sentin

The AI-ready NDT infrastructure Guide.

Wed, 02 Oct 2024 11:34:21 +0000

For years the hardware and software tools in non-destructive testing (NDT) have grown and added step by step. With the rise of Artificial Intelligence (AI) and the whole technology stack of NDT 4.0 new tools and approaches are needed to use the full potential of a company’s data. This article deals with AI-ready NDT infrastructure consisting of hardware and software, its benefits and requirements needed to use AI’s full potential.

What is an AI-ready NDT infrastructure?

An AI-ready NDT infrastructure is an ecosystem designed to support the integration of AI and other modern technologies into NDT processes. It seamlessly combines acquisition hardware, data storage, evaluation software, and advanced algorithms to create an intelligent, adaptable, and efficient inspection environment.

What are the components of an AI-ready infrastructure?

Acquisition Hardware: High-quality data acquisition hardware (sensors, scanners, detectors and cameras etc.) ensures that accurate, consistent, and detailed inspection data is captured. AI-ready acquisition systems store data digitally and pre-process and structure it for further handling. Data should be losslessly stored so that an AI can easily be trained and understand the data.
Revision-safe data storage / PACS: Effective storage solutions are crucial. AI-driven NDT systems require extensive datasets for training, validation, and evaluation. Often cloud and decentralized storage systems are used to be scalable, secure, and fast, ensuring that data can be accessed and processed as needed. In non-destructive testing it is also important to store the data in a revision-safe manner. Often a Picture Archiving and Communication System (PACS) is used in this context. To support the latest AI and data frameworks we suggest to use an newer, encapsulated data-format on top of DICOM / DICONDE (supported by second generation evaluation tools like the sentin EXPLORER), because this standard was introduced in the 90s without AI in mind.
Learn more here: What is a PACS in NDT?
Evaluation / Viewer Software: To become AI-ready the latest evaluation / viewer software like the sentin EXPLORER should be used. These not only have AI workflows integrated but also support working with data to build an AI model. The first generation of digital evaluation software may not be suitable for preparing and storing data with AI in mind. When „burning in“ image information / findings on top of an x-ray image for example, the information beneath is lost and cannot be used by an AI. The second generation of tools, on the other hand, store the data losslessly. These tools boost evaluation with AI models e.g. for text-recognition, defect recognition or fraud detection, and automatically generate reports as well as support the dataset cleaning and working with AI models (more on that later).
IMS / ERP Integration: Integration with Inspection Management Systems (IMS) and Enterprise Resource Planning (ERP) solutions ensures that the whole inspection workflow is considered when working with AI. It will get inputs from the other connected systems (e.g. label an order as done, when all inspection reports have been generated) and provide insights that can be leveraged across multiple departments and processes.
AI / Algorithms: At the heart of an AI-ready NDT infrastructure are the advanced AI models and algorithms. These use historical data to learn patterns, detect anomalies, and even predict failures before they happen (e.g. predictive maintenance). Continuous retraining of these models ensures they stay accurate and relevant. The applications of AI are so diverse they range from super-resolution algorithms reducing irradtion time in CT, over defect recognition in ultrasonic testing, to fraud detection and pattern recognition in the order data.
Learn more here: How does Artificial Intelligence (AI) in NDT work?

What are the benefits of an AI-ready NDT infrastructure?

Implementing an AI-ready NDT infrastructure can transform operations in numerous ways:

Unlocking AI Potential: Setting up an AI-ready infrastructure allows companies to work with data-driven insights and become future-proof.
Time and Cost Savings: Improving data acquisition, analysis, and reporting saves significant time. The reduction in manual processes also lowers operational costs, as AI-driven systems can work around the clock with minimal human intervention and higher precision.
Automated Workflows: AI-ready systems enable automated workflows, eliminating bottlenecks and streamlining tasks from acquisition to evaluation. This makes complex NDT processes easier to manage and reduces downtime.
Increased Transparency and Insights: AI systems provide deeper insights into testing processes and results. These systems can uncover hidden patterns and trends that humans might overlook, resulting in a more transparent and data-driven approach to NDT.
Maintainability and Scalability: AI-ready NDT infrastructures are designed to be easy to maintain and scale. As new testing requirements emerge, the system can evolve, and is ready for new datasets, hardware, and algorithms without major changes.
Revision-Proof Systems: The new systems store and track all changes and updates, making it easier to review historical data and ensure compliance with regulatory standards, while using top-notch technology to boost the inspection efficiency.

