KR102399597B1 - Intelligent safety analysis technology using AI - Google Patents

Abstract

The present invention relates to a failure type influence analysis automation system using artificial intelligence, capable of minimizing human errors. According to the present invention, the failure type influence analysis automation system using the artificial intelligence includes a pre-processing means and a post-processing means, wherein the post-processing device includes a robotic process automation engine unit.

Description

Failure mode impact analysis automation system using artificial intelligence {Intelligent safety analysis technology using AI}

The present invention relates to an automated system for failure mode impact analysis using artificial intelligence, and more specifically, for Failure Mode and Effect Analysis (FMEA), a safety analysis technique, failure mode impact analysis (FMEA) It relates to a failure mode impact analysis automation system using artificial intelligence that utilizes and automates the structural analysis, failure analysis, and risk analysis steps of .

As software such as artificial intelligence, self-driving cars, robotic technology, and the Internet of Things are converged throughout society and industry, they have emerged as a key driving force for high added value creation. are doing

In other words, as the possibility of damage to life and property due to the increase in software utilization and complexity increases, trust in software and interest in safety are increasing, but a systematic solution for safety analysis of software has not been established.

Compared to the strengthening of international software safety requirements, the technical capabilities (execution and management system, etc.) for securing software safety of domestic companies are insufficient, so it is necessary to enhance the capabilities of related software development activities.

Based on new technologies such as artificial intelligence, big data, Internet of Things, block chain, cloud, and mobile, which are rapidly emerging, functional safety standards and software life cycle processes in the industry required for the implementation of each diffusion technology The scope of use of software related to safety and security is being reviewed at the national level.

The software industry as a whole, across product and service areas, includes packaged software consisting of system software (stand-alone, embedded) and application software (applications, web browsers, etc.), and IT services including IT system development and management technology, Technology is becoming more complex with embedded software, etc.

In response, the proportion and complexity of development for commercialization is also rapidly increasing across the electronics and communication device industries, automobile industry, aviation industry, and medical industry. As a result, it is expanding to the area of potential disasters and disaster prevention.

Therefore, the industrial area requiring various intelligent safety analysis platforms based on software safety analysis and management technology for the prevention of national disasters, disasters and accidents is expanding, centering on software.

In addition, the software industry suffers from economic losses due to software analysis errors such as electronic control units (ECUs), signal processing systems, automatic navigation systems, and algorithms related to remote control systems, which are gradually developed, and loss of life caused by software safety problems. Policy and social business issues regarding the establishment of a platform that can respond to national disasters and crises are rapidly increasing.

In particular, the need for the introduction of an intelligent software safety analysis system is rapidly increasing due to the rapid increase in the complexity of electric parts of automobiles.

On the other hand, the vehicle-related industry is using safety analysis solutions that output outputs based on qualitative evaluation such as the existing FMEA (failure mode impact analysis), but it is possible to detect potential defects through accurate safety analysis of various and extensive next-generation automotive electronic control devices. It is not yet possible to provide an accurate guide to minimizing it.

In other words, when applying the FMEA method, structural analysis, function analysis, failure analysis, and risk analysis steps must be performed for each electrical component element. On average, more than 500 [items] of detailed items are subject to repeated and qualitative judgment using an Excel program. It requires a lot of time and manual work, and it is difficult to confirm whether the analyzed safety function is selected correctly.

Therefore, pre-processing of convergence services such as software safety analysis technique based on cognitive information convergence and related big data processing technology, such as driving situation, etc., due to the rapid increase in the electrification rate due to the development of driving support system, vehicle network control technology, and vehicle embedded software-related content technology There is an urgent need to introduce a software safety analysis system.

In addition, in the case of the automobile sector, the ISO 26262 standard for the application of functional safety in the automobile sector is based on IEC 61508, the parent standard for functional safety, in order to minimize accidents and loss of life due to malfunction of the electronic control device according to the improvement of the electric vehicle electrification rate of the vehicle. Although enacted and strengthened regulations, it is difficult to apply standard safety analysis techniques such as FMEA directly to practice.

Currently, in the automobile industry, due to the revision of ISO 26262 2nd edition, it is mandatory to comply with ISO 26262 for parts, modules, systems, etc. to be ordered from electronic parts companies that are vendors of automakers such as BMW, Mercedes-Benz, Toyota, Hyundai, and Kia. ,

To secure software safety required by ISO26262, software safety mechanisms such as FMEA are established, and an integrated safety system based on risk assessment, verification, and management, and element analysis specifications such as design verification and safety conditions are required.

