AI-POWERED, MULTI-PLATFORM (MOBILE AND WEB) OCCUPATIONAL HEALTH AND SAFETY RISK ANALYSIS AND AUDIT MANAGEMENT SYSTEM
Patent Information
- Authority / Receiving Office
- TR · TR
- Patent Type
- Applications
- Current Assignee / Owner
- TOKAT GAZİOSMANPAŞA ÜNİVERSİTESİ
- Filing Date
- 2026-06-04
- Publication Date
- 2026-06-22
Abstract
Description
AI-POWERED, MULTI-PLATFORM (MOBILE AND WEB) BUSINESS HEALTH AND SAFETY RISK ANALYSIS AND AUDIT MANAGEMENT SYSTEM Technological Field: The invention relates to images and inspections obtained from the field during occupational health and safety audits. data from mobile devices, server infrastructure, artificial intelligence analysis engine, central database and computer-based risk management through automated reporting modules. It is related to analysis and audit management systems. State of the Art: Risk assessments and field inspections carried out in the field of occupational health and safety, most of them The business still relies largely on manual processes. In these processes... Occupational safety specialists often take notes and photos of the risks they identify in the field. taking photos, recording relevant personnel or equipment information, and monitoring them. Subsequently, this information is transferred to different documents. This situation occurs during the audit. Converting the collected data into a report requires additional time and effort, also This can lead to data loss between field observation and the final report. One of the major problems in current practices is that data is transmitted across different environments. This is how photos are kept in the mobile device gallery, and descriptions are kept in personal notes. Personnel information is kept in separate tables, while risk assessment forms are in office programs. This allows for the integrated management of information related to the same audit. This makes things difficult, especially for a large number of employees, different work areas, or multiple roles. In businesses with branches, this decentralized structure makes it difficult to track past audits and assess risks. This makes comparative monitoring difficult. 1In known practices, the risk assessment process often relies on the expert's personal experience. and is shaped by interpretation. Different experts look at the same danger. the possibility of assigning different risk scores, suggesting different measures, or linking it to legislation It is possible that it is established inadequately. This situation affects occupational health and safety management. This makes standardization difficult and similar risks exist in different businesses or in the same business. This can lead to different evaluations in different departments of the business. Another problem encountered during field inspections is the effectiveness of photographic and visual data. The problem is that it cannot be processed in this way. The images taken during the inspection are mostly only... These images are used as supporting visuals added to the report, and are automatically extracted from these images. No risk assessment or classification is being carried out. Furthermore, it is high-dimensional. The images are then transferred to a computer, named, placed in the report, and Linking this to the relevant risk record is a cumbersome process. A significant portion of current software solutions are digital tools that only facilitate data entry. They operate in the form of forms or checklists. These types of systems use manual paper. while reducing its use, it analyzes data from the field and determines the nature of the risk. interpreting, suggesting appropriate measures, and directly converting these outputs into a report structure It does not offer an integrated technical infrastructure. Therefore, the user receives data in a digital environment. Even if it has been collected, it is still important in the risk assessment and reporting phase. They have to do a lot of manual work. Compliance with legislation in occupational health and safety processes also requires special attention. This is the subject. Legislative provisions depend on the type of risk, the work area, the equipment used, and This can vary depending on the business's field of activity. In known practices, the expert is relevant. They determine the legal information through their own knowledge or through further research. This This leads to a waste of time and is based on incomplete or outdated information. This increases the risk of an assessment being made. In conclusion, the known technique involves the digitalization of occupational health and safety audits. Although some solutions exist, visual and textual data obtained from the field... secure collection of data, analysis, and conversion of data into structured data. 2. Retrieving data, processing it into a central database, and automatically generating documents from this data. There is a continuing need for systems that manage processes in an integrated manner. This need especially data integrity, assessment standards, reporting speed, regulatory compliance, and This becomes evident in terms of field-center synchronization. Description of the invention: The main purpose of the invention is to utilize data obtained from the field during occupational health and safety inspections. time loss and errors resulting from disorganized, manual, and person-dependent processing. The aim is to eliminate the risk and reporting difficulties. In current practices, photography... Most of the processes include filming, note-taking, risk assessment, regulatory review, and report preparation. Since time is managed by independent tools, the process both slows down and Non-standard results may occur. The invention involves mobile data collection and artificial intelligence. Supported analysis, central database and automated document generation in a single technical system. by combining them under one umbrella, the audit process becomes faster, traceable, and integrated. It brings. One of the key advantages of the invention is the images and descriptions obtained during field inspections. The ability to directly link data to the reporting process. Mobile data by the user. Images captured and information entered via the device are stored in separate files or in different ways. instead of being stored in applications, it becomes a structured data