] For example, the server device (102) can determine a guide for response measures using a rule-based model for determining a guide for response measures. The server device (102) can determine a guide for response measures using a trained artificial intelligence model for determining a guide for response measures. The trained artificial intelligence model may include a generative model. For example, the trained artificial intelligence model may extract an embedding vector from at least one of the input construction site complaint risk level, geographic information system-based feature information, site work feature information, process management information, environmental measurement information, and external weather information, and decode the extracted embedding vector to generate text regarding a guide for response measures. Since details regarding the rule-based model, the trained artificial intelligence model, etc., can be clearly understood by referring to the above description, a more detailed explanation is omitted.

[0104] For example, Tables 1 through 5 below are examples of basic action templates by risk level (i.e., examples of guides for response measures).

[0105] Safety Level (0~20 points): Routine site management • Environmental monitoring at least once a week • Broadcasting standard announcements (twice a day) • Preparation for responding to resident complaints (designation of person in charge)

[0106] Low-risk stage (21–40 points) enhanced monitoring and mild prevention ㆍ Review of daily environmental measurements ㆍ Basic noise reduction (inspection of sound barrier conditions) ㆍ Dust suppression watering (intensified during high temperature / dry conditions) ㆍ Advance notice to residents (notification of schedule for the next 2 weeks)

[0107] Medium Risk Level (41–60 points): Proactive Preemptive Response • Real-time environmental monitoring (integrated with sensors) • Enhanced noise reduction (addition of sound barriers or increase in height) • Enhanced dust suppression (increased watering frequency, application of dust suppressants) • Reduction or prohibition of night work • Holding of resident information sessions (announcement of scheduled dates) • On-site presence of construction company safety managers • Notification of local government forecasts

[0108] Medium-Risk Level (61–80 points): Intensive Management and Work Coordination • 24 / 7 real-time environmental monitoring + alarm settings • 50% reduction in work intensity (limited daily work hours) • Maximized noise reduction (prohibition of perforated equipment) • Enhanced dust suppression (addition of dust nets, enclosed construction) • Suspension of work during severe weather (strong winds > 10 m / s) • Preparation for local government inspections (scheduled site visit by 담당 officials) • Deployment of dedicated personnel for resident complaint consultation

[0109] High-risk level (81–100 points): Immediate action or suspension of work • Recommendation to immediately halt work on the relevant construction phase or reduce work intensity by 80% • Elimination of all sources of noise and dust (perform only non-essential work) • Scheduled emergency inspection by local government • On-site inspection by construction company management • Formation of a special task force for responding to complaints • 24–48-hour risk tracking and management

[0110] As another example, dynamic guide adjustments may be performed. For instance, structural measures may be recommended when the same type of work is repeated (e.g., repeated noise during concrete pouring → "additional installation of high-frequency sound-absorbing material"), adjustments to the time period may be recommended when there is a concentration during a specific time period (e.g., dust concentration between 2 PM and 4 PM → "restrict outdoor work in the afternoon"), weather response recommendations may be recommended when adverse weather factors increase (e.g., forecast of increased wind speed → "pre-spraying of dust suppressants"), and distance-based measures may be strengthened when working in close proximity to residential areas (e.g., 20 or more houses within 300m → "prohibition of night work").

[0111] As another example, guides for site managers (construction companies) and public officials handling complaints (local governments) can be differentiated. The guide for site managers may include immediately actionable work adjustment items, proposals for changing equipment deployment plans, options for rescheduling the process, and cost impact estimations. The guide for public officials handling complaints may include proposals for inspection schedules, complaint response strategies, guidelines for administrative measures upon detection of violations, and materials for explanatory sessions with local residents.