Therefore, the automotive industry is an evaluation solution for ISO 26262 safety standard compliance, and through IQ-FMEA of German APIS, AIAG (Automotive Industry Action Group)-VDA (Verband der Automobilindustrie) self-diagnosis analysis and defect mitigation for risk sources , the risk management process, and risk source identification and evaluation management for each project.

However, although the IQ-FMEA solution requires professional judgment due to the nature of failure and risk analysis, it is currently implemented only at the level of entering and querying the database, and is sensitive to human errors such as reduced efficiency due to manual operation and user input errors. There is no countermeasure for this.

Conventionally, there is a qualitative evaluation model that detects the acceleration of the electric electrification rate of more than 50 [%], which has a progressively high proportion of software, and a qualitative evaluation model that detects through existing manual work. There are many factors that are insufficient to detect various risk factors that can occur due to

Therefore, it minimizes the occurrence of disasters and disasters caused by human errors caused by inaccurate safety analysis, and automatically inputs, generates, RPA-based intelligent safety analysis system that enables analysis and management is being talked about.

Nevertheless, in the case of SW robots used in RPA solutions, in general, GUI (Graphical User Interface) is lacking, as there are insufficient open sources and libraries for the development of safety analysis systems suitable for the automotive electronics sector for the establishment of domestic RPA-based intelligent safety analysis systems. It interacts with one or more systems through the system, calls various system APIs (Application Programming Interfaces), and converges with cognitive technologies such as text and voice recognition, natural language processing, and machine learning to broaden the field of automation. There is a lack of solutions to verify risk sources.

As described above, the proportion and complexity of the development of electronic control devices and software (SW) is rapidly increasing across the electronic and communication device industry, automobile industry, aviation industry, and medical industry according to the 4th industrial revolution, resulting in national disasters and disasters. In particular, it is important to apply the safety analysis technique for accident prevention, in particular, to minimize failures in the system and software (SW) development process for failure mode impact analysis (FMEA).

However, when applying FMEA, structural analysis, function analysis, failure analysis, and risk analysis steps must be performed for each system and software (SW) element. falls

Therefore, the present applicant intends to provide the present invention in order to overcome the above conventional problems.

Accordingly, as a prior art for an automated system for analysis of failure mode effects using artificial intelligence,

As shown in (a) of Figure 13, "Intelligent Status Monitoring and Defect Diagnosis System" (hereinafter referred to as 'Prior Art 1') of Korean Patent No. 10-1322434,

a data collection function to obtain temporal histories of selected data with respect to one or more components; a preprocessing function to compute particular characteristics of the temporal histories; and a preprocessing function to evaluate the characteristics to produce one or more assumptions of the state of the one or more components. An intelligent condition monitoring and fault diagnosis system comprising an analysis function and an inference function to determine the status of one or more components from one or more assumptions.

Another prior art,

As shown in (b) of FIG. 13, as a "safety system for human/robot integrated environments" of Korean Patent Application Laid-Open No. 10-2018-0074796 (hereinafter referred to as 'prior art 2'),

The computing device may determine information about any actors present within a predetermined area of the environment, and may determine a safety classification for the predetermined area based on the information, wherein the safety classification determines which actors the information is within the predetermined area. a low safety classification when the information indicates that none of the After determining the safety classification for the predetermined zone, including a higher safety classification when it indicates that one actor is of the second predetermined type, the computing device activates the safety rules for operating within the predetermined zone in the environment. It relates to a safety system for human/robot integrated environments that provides a robotic device for

As can be seen, the prior art 1 to the prior art 2 are,

As a technical field similar to the present invention and in the same technical field as the present invention, there are some similar and identical technical concepts in the problem to be solved in a comprehensive perspective compared to the present invention, but in a specific solution for solving the problem to be solved There is a difference.

Therefore, the present invention is different from the prior art related to the safety analysis system including the prior art 1 and the prior art 2, the problem to be solved only by the present invention (object of the invention), a solution for solving this Based on the (component) and the better effect that is achieved by solving it, we intend to try its technical characteristics.

(Document 01) Republic of Korea Patent Publication No. 10-1322434 (Registered on October 21, 2013) (Document 02) Republic of Korea Patent Publication No. 10-2018-0074796 (published on Jul. 3, 2018)

The present invention has been devised to solve the above-mentioned conventional problems,

It minimizes the rate of disasters and disasters caused by human error caused by inaccurate safety analysis, and automatically inputs, creates, analyzes, The purpose is to provide a cloud solution for automotive software safety analysis process automation using artificial intelligence technology with guaranteed management functions.