What tasks does an AI-enabled infrastructure assist with?

For an AI-ready NDT infrastructure has to support several tasks. These include the normal inspection tasks and new tasks that are arising from the work with AI.

Regular inspection tasks:

These tasks are part of the normal inspection workflows. The inspection infrastructure should support them to not interrupt the regular processes and ensure the company can work as before.

Data Acquisition: Capturing high-quality, consistent data from a variety of NDT methods (ultrasonic, radiographic, electromagnetic, etc.) using advanced sensors and acquisition devices.
Data Storage: Providing reliable and scalable storage solutions capable of handling vast amounts of data with the ability to quickly retrieve and process information for real-time decision-making.
Evaluation / Interpretation: Utilizing AI and ML algorithms to analyze data, identify defects, and detect anomalies with greater accuracy and speed compared to traditional methods.
Reporting: The generation of reports and insights should be automated, clear, and customizable to meet the needs of different stakeholders, from field technicians to upper management.

AI related tasks:

These tasks arise from working with AI. They are things that data scientists do regularly, but they are new to the world of NDT. Not every inspector needs to understand them or do them regularly, but an organisation’s success in applying AI will depend on good tools and managers who understand these processes.

Dataset Creation, Labeling, and Cleaning: AI models require clean, well-labeled datasets for effective training. The infrastructure should support automatic dataset creation, accurate labeling, and effective cleaning processes to eliminate noise / outliers and improve data quality.
(Re-) training: The process of “teaching” an AI model what to do is called training. It requires special hardware (like GPUs) and software tools. You can use a prior state and existing training pipeline with new data to re-train and improve the model’s performance or teach it new things. This is necessary if a new inspection challenge arises (e.g. new parts that are inspected or data points / images that look significantly different from what you have had before).
Validation: Before rolling out an AI model and use it in a real world environment you want to be sure it performs as expected e.g. having a certain accuracy. This process is called validation. Modern tools will allow you to work with the model e.g. on a validation data set and show you statistics and insights for you to make informed decisions whether you want to use a new model, its prior version or train a completly new one.
Deployment and Monitoring: The infrastructure should be capable of continuously supporting you in retraining and validating AI models as new data becomes available. For using it in an productive environment it must also deploy updated models efficiently (e.g. to make it available for all inspectors) and monitor their performance to ensure accuracy and reliability. It should be possible to see how a model is currently doing and deploy an update to a server with just a few clicks.

What are the requirements of an AI-ready NDT infrastructure?

For the infrastructure to support AI effectively, certain requirements must be met:

Interoperability: The system must be able to integrate with existing hardware and software platforms, supporting different NDT methods and ensuring seamless data exchange between systems. Using a mix of industry standard and modern interfaces is the key to success here. An example can be the DICONDE standard combined with RESTful web interfaces.
Workflow Awareness: The infrastructure should enhance existing workflows by making them AI- and data-centric, enabling smoother collaboration between human experts and AI algorithms and tools. Besides enhanced existing workflows there also will be new workflows that emerge from using AI like Critical Image/Item Detection (CID). It is therefore important to have a holistic overview of existing and new workflows.
Modern Interfaces and Data Formats: To create usable datasets and facilitate real-time analysis, the system should support modern data formats and interfaces, allowing for flexible interaction and integration with other digital systems. The DICONDE standard and PACS are commonly used, but to support AI-centric workflows new interfaces and data formats e.g. based on readable formats like JSON are the key to success.
Maintainability and Scalability: An AI-ready infrastructure must be both easy to maintain and scalable, ensuring that it can adapt to evolving industry requirements and increasing volumes of data without significant reconfiguration or downtime. NDT can learn from past years of cloud and web development which made huge advances with container and micro-service architectures.

How to get an AI-ready NDT infrastructure?

After describing benefits and requirements of AI-ready infrastructure solutions we now want to show you some applications of AI and the tools that are commonly used.

We support you from data acquisition to report generation
with AI and automation solutions.