[0112] For example, necessary response measures may be strengthened as the risk of complaints at a construction site increases. If the risk of complaints is high (e.g., high risk), work must be suspended and noise reduction and dust suppression measures reinforced; however, if the risk is not high (e.g., medium risk), environmental inspections or monitoring of the construction site may be performed. As mentioned above, the risk of complaints at a construction site varies depending on geographic information system-based feature information, site work feature information, process management information, environmental measurement information, and external weather information, and necessary response measures may be strengthened if there are factors that increase the risk. Furthermore, the necessary response measures may vary depending on spatial information around the construction site, work plans, and the actual work being performed. For example, if noise increases sharply due to the use of large excavators, the installation of additional sound barriers may be recommended. As another example, if there is a school near the construction site and excavation work is planned, it may be recommended to postpone work hours until after school dismissal time. Therefore, the risk of complaints at a construction site and the aforementioned information can be linked to guidelines for response measures.

[0113] In some embodiments, the server device (102) may provide a guide to the construction site complaint risk level and response measures to a user (e.g., construction site manager, civil servant in charge of complaints) when the construction site complaint risk level is greater than a predetermined threshold. When the construction site complaint risk level is greater than a predetermined threshold (e.g., 50 points, 50%, medium risk, etc.), the server device (102) may provide the construction site complaint risk level for a predetermined time to the user device (101). In this case, the server device (102) may further provide the user device (101) with a guide to response measures, the location of the construction site, etc.

[0114] In some embodiments, the server device (102) may determine whether a construction site is a priority inspection site based on at least one of the construction site complaint risk level, geographic information system-based feature information, site work feature information, and process management information, using a rule-based model or a trained artificial intelligence model. Whether a construction site is a priority inspection site may mean whether a construction site requires priority inspection. For example, if the construction site complaint risk level is high, priority inspection may be required to prevent complaints. Priority inspection may be required if the population density around the construction site is high, sensitive facilities are nearby, or dangerous processes (e.g., heavy equipment operations) are planned or performed, as there is a high risk of complaints arising. Details regarding the rule-based model, trained artificial intelligence model, etc., for determining whether a construction site is a priority inspection site can be clearly understood by referring to the foregoing description, so a detailed explanation thereof is omitted.

[0115] In some embodiments, the construction site may include multiple construction sites. For example, a civil complaint officer manages multiple construction sites within their jurisdiction, so they can determine priorities when inspecting construction sites. In this case, steps S301 through S304 may be performed for each construction site, that is, the server device (102) can predict the risk of civil complaints at each construction site at a predetermined time.

[0116] For example, the server device (102) can acquire geographic information system-based information, site shooting information, process management information, environmental measurement information, and external weather information for each construction site. The server device (102) can extract geographic information system-based feature information of the construction site based on the geographic information system-based information by using a rule-based model or a trained artificial intelligence model. The server device (102) can extract site work feature information of the construction site based on the site shooting information by using a trained artificial intelligence model. The server device (102) can predict the risk of construction site complaints at a predetermined time point based on the geographic information system-based feature information, site work feature information, process management information, environmental measurement information, and external weather information by using a rule-based model or a trained artificial intelligence model.

[0117] The server device (102) can determine inspection priorities for each construction site based on at least one of the construction site complaint risk level, geographic information system-based feature information, site work feature information, and process management information, using a rule-based model or a trained artificial intelligence model for determining inspection priorities. The rule-based model and the trained artificial intelligence model may include a multi-objective optimization algorithm. For example, if the construction site complaint risk level is high, sensitive facilities are nearby, and hazardous processes are being performed, the inspection priority may be high. The server device (102) can visualize the construction site complaint risk level of each construction site in the form of a heatmap on a geographic information system (GIS) map and provide it to the user device (101). The server device (102) can provide the user with a list in which the inspection priorities for each construction site have been determined. The server device (102) can further provide the user with a guide on response measures for each construction site. As the inspection priorities for the construction sites are provided, the user (e.g., a civil servant in charge of complaints) can efficiently inspect the construction sites. Details regarding rule-based models, trained artificial intelligence models, etc., can be clearly understood by referring to the aforementioned content, so a detailed explanation is omitted.