record within the system. This is brought about. Thus, a risk identified in the field may later be forgotten, leading to mistakes. This reduces the likelihood of misclassification or incomplete reporting. The invention, thanks to its artificial intelligence analysis engine, allows the user to make observations based solely on their own judgment. standardized for the identified risk, regardless of whether it is based on an assessment. It generates recommendations. This structure allows different experts to view the same danger in different ways. It reduces differences in assessment that may arise from interpretation. Risk definition, Systemic support for matching proposed measures and legislation leads to greater consistency and... It provides a repeatable monitoring approach. Another advantage of the invention is that the analysis result from artificial intelligence is directly integrated into the system. The first step is to check the data format to ensure it is not accepted. Incomplete, corrupted, or outputs that do not conform to the system's data structure are not processed into the database. It can be filtered out first. This feature is one that can be encountered in AI-powered systems. By reducing the inconsistent output problem, reports become more reliable and the system becomes more stable. It contributes to his work. The system works with a central database, providing information on companies, personnel, training, health examinations, It allows audit history and risk records to be linked to each other. This ensures that not only is a single audit report generated, but also the business's overall performance is assessed. Past risk situations, measures taken, and recurring non-conformities can also be monitored. Thus, occupational health and safety management becomes a periodic and discontinuous reporting activity. It is transforming from a simple process into a continuously monitored data management process. The invention's automated document generation feature also includes a lengthy report after the audit. It significantly reduces the need for preparation. The system provides analyzed risk data and Risk analysis report, emergency plan, business using registered company / personnel information. Automatically generates outputs such as safety instructions or action tracking forms. It can create. In conclusion, the invention facilitates data collection and risk management in occupational health and safety audits. evaluation, AI-powered analysis, records management, and reporting steps in a single system. by bringing them together within an integrated system, the speed, consistency and traceability of the process This structure both reduces the workload of field experts and increases productivity. a more organized, standardized and auditable occupational health and safety system for businesses Management infrastructure is being obtained. Detailed Description of the Invention: The terminology used here is intended solely to describe specific applications. and does not limit the scope of the invention. Used here The term "and / or" refers to any of the items listed as related. It includes 4 and all its combinations. Also, the singular "one" used here, "one" The terms "number" and "specified" are used in their plural forms unless the context explicitly indicates otherwise. It is designed to include singular forms such as those mentioned above. Furthermore, the terms used in this specification... The terms "includes" and / or "contains" refer to the specified features, steps, processes, It indicates the presence of elements and / or components, but one or more other features, steps, processes, elements, components and / or groups thereof It will be understood that this does not exclude its existence or addition. Unless otherwise noted, all terms used herein (including technical and scientific terms), in the sense that a person with general knowledge in the field to which this invention belongs would generally understand it They have the same meaning. Furthermore, the terms defined in commonly used dictionaries, in a way that is consistent with the context of the relevant field and this statement it should be interpreted as idealized or exaggerated unless otherwise explicitly defined here. It will be understood that it will not be interpreted in an official sense. The description of the invention will reveal a series of techniques and steps. Each of these... one provides separate benefits, and at the same time, one or more of them or some In these situations, all of the other techniques described can be used in combination. Accordingly, To ensure clarity, each step in the disclosure of the invention should be presented as accurately as possible. By doing this, unnecessary repetition of all combinations will be avoided. Together, the specification and claims, such combinations fully constitute an invention and claims. It should be read with the understanding that it falls within its scope. The invention utilizes images, text, and images obtained from the field during occupational health and safety inspections. Collecting user input via mobile devices ensures the security of that data. and transferred to the server side in a structured form, AI-powered analysis engine Risk analysis reports, emergency situations, based on evaluation and analysis results by the relevant authorities. Automatic generation of plans, safety instructions, and similar outputs. A multi-platform occupational health and safety risk analysis and audit developed for this purpose. It is related to the management system. 