In particular, intelligent safety analysis for vehicles that can systematically analyze risks that may impair vehicle safety due to failure of electronic control devices throughout the entire life cycle from the product concept stage, which is the beginning of software safety, to disposal, and cope with those risks. The purpose is to provide a system.

In addition, it aims to minimize human error by automating the structural analysis, failure analysis, and risk analysis steps of the failure mode impact analysis so that a prompt and accurate system safety analysis can be performed.

The present invention for achieving the above object has been devised to achieve the problem to be solved,

In the failure mode impact analysis automation system using artificial intelligence,

a potential processing means for implementing a user interface and a responsive web and mobile platform designed for user experience so that software safety analysis is performed;

Post-processing means for automatically performing safety analysis of software by activation of the front-end processing means by the user;

In the post-processing means,

Python-based robotic process automation engine unit that automates functional safety analysis of VDA FMEA, AIAG FMEA, AIAG-VDA standard software required by ISO 26262, the automotive functional safety standard for large-capacity electronic components;

By automating the complex step process of performing software safety analysis, a series of processes such as item entry, data management, processing, and hazard detection and optimization can be accurately performed in the minimum time,

Robotic Process Automation Engine Department,

a structural analysis module for classifying the functional structures of systems and software to be analyzed in failure mode impact analysis (FMEA);

a failure analysis module that analyzes the sentences of the failure mode impact analysis (FMEA) structure and automatically generates information on severity, detection level, and occurrence level;

A risk analysis module that automatically generates a risk priority number by analyzing the severity, detection degree, and occurrence of the failure mode impact analysis (FMEA) generated through the failure analysis module;

When performing failure mode impact analysis (FMEA), due to the structural analysis module, failure analysis module, and risk analysis module,

By automatically analyzing the structure, failure, and risk of systems and software,

It not only reduces analysis time, but also automates the complex steps of performing software safety analysis so that a series of processes such as item entry, data management, processing, and hazard detection and optimization can be accurately performed in the minimum time. characterized in that

On the other hand, prior to this, in this specification, the terms or words used in the claims of the patent registration should not be construed as being limited to conventional or dictionary meanings, and the inventor of the term in order to explain his invention in the best way. should be interpreted as meaning and concept consistent with the technical idea of the present invention based on the principle that can be appropriately defined.

Accordingly, the embodiments described in the present specification and the configurations shown in the drawings are only the most preferred embodiment of the present invention, and do not represent all of the technical spirit of the present invention. It should be understood that there may be equivalents and variations.

According to the present invention as described above in the above configuration and operation,

For a series of activities that minimize the occurrence of disasters and disasters caused by human error caused by inaccurate safety analysis and guarantee the safety of systems, hardware, software, and components that the user has not identified, AI technology is used to automatically We can provide a cloud solution for automotive software safety analysis process automation with guaranteed input, generation, analysis, and management functions.

In particular, from the product concept stage, which is the beginning of software safety, through the entire life cycle, from the product conception stage to disposal, the risk that can hinder the safety of the vehicle due to the failure of the electronic control device is systematically analyzed and the robot process automation that can cope with the risk It provides an intelligent safety analysis system for vehicles based on technology.

In other words, RPA (Robotic Process Automation) technology and automated RPA (Robotic Process Automation) technology provide rapid decision-making, structural analysis, risk analysis, and failure analysis analysis and execution management functions for compliance with the safety standards required by ISO 26262, the functional safety standard of the next-generation automobile industry. It provides a safety analysis system combined with remote monitoring function.

In addition, by automating the process of 6 [steps] for performing safety analysis, which has been conventionally performed, it is possible to accurately perform a series of processes such as item input, data management, processing, and hazard detection and optimization in a minimum amount of time. In addition, it is possible to minimize human errors that may occur throughout the software development life cycle (SDLC) for hazard analysis and risk assessment.

More specifically,

In the next-generation automobile industry, robotic process automation technology automatically performs 'structural analysis → function analysis → failure analysis → risk analysis → optimization step' for each electric part, and When setting test levels based on independent execution of 'requirements analysis → design → coding → testing', digital work can be done to minimize software defects based on test cases tailored to the requirements through the V-model, which models to be interconnected. It has the advantage of optimizing the quantitative judgment of force.