[0118] For example, inspection priorities can be determined as follows. For example, Equation 1 below is an example of priority score calculation.

[0119] [Equation 1]

[0120] Priority Score = (0.4 × Complaint Risk) + (0.2 × Population Density) + (0.15 × Proximity to Sensitive Facilities) + (0.15 × Past Complaint History) + (0.1 × Safety Accident Risk)

[0121] In Equation 1, the civil complaint risk level: a predicted value of 0 to 100 points (weight 40%), population density: the population within a 500m radius (1000 people = 100 points), proximity to sensitive facilities: the distance to schools / hospitals (within 300m = 100 points), past civil complaint history: the number of civil complaints in the last 6 months, and safety accident risk level: may be a predicted risk level score.

[0122] As another example, the priority level may include the following Grade A to Grade E (that is, classified into 5 stages). Grade A (score 90~100): Highest priority inspection target → Daily inspection recommended, inspector: At least the person in charge of the division chief level; Grade B (score 70~89): Priority inspection target → Inspection recommended once every 3 days; Grade C (score 50~69): Regular inspection target → Regular inspection once a week; Grade D (score 30~49): Patrol inspection target → At least once a month; Grade E (score 0~29): Minimum management target → At least once a quarter.

[0123] As another example, Table 6 below is an example of determining inspection priority.

[0124] Jurisdiction Construction Site Complaint Risk Dashboard ㆍ Grade A Sites: 3 (Checkbox) ㆍ Grade B Sites: 7 ㆍ Grade C Sites: 18 ㆍ Sites Scheduled for Inspection Today: 2 (Site Name, Time) ㆍ Sites Scheduled for Inspection This Week: 9 (Displayed on GIS Map) ㆍ Complaints Received This Week: 12 (-25% compared to previous week)

[0125] Referring to Table 6 above, the construction site civil complaint prediction system (100) provides construction site civil complaint risk levels, response action guides, and inspection priorities in real time through a web dashboard or mobile application, and the dashboard may include geographic information system-based heatmap visualization, time-series risk trend graphs, and a function for querying detailed information for each construction site.

[0126] In addition, the server device (102) is mounted on the central management server of a local government or construction company for managing a plurality of construction sites, and the at least one processor may compare and analyze the predicted civil complaint risk level for each construction site to automatically determine the inspection priority.

[0127] In some embodiments, the server device (102) can predict the risk of construction site safety accidents at a predetermined time point and provide the risk of construction site complaints and the risk of construction site safety accidents to the user device (101). The server device (102) can predict the risk of construction site safety accidents after step S304. The server device (102) can predict the risk of construction site complaints and the risk of construction site safety accidents simultaneously and in parallel at step S304. The server device (102) can predict the risk of construction site safety accidents before step S304. The server device (102) can further provide the user with a guide on safety accident prevention response measures. Specific details regarding the risk of construction site safety accidents and the guide on safety accident prevention response measures will be described later. For example, the server device (102) can comprehensively analyze the risk of construction site complaints and the risk of construction site safety accidents and recommend special response measures for construction types and / or time periods where the risk of construction site complaints and the risk of construction site safety accidents are simultaneously high.

[0128] FIG. 4 illustrates a flowchart of operations for performing a prediction of the risk of safety accidents at a construction site according to an embodiment of the present invention. In this case, the construction site complaint prediction system (100) of FIG. 1 may be applied as a construction site safety accident prediction system by referring to FIG. 1, and the construction site complaint prediction service of FIG. 1 may be applied as a construction site safety accident prediction service by referring to FIG. 1 (i.e., the content related to FIG. 4 can be clearly understood by referring to the description of FIG. 1 to FIG. 3). FIG. 4 may be operations performed by a server device (102). However, as previously mentioned, the construction site complaint prediction service may be provided by a program or application installed on a user device (101) without a server device (102), and in this case, the server device (102) described below may refer to the user device (101). Content that overlaps with the content related to FIG. 3 described above may be omitted.