5. The main purpose of the invention is to identify risks in the field during occupational health and safety audits. manual saving, later transferring photos to the computer, reports manual preparation and further investigation of legally compliant measures by an expert The goal is to technically improve time-consuming and error-prone processes like these. To this end... The invention includes; mobile device, application server, artificial intelligence analysis engine, security data transfer layer, validation middleware, central database, web-based administration a system architecture where the control panel and the automated document generation engine work together in an integrated manner. It is structured within. Within the scope of the invention, the user conducting the field inspection runs on a mobile device. the business, department, personnel, equipment or that is being monitored through the application The user enters data related to the work area into the system. The user identifies the following during the audit: The device can capture an image of the danger or risk element with its mobile device camera or device. It can select an existing image from its memory. It can also record sounds using the mobile device's microphone. It can record the captured image and sound directly as a raw file. Instead of sending it directly to the server, it pre-processes it on the device. This pre-processing involves: reducing the image size and adjusting its resolution to a suitable level for transmission. bringing in, cleaning up unnecessary file information, reducing data size, and This may include masking sensitive areas within the image when necessary. The pre-processed image data is then transmitted by the system's secure transmission layer. It is converted into a textual transmission format. This conversion allows the image to be transmitted over the network in a standard format. and enables it to be transmitted within a structured data packet. Raw image data not being transferred directly as a file, especially on mobile internet connections Possible issues include timeouts, file corruption, incompatible file formats, and transfer interruptions. It also helps reduce technical problems such as these. In addition, the image within the application It will be processed and converted into a transmission format, then sent to the artificial intelligence analysis engine. It allows for the standardization of the data structure. In a preferred application of the invention, image data prepared on the mobile device, audit information, user ID, business information, location of risk, description a structured version including the text and, if applicable, user-entered risk information. It is converted into a request packet. This request packet is then processed using a secure application programming interface. Data is transmitted from there to the application server. User authorization is required during data transfer. Session verification, timestamp, transaction log, and data integrity check. This is feasible. This ensures that only authorized users can send data to the system. This is provided and the audit records from the field specify which user, at what time, and which It becomes traceable that it was created for the business. The application server initially pre-validates the data coming from the mobile device. This verification checks whether the transmitted data packet contains missing fields, and whether the image and sound are correct. whether the data is in the expected format, and whether the user authorization is relevant to the business or control. whether it is valid for registration and whether the data size is within the limits that the system can process. It is checked whether it is present or not. Missing, corrupted or unauthorized data packets, It may be rejected by the system before it is transmitted to the artificial intelligence analysis engine, or A resend notification can be generated for the user. This allows for AI analysis. This prevents the engine from operating with faulty, incomplete, or inappropriate data. The verified data packet is sent to the artificial intelligence analysis engine. The artificial intelligence analysis engine, By processing the image, sound, and textual control data transmitted to it, it can create changes in the image or It attempts to identify the hazard element mentioned in the statement. In this context, the analysis engine... Risk of falls in the workplace, electrical hazards, lack of personal protective equipment, fire risk, lack of machine guards, improper storage, chemical exposure, transit It can classify obstacles or similar occupational health and safety risks. Analysis engine also the legislative headings that may be related to the identified risk, and the measures that need to be taken. measures, risk severity, probability and frequency (exposure) values, risk used risk score according to the assessment methodology (e.g., 5x5 L-Type, Fine Kinney) and It generates suggested actions. One of the key aspects of the invention is that the result from the artificial intelligence analysis engine is freely available. not as text, but as a predefined structured data format returned to the system. This is the structured output that is returned. This output includes hazard definition, risk description, risk category, probability value, severity value, frequency (exposure) value, calculated risk methodology (e.g., 5x5 Matrix, Fine Kinney), risk score, recommended measure, explanation of relevant legislation, 7 responsible persons, completion time, priority level, and description to be included in the report. It may include these fields. The structured nature of the output means that the analysis result is directly presented. This enables the data to be processed into the database and used by the reporting engine. Outputs returned by AI analysis engines may sometimes contain missing information. It may not conform to the expected data type or may be generated invalidly. Invention Within this scope, a connection between artificial intelligence output and the database is established to address this problem. A verification and error management middleware is used. This middleware is controlled by artificial intelligence. It checks whether the generated output conforms to a predefined data schema. For example, whether the risk score is within the numerical range, and whether the required fields are empty. whether or not, legislation and precautionary measures are within the structure accepted by the system It is checked whether it exists and whether the output integrity is preserved. Invalid When an output is detected, the system can request a new output from the artificial intelligence analysis engine. The system may request additional information from the user or place the relevant record on hold for manual review. Verified analysis results are recorded in a central database. The central database contains information about the company. records, personnel information, training dates, health examination validity dates, audit history, risk records, photo links, recommended precautions, actions It is configured to store their status and the history of generated reports. Each