Software Products:

Services:

Analyzing the status quo and potentials of analog and digital workflows
AI dataset acquisition, annotation and cleaning
AI model training & validation
Development of digital automation & assistance tools as well as customized workflows
External interface implementation
User-Experience & User-Interface Optimization
Software or cloud integration, updates and monitoring

More…

AI & Automation Tools:

Wall Thickness Measurement / Corrosion Detector [X-Ray]
Weld Defect Detector (in-field, manual welds) & Standard Validation [X-Ray]
Weld Defect Detector & Validator (inline, manufacturing) [X-Ray]
CT & X-Ray Image Quality Enhancement / Irradition time reduction
OCR / Text Recognition [X-Ray & RGB]
Inline Surface Inspection [X-Ray, RGB]
Aerial Drone Inspection [RGB]
Inspection Data Anonymization e.g. personal data [RGB]

More…

Hardware & Services with partners:

Inspection Hardware setup
Digitization of analog X-Ray archives
Setup of digital archives like DICONDE-compliant PACS
And more to make your inspection faster and more reliable…

Found or missing something?

Der Beitrag The AI-ready NDT infrastructure Guide. erschien zuerst auf sentin.

The NDT 4.0 Guide – Everything you need to know.

Wed, 24 Apr 2024 10:39:37 +0000

The industrial revolution is widely known. However, this article deals with the fourth revolution in non-destructive testing, the so-called NDT 4.0. It first discusses the development of industry and non-destructive testing and then some new technologies, as well as the advantages and use cases of NDT 4.0.

What is NDT 4.0?

The term “NDT 4.0” describes the fourth revolution in non-destructive testing. It relies on new technologies such as artificial intelligence, networked and decentralized systems (e.g. cloud) and other technical and digital solutions.

The NDT revolutions can be defined as follows:

NDT 1.0 – Tools (to sharpen the senses)
NDT 2.0 – Analog systems (for viewing within the components)
NDT 3.0 – Digital processing and automation
NDT 4.0 – Transparency of information, technical & digital aids and autonomous and decentralized decisions

It can be understood in a similar way to Industry 4.0. Its revolutions can be categorized as follows:

Industry 1.0 – Mechanical (steam, energy)
Industry 2.0 – Technical (electrical energy, mass production)
Industry 3.0 – Digital (data processing, microelectronics)
Industry 4.0 – Increasing digitalization, such as modern technologies and production as well as faster reactions to the sales market.

How do industry and NDT compare?

Industry and NDT are each divided into four revolutions. In the 1st and 2nd revolutions, there are still hardly any similarities. However, they are already very similar in the 3rd (digital stage). The NDT and NDE sectors and the industry are currently in the digital revolution. Both use networked systems that can make autonomous and decentralized decisions thanks to modern technologies and technical support.

What technologies does NDT 4.0 use?

NDT 4.0 is driven by many different technologies. It uses the following technologies, among others:

Big Data & Artificial Intelligence
Digital twins & simulations
Cloud computing & storage
Internet of Things (IoT)
5G
Blockchain
Augmented & virtual reality

Big data: Describes the increasing amounts of data that are available and analyzed with the help of modern systems. Companies that can use a longer history and a large amount of collected data are well prepared.

Artificial intelligence (AI): is a system that can interpret data and images, react to them and adapt. So-called neural networks are often used for this purpose. The subcategories of machine learning, which describes algorithms that improve with increasing data, and deep learning, which is based on neural networks and can improve with a lot of data even better, are particularly important for NDT 4.0.

You can find out more about AI in NDT here.

And you can find out more about AI-ready NDT infrastructure here.

Digital twins and simulations: On the basis of collected data, simulations can be used, for example, to make recommendations for action and predict decisions or events. The digital twin is then a digital entity that has exactly the same properties as the physical object. As a result, the behavior and condition of the real object can be precisely viewed and predicted. An AI can be used as the basis for such simulations.

Cloud Computing and Storage: Data can be securely stored in a cloud so that it can be accessed from any location. It consists of a distributed server network and offers redundant storage and a high level of compatibility thanks to modern web standards.

Internet of Things (IoT): All data, devices and sensors are connected with each other and can be accessed via a cloud. This can be a prerequisite for a good database and thus enable AI and digital twins. It is often abbreviated to IoT or IIoT for (Industrial) Internet of Things.

5G: Is a mobile data transmission technology and the successor to 4G/LTE and enables the connection of a large number of devices and ensures a robust real-time data connection. The large bandwidths enable the transmission of up to 10 Gbit/s, so that even large images or inspection results can be sent quickly.

Blockchain: With a secure blockchain, data can no longer be changed without being recognized, which improves traceability and trust. This can be interesting for archives, for example. Newer blockchain generations also make it possible to execute smart contracts, which are similar to smaller decentralized software programs.