[0129] Referring to FIG. 4, the server device (102) can acquire geographic information system-based information of the construction site, site shooting information, process management information, environmental measurement information and external weather information (e.g., S401).

[0130] The server device (102) can extract geographic information system-based feature information based on geographic information system-based information using a rule-based model or a trained artificial intelligence model (e.g., S402).

[0131] The server device (102) can extract field work feature information based on field shooting information using a trained artificial intelligence model (S403).

[0132] In the case of S401 to S403, since the contents of the aforementioned S301 to S303 can be applied substantially identically, a detailed explanation thereof is omitted.

[0133] The server device (102) can predict the risk of safety accidents at a construction site at a predetermined time point based on geographic information system-based feature information, site work feature information, process management information, environmental measurement information, and external weather information by using a rule-based model or a trained artificial intelligence model for predicting the risk of safety accidents at a construction site (e.g., S404). Since the details regarding the predetermined time point, preprocessing, rule-based model, trained artificial intelligence model, etc. can be clearly understood by referring to the above-mentioned content, a detailed explanation thereof is omitted.

[0134] The risk of a construction site safety accident at a given point in time may refer to the degree to which a safety accident is likely to occur at the construction site at that time. The risk of a construction site safety accident may include at least one of a quantitative indicator and a categorized indicator regarding the degree to which a safety accident is likely to occur at the construction site. The risk of a construction site safety accident may be expressed as a score, probability, grade (e.g., safe, low risk, medium risk, moderate risk, high risk, extremely high risk), or a combination thereof. The risk of a construction site safety accident may be based on criteria predefined by experts.

[0135] For example, if the area surrounding a construction site is densely populated and traffic is heavy, the likelihood of accidents occurring during the movement of construction vehicles may be high. On terrain with steep slopes, there may be a high risk of equipment overturning. Safety accidents can easily occur when hazardous work is planned or performed. The likelihood of safety accidents at a construction site may increase as heavy equipment is deployed, the number of personnel involved increases, and safety equipment is inadequate. Noise and dust generated at the construction site can affect workers' cognitive abilities, visibility, and the operational stability of equipment. During adverse weather conditions (e.g., strong winds, heavy rain, heatwaves, cold waves), equipment safety, ground conditions, and workers' health may deteriorate. Therefore, Geographic Information System-based feature information, site work feature information, process management information, environmental measurement information, and external weather information can be correlated with the risk of safety accidents at a construction site.

[0136] The server device (102) can provide the risk of a construction site safety accident at a predetermined point in time to the user device (101). The server device (102) can predict the risk of a construction site safety accident at a predetermined point in time by accident type (e.g., fall, collision, fire, etc.) and provide the predicted risk of a construction site safety accident to the user device (101). Since the details regarding the provision of the risk of a construction site safety accident can be applied substantially identically to the above-mentioned details, a detailed explanation thereof is omitted.

[0137] In some embodiments, the server device (102) may determine a guide for response measures based on at least one of the construction site safety accident risk level and geographic information system-based feature information, site work feature information, process management information, environmental measurement information, and external weather information using a rule-based model or a trained artificial intelligence model. In this case, the guide for response measures may include a guide for safety accident prevention response measures.

[0138] Guides on response measures can be used as a reference for determining measures to minimize safety accidents at construction sites. Guides on response measures may include at least one of work suspension, reduction of work intensity (e.g., reduction of work intensity by 50%), waiting for weather recovery, inspection of safety equipment condition, extension of rest time, inspection of work equipment, inspection of construction site environment, and adjustment of work time.

[0139] For example, as the risk of safety accidents at a construction site increases, necessary response measures may be strengthened. If the risk of safety accidents at a construction site is high (e.g., high risk), work may be suspended or a standby period may be observed until weather conditions improve. If the risk of complaints at the construction site is not high (e.g., medium risk), safety equipment and work machinery may be maintained, and the surrounding environment may be inspected. The necessary response measures may vary depending on spatial information around the construction site, work plans, and the actual work being performed. For instance, in terrain with steep slopes, it may be recommended to install safety nets and inspect safety railings. As another example, if hazardous work is planned during inclement weather, it may be recommended to manage heavy equipment routes, inspect equipment safety conditions, and extend rest periods. Therefore, the risk of safety accidents at a construction site and the aforementioned information can be linked to guidelines for response measures.