audit record in the database contains information about the relevant company, department, personnel, image, risk, and report. It is linked to the data. This allows audits conducted at different times to be compared. The evolution of risks within the same business can be monitored and repeated based on previous risk records. The irregularities can be identified. The system will work synchronously between the mobile application and the web-based administration panel. It has been designed. The mobile application handles data entry from the field, image and sound capture, and While enabling the audit initiation process, the web-based administration panel allows access to the company and personnel. It enables centralized monitoring, reporting, and user management processes. It recognizes the risk record the user created in the field with a mobile device, synchronizing it with the system. Thanks to its infrastructure, it can be viewed in the web panel. Similarly, in the web panel... Updated company or employee information becomes available on the mobile application. In one of the inventions, the mobile application is available on Android and iOS operating systems. It is developed with a multi-platform software architecture that can work together seamlessly. This architecture shares common characteristics with multiple platforms. business logic, data model, and communication layers should be as single-codebase as possible. While managed through, camera access, file selection, notifications and the operating system Platform-dependent processes such as permission management require operating system-specific hyperlink codes. This is achieved through [method / way]. Thus, the same control flow is maintained across different operating systems. This ensures that the same data structure is preserved. The server architecture used within the scope of the invention, and the information coming from the mobile application and web panel. an application server that fulfills requests, and an AI that performs analysis processes. The analytics service includes a database server that stores audit data and report outputs. It may include a document production service that prepares documents. These components are located on a single physical server. It can run on a single server, as well as on separate server environments. In high-usage situations, analysis processes can be queued and processed asynchronously. This can be executed via the queue. This allows for the analysis of high-resolution images. the user interface freezing, the request timing out, or during the process Technical issues such as the server being unable to respond due to sudden load are reduced. The automated document generation engine uses verified risk analysis from the AI analysis engine. Generates reports using data and company / personnel information registered in the database. It creates a system that includes a risk analysis report, an emergency plan, and occupational safety instructions. It can create a list of nonconformities, an action tracking form, or similar documents. The production engine matches the fields in the database with the fields in the predefined document templates. It matches and analyzes the resulting hazard, risk, legislation, and precautionary information with the relevant parties. It automatically places the data into document sections. This allows the user to easily access information created in the field. Based on the audit log, a manual report must also be prepared later. You can download the document generated by the system without delay. One of the technical advantages offered by the invention is the image and risk data generated in the field. Instead of keeping it scattered across different applications, a standard data structure It is centrally managed through the device. In current applications, photos are managed through the device. In the gallery, notes are in separate applications, personnel information is in separate files, and reports are in the office. It can be stored in 9 programs. In the invention, image data, risk record, analysis result, personnel. The information and report output are linked within the same data model. This prevents data discontinuity. is reduced and the flow from the beginning of the audit process to report generation is technically streamlined. It is made traceable. Another technical advantage of the invention is that the artificial intelligence analysis result is directly fed into the system. This means that the data is validated before processing. This feature allows for the identification of corrupted, incomplete, or incorrect data models. Non-compliant AI outputs are prevented from being recorded in the database. Thus, the report... the occurrence of faulty areas in production, the system unexpectedly stopping, or This prevents the user from being presented with incomplete reports. This middleware is used in AI-powered systems. It improves system stability by reducing the frequently observed inconsistent output problem. The invention also enables field images with high data sizes to be transmitted over mobile connections. It reduces technical problems encountered during transmission. Image processing on the device, Converting it into a transmission format and sending it over a secure communication layer, It provides a more controlled data flow compared to raw file transfer. This structure allows for low-connection. transferring audit data to the system even in field environments where quality is poor It makes it easier. In one example use case, a workplace safety expert conducts an inspection at a production facility. The expert identifies a work area where protective equipment is not being used. The mobile The application selects the relevant company and audit area, and the device displays the risk image. It takes a picture with its camera and enters a short description. The mobile application displays the image on the device. It converts the data into a suitable format for processing and transmission. The generated data packet is secured via API. The data is transmitted to the application server via [intermediate process]. The server verifies the data packet, and artificial intelligence [controls the process]. It sends the image and description to an analysis engine. The artificial intelligence analysis engine analyzes the image and description. It identifies deficiencies in