In particular, it applies and automates AI-based RPA (Robotic Process Automation) technology in the structural analysis, failure analysis, and risk analysis stages to minimize human errors and perform prompt and accurate system safety analysis.

In addition, it can contribute to minimizing the possibility of operation errors, component errors, accidents and recalls by automating failure mode and effect analysis such as FMEA (failure mode impact analysis) at the software level used for electronic components.

In addition, as described above, it is possible to accurately perform a series of processes such as item input, data management, processing, hazard detection and optimization, etc. in the minimum time by automating the process of conventionally complex steps of generally performing safety analysis. make it possible

In addition, due to the intelligent safety analysis system that automatically analyzes and executes structural analysis, risk analysis, failure analysis, etc. for rapid decision-making for compliance with safety standards required by ISO 26262, the present inventor, cognitive-based Accidents and loss of life due to malfunction of various electronic control devices can be minimized with the intelligentization of safety analysis systems in the automotive industry that require customizing and risk-based testing techniques in the process automation area and robotic process automation technology functions.

In addition, it is possible to maximize worker production and quality efficiency of the company by supporting the replacement of commercialized safety analysis software and programs.

Specifically,

1. Safety analysis software and programs that are commercialized in Korea can secure an alternative market for solutions that rely entirely on imports, including 'IQ-FMEA' of German APIS.

2. In addition, it can contribute to the localization of the source technology of new software to which the process automation technique is applied to the present invention.

3. In addition, through cloud-linked services, between remote managers and users, and through smooth communication, we provide a variety of communication and monitoring platforms in web and mobile environments to provide standard processes for a series of processes related to safety analysis. It can be said to be a very effective invention.

1 is a diagram showing the configuration of the failure mode impact analysis automation system utilizing artificial intelligence of the present invention.
2 shows a basic configuration plan for an embodiment of the failure mode impact analysis automation system utilizing artificial intelligence of the present invention.
3 shows a schematic diagram of an embodiment of an automated failure mode impact analysis system using artificial intelligence of the present invention.
4 is a table showing details of an embodiment of the failure mode impact analysis automation system utilizing artificial intelligence of the present invention.
5 shows an embodiment of a potential processing means among the components of an automated failure mode impact analysis system using artificial intelligence of the present invention.
6 shows an algorithm of the failure analysis step among the components of the failure mode impact analysis automation system using artificial intelligence of the present invention as an example.
7 shows an algorithm of the structural analysis step among the components of the failure mode impact analysis automation system using artificial intelligence of the present invention as an example.
8 shows an algorithm of the risk analysis step among the components of the failure mode impact analysis automation system using artificial intelligence of the present invention as an embodiment.
9 is a block diagram schematically illustrating an automated processor of the robotic process automation engine unit among the components of the failure mode impact analysis automation system utilizing artificial intelligence of the present invention.
10 shows a technical flow diagram for an automated system for failure mode impact analysis using artificial intelligence of the present invention.
11 shows, as an example, the execution code of the robotic process automation engine unit among the components of the failure mode impact analysis automation system using artificial intelligence of the present invention.
12 shows an example of an operation by the execution code of the robotic process automation engine part among the components of the failure mode impact analysis automation system utilizing artificial intelligence of the present invention as an embodiment.
13 shows a representative diagram of the prior art for an automated system for failure mode impact analysis using artificial intelligence of the present invention.

Hereinafter, the function, configuration and operation of the failure mode recommendation system 1 of the failure mode impact analysis-based failure analysis according to the present invention will be described in detail with reference to the accompanying drawings.

1 is a diagram showing the configuration of the failure mode impact analysis automation system using the inventor's artificial intelligence 3 is a schematic diagram of an embodiment of the failure mode impact analysis automation system using the inventor's artificial intelligence, and FIG. 4 is a table and diagram showing details of an embodiment of the failure mode impact analysis automation system using the inventor's artificial intelligence 5 shows an embodiment of the potential processing means among the components of the failure mode impact analysis automation system using artificial intelligence of the present invention.