[0140] In some embodiments, the server device (102) may provide a user (e.g., a construction site manager) with a guide on the risk of a construction site safety accident and safety accident prevention response measures when the risk of a construction site safety accident is greater than a predetermined threshold. When the risk of a construction site safety accident is greater than a predetermined threshold (e.g., 50 points, 50%, medium risk, etc.), the server device (102) may provide the user device (101) with the risk of a construction site safety accident at a predetermined point in time. In this case, the server device (102) may further provide the user device (101) with a guide on safety accident prevention response measures, the location of the construction site, etc.

[0141] In some embodiments, the server device (102) may determine whether a construction site is a priority inspection site based on at least one of the construction site safety accident risk level, geographic information system-based feature information, site work feature information, and process management information, using a rule-based model or a trained artificial intelligence model. Since the details regarding the determination of whether a construction site is a priority inspection site can be clearly understood by referring to the above description, a specific explanation thereof is omitted.

[0142] In some embodiments, the construction site may include multiple construction sites. For example, a construction site manager may manage multiple construction sites and determine priorities during construction site inspections. In this case, steps S401 through S404 may be performed for each construction site, that is, the server device (102) may predict the risk of construction site safety accidents at a predetermined time for each construction site. The server device (102) may determine the inspection priority for each construction site based on at least one of the construction site safety accident risk of each construction site, geographic information system-based feature information, site work feature information, and process management information, using a rule-based model or a trained artificial intelligence model for determining safety accident inspection priorities. Since the details regarding the determination of inspection priorities can be clearly understood by referring to the above description, a detailed explanation thereof is omitted.

[0143] In this specification, the designation of 'one embodiment' of the principles of the present invention and various variations of such expression means that specific features, structures, characteristics, etc., associated with this embodiment are included in at least one embodiment of the principles of the present invention. Accordingly, the expression 'in one embodiment' and any other variations disclosed throughout this specification do not necessarily refer to the same embodiment.

[0144] The implementation of the device and system according to the various embodiments of the present invention described above may be realized with digital electronic circuits, integrated circuits, ASICs (application specific integrated circuits), hardware, firmware, software, or a combination thereof.

[0145] The method according to the various embodiments of the present invention described above may be implemented as a computer program or mobile application and stored on a medium so as to be executed in combination with hardware. The steps of the method or algorithm described in relation to the embodiments of the present invention may be implemented directly in hardware, implemented as a software module executed by hardware, or implemented by a combination thereof. The software module may reside in RAM, ROM, EPROM, EEPROM, flash memory, hard disk, removable disk, CD-ROM, or any form of computer-readable recording medium well known in the art to which the present invention belongs. Additionally, the algorithm may be produced in the form of an installation file and provided in the form of an online download, and for this purpose, may be stored on a server accessible through an online software market.

[0146] All embodiments and conditional examples disclosed herein are intended to help those skilled in the art to understand the principles and concepts of the invention. Those skilled in the art will understand that the invention may be implemented in modified forms without departing from the essential nature of the invention. Therefore, the disclosed embodiments should be considered in an illustrative rather than a limiting sense. The scope of the invention is defined by the claims, not by the foregoing description, and all variations within the scope of the claims should be interpreted as being included in the invention.