Augmented / virtual reality: With the help of data glasses or intelligent displays, e.g. via the smartphone camera and display in real time, additional information and context-relevant representations can be shown to users. This can, for example, make it easier for inspectors to recognize the installation situation of components or certain work steps.

What advantages does NDT 4.0 offer?

The possibilities of NDT 4.0 create a number of advantages for user companies. The availability of lots of data, for example, makes it possible to use machine learning and artificial intelligence (AI). With the help of such AI systems, for example, the inspection of systems and components can be improved, thereby increasing efficiency.

In addition, the efficiency of inspections can be increased through modern connectivity and advanced computers. Further more the tools may also increasingly simplify data exchange and feedback between customers and clients. Such systems are more reliable and inspections are easier to trace.

The benefits of NDT 4.0 can be categorized and summarized as follows:

Improved NDT methods
Improved equipment
More control & traceability
Simplified data transfer
Better data evaluation
Improvement of the production process

Advantage 1: Improved NDT merhods

Relevant technologies:

Cloud computing & storage – Decentralized evaluation
Artificial Intelligence – Critical Image Detection, Automatic Defect Recognition, Batch Processing
Robotics

Impact & examples:

NDT 4.0 utilizes a wide range of new technologies that improve current NDT methods. These have a direct impact on the way work is done in NDT.

For example, cloud computing enables decentralized evaluation of recordings and data. For example, data can be recorded at one facility and then transferred to an evaluation station at another location.

Another example could be the use of artificial intelligence (AI), for example to process all newly captured images of an X-ray inspection as a so-called batch. With the help of critical image detection, no time is wasted on “good images”, but the critical images that have poor quality or indications, for example, are found.

AI can also help to automatically detect indications and defects in inspection data and on images (ADR – Automatic Defect Recognition).

Other applications can include, for example, automatic recognition and assignment of texts, e.g. for a measuring point, or pattern recognition to detect suspicious or fraudulent activities (e.g. an image used several times for different measuring points).

You can find out more about decentralized evaluation and AI here.

And you can find out more about AI-ready NDT infrastructure here.

Advantage 2: Improved Equipment

Relevant technologies:

Cloud computing & storage – digital archives, PACS, APIs
Artificial intelligence – super resolution, image enhancement
Robotics – crawlers, drones
New sensors & inspection technology
Augmented reality

Impact & examples:

The improvements to testing equipment as part of NDT 4.0 make it possible to test faster and more efficiently. Among other things, the improvements relate to the way in which humans and machines work together. The improved input and output of the devices can take the form of processed system and inspection information using augmented reality, automatic validation of results and meaningful error messages, for example.

Other methods can include data storage and processing, for example, so that the analog archive is replaced by a digital one. The so-called Picture Archiving & Communication Systems (PACS), ensure that the data is stored in an audit-proof manner. Together with interfaces such as DICOM/DICONDE or modern cloud APIs, the devices can be better connected.

You can find more on PACS here.

AI algorithms can also be used to achieve an improved representation of the recorded data. The “super-resolution” and “image enhancement” techniques are particularly interesting in this respect. For example, irradiation times can be reduced and more details can be found without having to accept greater image noise. A further consideration can be the improvement of the Probability Of Detection (POD), so that smaller and smaller defects can be found better.

You can find out more about POD here.

Other applications include the use of intelligent robots, so-called “crawlers” and drones or new sensors and inspection technology. These networked devices make it possible to inspect systems or components automatically in some cases, e.g. in the thermography of solar systems or ultrasonic testing of large pipes. This has the simple advantage that even measuring points that are difficult to reach or large areas can be inspected with little effort and without the need for a person to climb up somewhere, for example.

Advantage 3: More control & traceability

Relevant technologies:

Cloud computing & storage – Digital archives, PACS
Digital interfaces – DICONDE
Digital twin
Data analytics

Impact & examples:

In NDT, there are various measures and standards that must be taken into account when conducting an inspection. One example is the traceability, which determines whether the check was carried out in the right place and with which procedures/parameters and according to which instructions. The implementation of audit-proof data storage is therefore necessary. The use of digital solutions from NDT 4.0, such as improved archives on the cloud and PACS solutions for storage or blockchains to protect against tampering, can help to increase this traceability. You can also store data in a more structured way and make it searchable more quickly.