1 to 5, the present invention is

In the failure mode impact analysis automation system (1) using artificial intelligence,

A potential processing means 100 for implementing a user interface and a responsive web and mobile platform designed for user experience so that software safety analysis is performed;

With activation of the front-end processing means 100 by the user, the back-end processing means 200 for automatically performing safety analysis of software;

In the post-processing means 200,

Python-based robotic process automation engine unit 240 that automates functional safety analysis of VDA FMEA, AIAG FMEA, AIAG-VDA standard software required by ISO 26262, the automotive functional safety standard for large-capacity electronic components;

By automating the complex step process of performing software safety analysis, a series of processes such as item entry, data management, processing, and hazard detection and optimization can be accurately performed in the minimum time,

Robotic process automation engine unit 240,

Structural analysis module 241a for classifying functional structures of systems and software to be analyzed in failure mode impact analysis (FMEA);

a failure analysis module 242a that analyzes the sentence of the failure mode effect analysis (FMEA) structure and automatically generates information on severity, detection level, and occurrence level;

A risk analysis module 242b that automatically generates a risk priority number by analyzing the severity, detection degree, and occurrence of the failure mode impact analysis (FMEA) generated through the failure analysis module 242a; composed,

When performing the failure mode effect analysis (FMEA), due to the structural analysis module 241a, the failure analysis module 242a, and the risk analysis module 242b,

By automatically analyzing the structure, failure, and risk of systems and software,

It not only reduces analysis time, but also automates the complex steps of performing software safety analysis so that a series of processes such as item entry, data management, processing, and hazard detection and optimization can be accurately performed in the minimum time. characterized in that

That is, the present invention enables the structural analysis, failure analysis, and risk analysis areas to be automatically analyzed when the failure mode impact analysis (FMEA) of the system and software (SW) is performed by the robotic process automation engine unit 240,

As described above, it relates to an analysis automation system that minimizes human error and enables rapid and accurate safety analysis of systems and software (SW) to be performed.

The present invention is described in more detail,

As shown in FIG. 5, the potential processing means 100 is

The safety analysis model identified through the post-processing means 200, the requirements of the robotic process automation engine unit 240, and the FMEA solution are interlocked to provide users with Javascript, HTML5, jQuery, and CSS3-based web and mobile apps. to be provided

In addition, the post-processing means 200,

a controller unit 210 that manages communication between the front-end processing means 100 and the post-processing means 200 through the REST API;

a service creation unit 220 including a business logic module 221 for processing general business logic and a database mapper module 222 for interworking with a database and business logic;

A base rule module 231 that defines the work contents according to step-by-step analysis of FMEA structure, failure, risk, etc., and a user define rule module ( 232) is included in the rule generator 230;

Python-based robotic process automation engine unit 240 that automates functional safety analysis of VDA FMEA, AIAG FMEA, AIAG-VDA standard software required by ISO 26262, the automotive functional safety standard for large-capacity electronic components;

A data storage storage unit 250 including a database module 251 for storing system data, and a file module 252 for managing various file information such as rules registered through the rule generator 230 and analytes resulting therefrom ); is composed of;

Robotic process automation engine unit 240,

a parser module 241 that analyzes the content of the rule generator 230 and converts it into a business procedure;

Consists of; an actuator module 242 for generating data by applying the business procedure converted through the parser module 241 to the system;

By automating the complex step process of performing software safety analysis, a series of processes such as item entry, data management, processing, and hazard detection and optimization can be accurately performed in a minimum amount of time.

That is, the present invention is an intelligent safety analysis system that performs accurate safety analysis through the robotic process automation engine unit 240, and a series of data related to automatic creation of structures, automatic input of failure analysis items, and automation and optimization of risk analysis There is a feature that automatic input is performed.

Summarizing the core of the technology of the present invention,

As described above, in the front-end processing means 100 that is linked to the web and mobile platforms based on user interface (UI) and UX (User Experience),

The identified safety analysis model (Waterfall, Iterative, AGILE), the requirements of the robotic process automation engine unit 240 and the FMEA solution are interlocked to provide users with Javascript, HTML5, jQuery, and CSS3-based web and mobile apps. provided,

In addition, in connection with the cloud server means 300, the process or result of safety analysis is jointly monitored between the remote manager and the user, and cloud server access rights and licenses are provided for performing common quality assurance activities.

In addition, in the post-processing means 200 in which the main functions of the intelligent safety analysis system are implemented,

The business logic module 221 and the database mapper module 222 of the controller unit 210, the service generation unit 220, which manages communication with the REST API-based prefix processing means 100, and the rule generation unit 230 Base rule module 231 and user define rule module 232, parser module 241 and actuator module 242 of robotic process automation engine unit 240, database module 251 of data storage and storage unit 250 And the file module 252 is configured,

In order to stabilize the function and improve the quality of the present invention,

Configure the Spring Framework using Spring boot2 based on object-oriented Java language and Mybatis Framework for DB programming,

The rule generator 230 is an open standard format JSON (Jason [1], JavaScript , Object Notation) is used,

In the case of the data storage storage unit 250, it is provided through the open source relational DB system Maria DB based on the same source code as MySQL,

The parser module 241 and the actuator module 242 used in the robotic process automation engine unit 240 are configured based on the Django open source-based Web Framework, and large-capacity safety to be analyzed in more than 100 [more than] electronic components. For analysis data, pattern analysis and visualization are performed through Python.