personal protective equipment, and provides a risk statement and measures to address them. It generates a proposal. The generated output is passed through a validation middleware and, if found suitable, This risk record is then stored in the central database. The system then uses this risk record to identify the relevant risk. the company's risk analysis report and, if deemed necessary, instructions or action to follow. It is automatically processed into the form. The invention is not merely an office automation software that enables report generation, Field footage and surveillance data are analyzed using mobile devices, servers, and artificial intelligence. a technically structured data processing between the engine and the central database It ensures that the flow is subjected to it. Therefore, the system processes image data securely. transfer, validation of the AI output against the schema, consistent transfer to the database. endpoints that include recording and automatic use by the reporting engine It offers a cutting-edge technical architecture. In different applications of the invention, the artificial intelligence analysis engine uses an external artificial intelligence. from the service, from an on-premises AI server, or from a local server running on the server. It can consist of an analysis model. Similarly, a database can be a relational database, a document database. It can be implemented in the form of a database-based or hybrid data storage structure. Report outputs spreadsheet, portable document format, or web-based viewable report It can be produced in this format. These differences do not alter the fundamental working principle of the invention. applicable. In conclusion, the invention relates to the image and data inputs obtained during mobile field inspections. Preparation on the device, transmission to the server via a secure data transfer layer, artificial intelligence analysis engine's structured evaluation, artificial The intelligence output is checked with a validation middleware and sent to the central database. processing and automated production of occupational health and safety documents from this data In this way, technically advanced occupational health and safety inspection processes were carried out manually and in a scattered manner. as an integrated, traceable, standardized and fault-tolerant data processing system It is transforming into a system. 11
Claims
1- The invention is an occupational health and safety risk analysis and audit management system, characterized by: Mobile device-based imaging and monitoring related to field inspections. a mobile data collection interface that enables the entry of data, an application server that communicates with the mobile data collection interface in question, Application of data packets generated by the mobile data collection interface a secure data transfer layer that ensures the data is transmitted to the server, By analyzing inspection data such as images, audio, and text related to field inspections An artificial intelligence analysis engine that generates risk, regulatory, and preventive information. The analysis output returned by the artificial intelligence analysis engine is predefined data. a validation and error management middleware that checks according to its structure, a central location where verified analysis outputs and audit data are stored database, Occupational health and safety documents from data registered in the central database It includes a document generation engine that creates. 2- The system mentioned in Claim 1, its feature is; mobile device during field inspection. The image captured by the camera is sent directly to the application server as a raw file. Mobile data collection that pre-processes data on a mobile device before sending it. It is characterized by having an interface. 3- The system mentioned in Claim 2, its characteristic is that the pre-processing in question is done on the image. reducing size, adjusting resolution for transmission, removing unnecessary files cleaning of information, reduction of data size and / or sensitive image areas It is characterized by the fact that it includes at least one of the masking processes. 4- This is the system mentioned in any of the previous requests, and its features include image, sound, and Text data along with information about the audited company, NACE information, and audit area information. at least user information, location information, description data and timestamp data. a secure data structured data packet that transmits one of the data packets to the application server It is characterized by having a transport layer. 12 5- This is the system mentioned in any of the previous requests, and its feature is: mobile data collection. Before transmitting the data packet from the interface to the artificial intelligence analysis engine, the data integrity, user authorization, data size, presence of required fields, and data format It is characterized by having an application server that controls it. 6- This is the system mentioned in any of the previous requests, and its characteristic is; field inspection. Hazard definition, risk description, risk assessment based on related image, sound and control data. category, probability value, severity value, frequency (exposure) value, calculated risk methodology (e.g., 5x5 Matrix, Fine Kinney), risk score, recommended measures, and relevant legislation. an artificial intelligence analysis engine that produces analysis output containing at least one of the fields of knowledge It is characterized by having 7- This is the system mentioned in any of the previous requests, and its characteristic is; in the analysis output missing field, incorrect data type, risk score outside acceptable range, empty required field, or If it detects an invalid data structure, the analysis output is sent to the central database. It has a validation and error management middleware that prevents processing. It is characterized by its being. 8- This is the system mentioned in any of the previous requests, and its feature is; company records, personnel information, training dates, health examination validity dates, audit history, risk records, image data, recommended measures, action plans, and generated report. It has a central database structured to store at least one of its pasts. It is characterized by its being. 13