In addition, NDT 4.0 technologies and interfaces support the digital workflow by enabling the electronic transfer of inspection and order data. This enables paperless inspections and ensures greater overall control over processes and workflows. This is also necessary in order to derive new findings and recommendations for action with the help of data analytics and thus remain successful in the long term.

You can find more on PACS here.

And you can find out more about AI-ready NDT infrastructure here.

Advantage 4: Simplified data transfer

Relevant technologies:

Cloud computing & storage – Databases, APIs
Digital interfaces – DICONDE
5G

Impact & examples:

Networked systems and decentralized storage are important factors in NDT 4.0. Secure and fast transmission is necessary to make full use of these. Technologies such as 5G and modern interfaces make it possible to send data from the recording location to the evaluation station almost in real time and at the same time provide the manager with status information. User companies can benefit in particular from improved interoperability, allowing various solutions and devices to communicate with each other and thus enable more efficient testing.

Advantage 5: Better data evaluation

Relevant technologies:

Artificial intelligence
Data analytics
Big Data

Impact & examples:

NDT 4.0-supported understanding & transparency of data is a valuable competitive advantage. With structured data formats and a sufficient history, it is possible, for example, to make predictions for the future, create AI models and digital twins and optimize business processes. The challenge in NDT lies in the complexity of the inspection processes. With the help of data analytics, new insights can be gained and more efficient processes can be designed. For example, more precise service life calculations for components, predictive maintenance and reliability technologies are well known.

Advantage 6: Improvement of the production process

Relevant technologies:

Artificial intelligence
data analytics
Digital twin
New sensors & testing technology

Impact & examples:

NDT 4.0 also offers potential for improvement in production that goes beyond the traditional understanding of Industry 4.0 (such as predictive maintenance). One application, for example, could be faster and better quality assurance. If components can be inspected more quickly in production, for example, because an AI with super resolution is used for radiographic testing, this increases the efficiency of the production process from within NDT. In industry, NDT is often reduced to visual inspections with cameras or as a visual inspection by humans, but this falls short. In addition, digital archive systems in which the inspection data is stored provide an important basis for optimizing machine parameters in the production process, for example, based on the quality determined.

What are the challenges for NDT 4.0?

Despite all the benefits that NDT 4.0 brings, there are also some challenges. These challenges are primarily

Technical hurdles, data and infrastructure
Relevance of the skills learned
Hesitant managers
Insufficient understanding

Challenge 1: Technical hurdles, data and infrastructure

Although the new NDT 4.0 technologies mentioned above are developing rapidly, there are still some hurdles that are delaying their introduction into user companies. When it comes to artificial intelligence, for example, there are serious differences between consumer AIs, which are known for speech recognition and large language models such as ChatGPT or Google Bard, and industrial AIs, as the amount of data available in industry is much smaller. While millions of images of cats from the Internet can be used for consumer AIs, for example, the amount of public weld seam images for an industrial AI is close to zero. In addition, there is often no interest in making this data available to the competition.

Another factor can be supply bottlenecks or supply chains. Especially during the COVID-19 pandemic in 2020/21, a lack of semiconductors and graphics cards became a problem for the industry. Since the energy crisis in Europe in 2022/23, the cost of maintaining the digital infrastructure has also increased.

Furthermore, there is of course a difference between research or the innovation departments of tech giants such as Google or Facebook and application in the field and in practice. It is well known that it takes some time for technologies tested in the laboratory to meet real-world requirements and for the necessary infrastructure to be established in the company.

Challenge 2: Relevance of the skills learned

The tasks and technologies of NDT 4.0 differ from traditional NDT functions and systems. Therefore, organizations that want to use these new technologies need to develop new skills. They need to acquire skills in information and communication technologies (ICT) and be able to work with intelligent systems such as collaborative industrial robots and desktops.

In addition, the people who will interact with these new advanced systems should have a flexible mindset. They should be prepared to accept that the knowledge they have today may become obsolete before the technology is used in the field.

NDT 4.0 requires employers and employees to learn more and faster than the previous revolution. Employers and their employees must see learning as an investment.

The skills required will also differ at different hierarchical levels. The evaluators in the field, for example, will need to receive training on new equipment. The managers will need to properly understand and manage the new digital processes, while corporate management will need to develop new business models. It will be crucial for them all to have the ability to learn and develop new skills faster than they were used to in previous innovations.

Challenge 3: Hesitant managers

Business leaders are often reluctant to introduce new innovations for a variety of reasons. One of these reasons is budgeting and the incentive for annual investment.