In the future, the pattern analysis and visualization data of Python of the robotic process automation engine unit 240 will be expanded and built as an AI-based basic big data platform.

That is, the potential processing means 100,

Identify and apply safety analysis models (Waterfall, Iterative, AGILE),

Identify and define the requirements of the robotic process automation engine unit 240,

Deriving a common framework for interworking with FMEA solutions,

Javascript, HTML5, jQuery, CSS3 based UI, UX responsive web and mobile platforms that consider user convenience are provided.

In addition, the post-processing means 200,

A controller unit 210 that manages communication with the REST API-based prefix processing means 100, a service generation unit 220, a rule generation unit 230, a robotic process automation engine unit 240, data storage and storage Through the unit 250, the progress and evaluation of the system integration test is performed.

At this time, the robotic process automation engine unit 240,

Through the structural analysis module (241a), the failure analysis module (242a), the risk analysis module (242b),

A failure analysis step (A) of inquiring the information in the worksheet for failure analysis, extracting a character string, and determining the occurrence value;

Structural analysis step (B) of inquiring the information of the worksheet and the list of submodules of the worksheet for structural analysis, and copying the important information diagram of the existing worksheet;

For risk analysis, the information in the worksheet is checked, the type of worksheet is checked, and the risk analysis step (C) is analyzed to determine the AP (action priority) value;

By automating the failure analysis step (A), the structural analysis step (B), and the risk analysis step (C),

It minimizes human error and enables rapid and accurate system safety analysis to be performed.

6 to 10, if more specifically described,

The failure analysis step (A) is,

Failure analysis worksheet information inquiry element (A-1) for inquiring worksheet information for failure analysis;

A character string reading element (A-2) that reads each character string in the severity, occurrence, and detection standard columns for failure analysis of the worksheet;

A string extraction element (A-3) for extracting a string through a morpheme analysis engine;

It consists of a; occurrence value determination element (A-4) that determines the occurrence value based on the character string registered in the parser module 241 and the occurrence value.

to automatically perform system failure analysis,

Structural analysis step (B),

Structural analysis worksheet information inquiry element for inquiring worksheet information for structural analysis (B-1);

Sub-module list inquiry element (B-2) for retrieving the sub-module list of the worksheet;

Consists of; FMEA worksheet addition element (B-3) for adding the sub-module list searched through the sub-module list inquiry element (B-2) to the FMEA worksheet;

to automatically perform structural analysis of the system,

The risk analysis step (C) is,

Risk analysis worksheet information inquiry element (C-1) for inquiring worksheet information for risk analysis;

a worksheet form confirmation element (C-2) for confirming the type of worksheet inquired by the risk analysis worksheet information inquiry element (C-1);

AP value discrimination factor (C-3) that determines the severity, incidence, detection level, and AP (action priority) value of the worksheet;

After the form of the worksheet is confirmed by the worksheet form confirmation element (C-2), the RPN value is generated to generate the RPN (risk priority number) value by calculating the severity, occurrence, and detectability values of the worksheet Element (C-4); consists of,

It enables to automatically perform risk analysis of the system.

As such, the robotic process automation engine unit 240 for automating the failure mode impact analysis (FMEA) for the system and software (SW),

It is designed as an executable code as shown in FIGS. 11 to 12 .

On the other hand, in the failure mode impact analysis automation system (1) utilizing artificial intelligence of the present invention,

The cloud server means 300 for allowing the user and the remote manager to jointly monitor the process or result of the safety analysis of the software performed by the backend processing means 200; may be further included or configured.

Cloud server means 300,

As a server for building a remote monitoring platform of the present invention, the analyzed data of the number of safety analysis of 1000 [EA] or more per session based on the analysis information monitoring cycle of the robotic process automation engine unit 240 2 [times] per second, It is built as a dedicated server to which web, Android, and iOS platforms that perform cloud service traffic forwarding over 3[EA] are accessed.