Claims

Claim 1 A method for predicting construction site complaints, comprising: a step of obtaining geographic information system-based information, site shooting information, process management information, environmental measurement information, and external weather information of a construction site, executed by at least one processor; a step of extracting geographic information system-based feature information based on the geographic information system-based information using a rule-based model or a trained artificial intelligence model; a step of extracting site work feature information based on the site shooting information using a trained artificial intelligence model; and a step of predicting the risk of construction site complaints at a predetermined time point based on the geographic information system-based feature information, the site work feature information, the process management information, the environmental measurement information, and the external weather information using a rule-based model or a trained artificial intelligence model; wherein the site work feature information includes at least one of site work type feature information, site work location feature information, site equipment feature information, and site personnel feature information. Claim 2 A method for predicting construction site complaints according to claim 1, wherein the geographic information system-based information comprises at least one of surrounding population density, surrounding facility location information, past complaint occurrence information, surrounding traffic information, and topographic information. Claim 3 A method for predicting construction site complaints according to claim 1, wherein the process management information includes at least one of work type information, process progress rate, scheduled work time, scheduled work location, scheduled equipment information, and scheduled personnel information; the environmental measurement information includes at least one of noise data, dust data, and weather data of the construction site; and the external weather information includes at least one of wind direction, wind speed, humidity, atmospheric pressure, temperature, solar radiation, and precipitation. Claim 4 A method for predicting construction site complaints according to claim 1, further comprising the step of determining a guide for response measures based on at least one of the construction site complaint risk level, geographic information system-based feature information, site work feature information, process management information, environmental measurement information, and external weather information using a rule-based model or a trained artificial intelligence model. Claim 5 A method for predicting construction site complaints according to claim 4, further comprising the step of providing a user with a guide on the construction site complaint risk and the corresponding countermeasures when the construction site complaint risk is greater than a predetermined threshold. Claim 6 A method for predicting construction site complaints according to claim 1, further comprising the step of determining whether a construction site is a priority inspection site based on at least one of the construction site complaint risk level, geographic information system-based feature information, site work feature information, and process management information, using a rule-based model or a trained artificial intelligence model. Claim 7 A method for predicting construction site complaints, further comprising the step of predicting the risk of a construction site safety accident at a predetermined time point based on the geographic information system-based feature information, the site work feature information, the process management information, the environmental measurement information, and the external weather information, using a rule-based model or a trained artificial intelligence model according to claim 1. Claim 8 A method for predicting construction site complaints according to claim 7, further comprising the step of determining a guide for response measures based on at least one of the construction site safety accident risk level, geographic information system-based feature information, site work feature information, process management information, environmental measurement information, and external weather information using a rule-based model or a trained artificial intelligence model. Claim 9 A device for predicting construction site complaints, comprising: at least one memory; and at least one processor that executes instructions stored in the at least one memory; wherein the at least one processor acquires geographic information system-based information of a construction site, site shooting information, process management information, environmental measurement information, and external weather information; extracts geographic information system-based feature information based on the geographic information system-based information using a rule-based model or a trained artificial intelligence model; extracts site work feature information based on the site shooting information using a trained artificial intelligence model; and predicts the risk of construction site complaints at a predetermined time point based on the geographic information system-based feature information, the site work feature information, the process management information, the environmental measurement information, and the external weather information using a rule-based model or a trained artificial intelligence model; wherein the site work feature information includes at least one of site work type feature information, site work location feature information, site equipment feature information, and site personnel feature information. Claim 10 An application program stored on a recording medium to execute the method according to claim 1 when operated by at least one processor.

Citation Information

Patent Citations

  • System for smart safety integrated control and method for construction site risk assessment using the same

    KR1020250090958A

  • Method, apparatus and program for construction complaint management based on integrated environmental measurement module

    KR1020250125047A

  • Apparatus and method for predicting civil complaints using data-based spatial analysis

    KR102249524B1

  • A SYSTEM AND METHOD FOR CALCULATING THE COMPREHENSIVE CIVIL COMPLAINT INDEX AND CONTROLING THE CRITICAL CIVIL COMPLIAINT POINTS USING IoT TIME SERIES NOISE, VIBRATION DATA AND REGULATORY STANDARDS AND THE CIVIL COMPLAINT INFORMATION

    KR102394464B1