Companies seek short-term visible gains at the expense of greater and long-term intangible benefits. So, considering that NDT 4.0 is a long-term investment that requires higher technological investments and skills development, it may not be attractive for companies that are used to focusing on short-term gains.

Secondly, in most cases, people are extremely risk averse. They tend to focus on the uncertainties and problems, underestimating opportunities and potential. However, recent years have shown, both in industry and in NDT, that new technologies offer real added value and have arrived in practice. Companies that have invested early have been able to secure competitive advantages and overcome current challenges such as demographic change and the shortage of skilled workers.

NDT 4.0 therefore requires a new management style – Leadership 4.0, so to speak – flexible and digitally competent managers who know how to assess and use the new technologies with confidence.

Challenge 4: Insufficient understanding

NDT 4.0 offers many opportunities for inspection service providers, system operators and many other market participants. However, there is often little understanding of the technologies and a rather negative mentality. We occasionally hear that these technologies are not needed or that they are nothing special.

It is only when the potential of the technology is demonstrated and enriched with practical examples that it becomes clear how it can be used in your own company. At the latest with the large language models such as ChatGPT or Google Bard, artificial intelligence has at least reached the general public. For the NDT, the aim is to improve the understanding of auditors and managers through presentations, publications and training courses.

Furthermore, national associations such as the American, German or Indian Society for Non-Destructive Testing (ASNT, DGZfP, ISNT) and their working groups offer various training and further education opportunities.

We also offer management courses for AI and NDT 4.0.

What do experts say about NDT 4.0?

"Working with sentin, I realised that modern technologies such as artificial intelligence have arrived in the industry. Especially the combination of conventional and AI algorithms can offer real added value for digital radiography and NDT."

Dr. Uwe Ewert
Expert for (Digital) Radiography

Professor and Director off duty
Chairman of the Technical Committee “Radiology” at the DGZfP for many years
as well as former head of division at BAM and active at DGZfP
Member of various standardisation committees of ISO, CEN and ASTM
received the Berthold Award of the DGZfP (2005), the Röntgen Medal of the city of Remscheid (2009) and the Briggs Award of ASTM-International (2010).

How can I use NDT 4.0 today?

NDT 4.0 has already arrived in practice. Although it will certainly take some time before all companies are working entirely according to the principles, it is already possible to use some of the technologies presented.

Artificial Intelligence

Wall thickness measurements or corrosion – sentin GmbH

Weld seam defects – sentin GmbH

Text recognition / OCR – sentin GmbH

Duplicates, fraud and manipulation detection – sentin GmbH

Test data anonymization – sentin GmbH

Digital Tools and Workflows

Cloud & Infrastructure

Infrastructure & Integration Services – sentin GmbH

Automation solutions – sentin GmbH

Digital examination management – sentin GmbH

Examination archives and PACS – sentin GmbH & BW Plus NDT

Other

Digital Archive – BW PLUS PACS
NDT (Order) Management Software – DriveNDT
Digital Twins – PipelineSentry
Crawler / Robots – HausBots

Sources and further links

Literature:

Der Beitrag The NDT 4.0 Guide – Everything you need to know. erschien zuerst auf sentin.

What is a PACS in NDT?

Fri, 26 Aug 2022 09:35:50 +0000

It is no longer possible to imagine non-destructive testing (NDT) without PACS. What has been standard in medical technology for years (under the keyword – DICOM) is now also increasingly used in digital NDT (DICONDE) to securely archive digital images, e.g. from radiographic examinations, and evaluation results.

What does PACS stand for in NDT?

The abbreviation PACS stands for Picture Archiving & Communication System. It serves as a secure and central data storage for inspection images and signals from various sources such as RT scanners or CT systems.

For this purpose, a server system is used as a kind of database, which can then accept requests from evaluation workstations. A separate viewer (display software) can then be used by the evaluators on their own computers to manage the studies/test orders and perform the evaluation of the images.

A PACS can be part of an AI-ready NDT infrastructure and support the inspection workflows with convenient interfaces and fast / reliable data access.

Learn more about an AI-ready NDT infrastructure here.

What are the advantages of a PACS?