The present inventors constructed and constructed in this way,

Through an automated intelligent safety analysis system that enables rapid decision-making for compliance with the safety standard required by ISO 26262, the functional safety standard of the next-generation automobile industry, and automatic analysis and execution management functions such as structural analysis, risk analysis, and failure analysis,

1. Standard SW functional safety analysis and FMEA technique automatic analysis technology of VDA FMEA, AIAG FMEA, AIAG-VDA required by ISO 26262, the automotive functional safety standard;

2. In the structural analysis, failure analysis, and risk analysis stages of the existing FMEA qualitative safety analysis (failure mode analysis and risk assessment), AI-based RPA technology is applied to automate the safety analysis and make decisions or judgments of the analyzed results. The fusion technology of the robotic process automation engine unit 240 that minimizes human error based on the improvement of accuracy and the minimization of the execution time required for analysis,

3. Interworking technology with responsive web and mobile platforms that enhance usability, such as central database management;

4. Cloud server-based web and mobile platform display technology is provided for jointly monitoring the safety analysis process or result between the remote manager and the user.

In other words, for the automation of safety analysis techniques, a robotic process automation engine within the safety analysis mechanism that can minimize human errors based on the improvement of the accuracy of judgments such as structures, failures, and risk sources for multiple vehicle electronic control devices and the minimization of execution time. By mounting the unit 240, it provides an intelligent safety analysis system based on web and mobile platforms for automation of automotive software safety analysis and central database management of corresponding information.

In addition, the present invention provides a safety analysis system that provides a license and a system to a number of companies or users to output to a web and mobile platform in conjunction with a cloud server to monitor the safety analysis process or result from a remote location do.

With reference, described in the present invention,

'ISO 26262' is,

C programming language development guidelines for improving code stability, compatibility, and reliability of embedded system software for the automotive industry developed by MISRA as a safety standard for automotive electrical and electronic systems.

In addition, 'software stability' means,

Based on the preceding process of risk analysis at the overall system level and the system safety requirements derived from the risk analysis results, it is checked whether the safety function is correctly selected, and the state in which the safety function performed by the final result is operating normally say

In addition, '6 [step]' means,

It refers to range definition, structural analysis, function analysis, failure analysis, risk analysis, and optimization, and in particular, the purpose of the present invention is to automate structural analysis, failure analysis, and risk analysis.

In addition, characters written through '[]' indicate units or quantities.

In addition, if the terms described in the present invention are briefly summarized,

FMEA, an abbreviation of Failure Mode and Effect Analysis, is one of the reliability analysis methods that sequentially analyzes the relationship between accidents and causes in accidents that occur while using the system.

IQ-FMEA is the product name of the functional safety evaluation solution of APIS, a German competitor. can

However, there is a disadvantage in that the reliability of the output changes according to user efficiency.

RPA, an abbreviation of Robotic Process Automation, is a program that learns simple and repetitive human processes and uses artificial intelligence to handle tasks as it is.

Component refers to the technology of assembling the module when developing the necessary application technology after making a software component unit (module) in advance, out of development by the existing coding method.

In the V-model, the work of testing a system, system component, or software program, etc. is not just a task of finding simple coding errors included in these tasks, but also tests of errors in requirements analysis and development stage tasks such as design. It refers to the concept of systematization by integrating the contents of inclusion with the development process.

Workforce refers to the workforce.

REST is an abbreviation of Representational State Transfer. It refers to anything that classifies resources by name (representation of resources) and exchanges the state (information) of the resource (that is, by Representation of Resources). State transfer The specific concept means specifying a resource through HTTP URI (Uniform Resource Identifier) and applying CRUD operation to the resource through HTTP Method (POST, GET, PUT, DELETE) do.)

API, an abbreviation of Application Programming Interface, refers to providing a set of data and functions to facilitate interaction between computer programs and to exchange information with each other.

REST API is a compound word of Representational State Transfer API, and refers to implementing a service API based on REST.

Spring boot refers to a project of the Java framework called Spring, and by using it, you can easily create an application.

Spring Framework is an open source application framework for the Java platform, also called simply Spring. (Various services for developing dynamic web sites are provided.)

That is, for the efficient development of the post-processing means 200, an open source framework is built based on the Java language.

Mybatis Framework is a development framework that helps to facilitate relational database programming of Java, an object-oriented language.

Thus, with it, a persistence framework is built that supports developer-specified SQL of programming, stored procedures, and advanced mapping.

JSON (JavaScript Object Notation) is an open standard format that uses human-readable text to convey data objects consisting of 'property-value pairs' or 'key-value pairs'.