A PACS is the basis for digital evaluation and secure data storage in NDT. In addition to the advantages of digital solutions, a PACS also offers further arguments that can accelerate your processes and make them future-proof:

A PACS stores data in an audit-proof and traceable manner.
It can be used decentrally, so that you can access it from anywhere in the world – for example, remote examinations and new processes can be established.
You save storage costs for films and reports because they can now be stored digitally. New scanners can digitize your old images in minutes.
You can use state-of-the-art image processing to get the best out of your images. Various filters and artificial intelligence can help you examine even more accurately and quickly.
A PACS can also be connected to your ERP system and thus seamlessly integrated into your processes. This allows you to plan and manage inspection jobs and store the corresponding data in an organized and linked manner.
Modern software solutions also allow you to quickly receive feedback and updates if you need help – keyword: remote maintenance.

Learn more about benefits of modern technologies in NDT:

“The NDT 4.0 Guide – Everything you need to know.”

What interfaces does a PACS have?

A PACS is a type of database and should therefore support the common CRUD operations (Create, Read, Update, Delete).

A PACS system uses DICONDE (Digital Imaging and Communication in Non Destructive Evaluation), which is similar to DICOM (Digital Imaging and Communications in Medicine).

What is the DICOM or DICONDE format?

A so-called DICOM/DICONDE container is used to store image & metadata. So-called studies are created, which correspond to an inspection order. These can contain several series, e.g. different components of a pipeline. In the series again are different instances which can contain a certain picture. As meta-data then e.g. recording time, devices hardware, tester etc. can be stored.

Further information can be found in the documentation / blog of AWS here.

What are DICOM services?

The best known interface of a PACS is represented by the DICOM Services, which are implemented using the DIMSE-C protocol. The DICOM Services include the following actions:

– C-STORE – storage of composite SOP instance information.

– C-FIND – search using a set of attribute strings. For each match, the C-FIND service returns a list of requested attributes and their values.

– C-MOVE – Move information for one or more composite SOP instances from one user to a third user.

– C-GET – Retrieve information for one or more composite SOP instances based on the attributes provided by the calling DIMSE service user.

– C-ECHO – Verifying an end-to-end communication with a peer DIMSE service user.

Some developers have created custom software libraries for this communication to make it available using modern programming languages such as Python. Probably the best known package for this is pydicom.

You can read more about this standard here: 7.5 DIMSE Services

What is DICOMWeb?

With the increasing use of web technologies, a web standard for PACS has been implemented. This works via HTTP(S) and JSON standard and can be accessed more easily with modern frameworks like Node.js or Python. There are 3 methods that map the DICOM services (C-STORE, GET & FIND) to it.

– QIDO-RS – analogous to C-FIND- for querying collections of DICOM objects & meta-data (not images) – view documentation

– STOW-RS – analogous to C-STORE- for storing DICOM files or separate metadata and bulk data – view documentation

– WADO-RS – analogous to C-GET- for retrieving DICOM files, metadata in XML or JSON form, bulk data separated from metadata and rendered images – view documentation

For example, using the QIDO interface, you can serach for studies that were created on a certain date or match a certain hardware. Then using the WADO interface, retrieve the corresponding instances and image data. Evaluate these and transmit them back to the PACS using the STOW interface.

How do you set up a PACS?

A PACS can be easily integrated into your existing infrastructure. It is best to involve your IT department in the integration process and have an experienced provider advise you on which specifications you should meet. Basically, you can proceed in the following way:

Set up PACS server – often a separate computer is used to run the database software of the PACS. This can be e.g. a Windows system or a virtual machine / container solution.
Set up network – The PACS must then be made available on the network so that the other devices and software solutions can connect to it. Find out beforehand whether, for example, certain firewall settings are necessary.
Connect evaluation computers – The next step is to connect evaluation computers to the PACS. A viewer solution is then used to manage the archive and to evaluate images.
Connect acquisition hardware – Once you have made your PACS archive available, you can also connect your scanners and equipment to it. Now you can, for example, send orders from a computer to the hardware, which then automatically links studies/orders to images and stores them in the PACS.
Digitize existing archives – It can also be useful to digitize historical images. This allows you to archive your old films in an audit-proof manner and, if necessary, clean up your archive basement and free more space.
Connect other software – If your PACS has the right interfaces, you can also connect other software solutions that, for example, create reports or build a bridge to your ERP system.

Is a PACS future-proof?

A PACS is the foundation for a digitized NDT environment. It is the first step towards making your processes sustainable and future-proof. Just as computers and the Internet have permanently changed our working world, there are promising technologies that we will use as tools in the future.

One of these technologies is, for example, Artificial Intelligence. But this technology needs digital and sorted data to learn from. A PACS can ensure that your NDT images/data meet this standard.