Object Notation is an object notation.

Maria DB, an open source relational database management system, is based on the same source code as MySQL and follows the GPL v2 license.

Python, a high-level programming language, is a platform-independent, interpreted, object-oriented, dynamic typing interactive language.

Framework refers to providing classes in a collaborative form to enable reuse of design and implementation corresponding to specific parts of software.

In addition, RPN (risk priority number) is a risk priority number, and is calculated from severity, frequency of occurrence, and degree of detection.

Severity, how severe the pain felt by the customer,

Occurrence frequency (occurrence rate) is how often the failure occurs,

The degree of detection refers to how accurately the cause of a failure is discovered.

As described above, the present invention is not limited to the described embodiments, and it is apparent to those skilled in the art that various modifications and variations can be made without departing from the spirit and scope of the present invention.

Therefore, since it can be implemented in other various forms without departing from the technical spirit or main characteristics, the embodiments of the present invention are merely examples in all respects and should not be construed as being limited and may be implemented with various modifications.

The present invention relates to a failure mode impact analysis automation system using artificial intelligence,

It can contribute to the promotion of industrial fields that require various intelligent safety analysis platforms based on software safety analysis and management technology, such as software industries used in electronic and communication devices, automobiles, aviation, and medical industries.

In particular, rapid decision-making for compliance with the safety standards required by ISO 26262, structural analysis, risk analysis, failure analysis, etc. are automatically analyzed using artificial intelligence, and the recognition-based process automation area and RPA function customization and risk-based It can be used as a technology that can minimize accidents and loss of life due to malfunction of various electronic control devices by making intelligent safety analysis solutions in the automotive industry that require testing techniques.

1: Failure mode impact analysis automation system using artificial intelligence
100: front-end processing means 200: post-processing means
210: controller unit 220: service generation unit
221: business logic module 222: database mapper module
230: rule generator 231: base rule module
232: user define rule module 240: robotic process automation engine unit
241: parser module 241a: structure analysis module
242: actuator module 242a: failure analysis module
242b: risk analysis module 250: data storage storage unit
251: database module 252: file module
300: cloud server means
A: Failure analysis step A-1: Failure analysis worksheet information inquiry element
A-2: String reading element A-3: String extraction element
A-4: Occurrence value judgment factor
B: Structural Analysis Step B-1: Structural Analysis Worksheet Information Inquiry Element
B-2: Submodule List Query Element B-3: FMEA Worksheet Additional Element
C: Risk analysis stage C-1: Risk analysis worksheet information inquiry element
C-2: Worksheet Form Check Factor C-3: AP Value Discrimination Factor
C-4: RPN value generation factor

Claims (1)

In the failure mode impact analysis automation system (1) using artificial intelligence,
A potential processing means 100 for implementing a responsive web and mobile platform designed with a user interface and user experience so that safety analysis of software is implemented;
With activation of the front-end processing means 100 by the user, the post-processing means 200 for automatically performing safety analysis of software;
In the post-processing means 200,
Python-based robotic process automation engine unit 240 that automates functional safety analysis of VDA FMEA, AIAG FMEA, AIAG-VDA standard software required by ISO 26262, the automotive functional safety standard for large-capacity electronic components;
A series of processes such as item input, data management, processing, hazard detection and optimization by automating the process of failure analysis step (A), structural analysis step (B), and risk analysis step (C) that performs software safety analysis to be performed accurately in the least amount of time,
Robotic process automation engine unit 240,
Structural analysis module 241a for classifying functional structures of systems and software to be analyzed in failure mode impact analysis (FMEA);
a failure analysis module 242a that analyzes the sentence of the failure mode effect analysis (FMEA) structure and automatically generates information on severity, detection level, and occurrence level;
A risk analysis module 242b that automatically generates a risk priority number by analyzing the severity, detection degree, and occurrence of the failure mode impact analysis (FMEA) generated through the failure analysis module 242a; composed,
When performing the failure mode effect analysis (FMEA), due to the structural analysis module 241a, the failure analysis module 242a, and the risk analysis module 242b,
By automatically analyzing the structure, failure, and risk of systems and software,
By automating the process of failure analysis step (A), structural analysis step (B), and risk analysis step (C) that performs software safety analysis as well as shortening the analysis time, item input, data management, processing, and risk sources Characterized in that a series of processes such as detection and optimization can be accurately performed in a minimum amount of time,
Failure mode impact analysis automation system using artificial intelligence.

Images