So, with the help of these solutions, you can address today’s and tomorrow’s problems, such as demographic change and skills shortages.

Learn more about AI in NDT in our article:

“How does Artificial Intelligence (AI) work in NDT?”.

Free checklist
with 24 questions.

Is your company ready for AI & Machine Vision projects?

Where can I buy a PACS (for NDT)?

For the commercial operation of a PACS in NDT we recommend competent partners from the NDT industry. We therefore cooperate with the German company BWPlus NDT to offer you a suitable total package of PACS, hardware & software. Also covered are maintenance, support and updates, which are crucial for a smooth commercial use.

For research projects and development, there are a few open source projects from medical technology that can be used, for example, with Docker containers or software images. Here you can find an overview of 10 open source PACS solutions from medical technology.

Which viewer & evaluation softwares are available?

Many hardware manufacturers offer separate software for their imaging devices that can access the PACS and with which the evaluation can be performed. However, these are usually only optimized for one manufacturer and may not offer state-of-the-art methods.

We have developed sentin EXPLORER so that you can perform your evaluations independently of the hardware and also store your data in the PACS in a future-proof manner. This supports various filter and display options for your images as well as the management of studies/orders in the PACS. It is compatible with various interfaces and other systems. It is also ready to work with artificial intelligence and thus ideal for your digital work processes.

Der Beitrag What is a PACS in NDT? erschien zuerst auf sentin.

How does Artificial Intelligence (AI) in NDT work?

Tue, 26 Jul 2022 08:54:54 +0000

Industry 4.0 refers to the digitization and networking of industrial production after the three great industrial revolutions. Non-Destructive Testing and Evaluation 4.0 (NDT & NDE) mainly depend on improvements of NDT/E equipment. At this point, there is a special relationship between progress in industry 4.0 and NDT 4.0. One technology which influences both is Artificial Intelligence (AI), which in this application focuses on image processing and evaluation.

Probability of Detection (POD) is a common metric for NDT that helps to establish compliance with standards for inspection quality. This article discusses the contribution of these new technological developments in the NDT process, to improve the Probability of Detection (POD) by making use of the digitalization brought by the industrial revolution.

Keywords: Probability of Detection, Artificial Intelligence, Discontinuity, Defects, True Positive, False Positive, Non-Destructive Testing, Non-Destructive Evaluation, NDT 4.0, NDE 4.0, Industry 4.0, Industrial Revolution, Automatic Defect Recognition, Image Enhancement, Denoising.

It is based on the Paper “How can NDT 4.0 improve the Probability of Detection (POD)?” in The e-Journal of Nondestructive Testing – ISSN 1435-4934

What is Artificial Intelligence (AI)?

Artificial Intelligence (AI) is the term used to describe programs that can perceive, understand, react and adapt their behavior to the situation. The term is often used as a synonym for Machine or Deep Learning and is seen as a contrast to rule-based programming.

An AI model can be regarded as an evaluation software. In the case of Deep Learning, it is a so-called network of evaluation functions, which can thus learn arbitrarily complex relationships. The parameters of the evaluation functions, the network architecture and the interface for querying evaluations then represent a model that can be used elsewhere, e.g. in another software or plant.

AI is one of the most important technologies in NDT 4.0.

If you want to learn more about AI in general checkout our FAQ.

Where can Artificial Intelligence (AI) be used for NDT 4.0?

In the context of NDT 4.0, many new and old areas of application for AI can be found along the NDT value chain. These include, for example

Data analytics – insights into order and test data
Digital twins – modeling and predictions
Use in the evaluation / inspection process – Improved equipment / inspection methods
- Critical Item / Image Detection (CID)
- Single Item / Image Detail Analysis (SIDA)
- Denoising, Superresolution, Data & Image Enhancement
- Automatic Defect Recognition (ADR)

You can find out more about NDT 4.0 in our guide.

Use case 1: Data Analytics

Artificial intelligence can be used in the area of order and inspection data to gain new insights and generally create transparency. Data analytics in NDT can be used to analyze test sequences and processes more precisely. However, this requires sufficient digital data.

Sales increase & cost savings:

One area of application is the analysis of customer, order and supplier data. Here, sales increases and cost savings can be realized by using AI to uncover patterns that arise in orders, for example. It is conceivable, for example, that many individual orders for testing equipment or consumables (such as films or chemicals) from different company divisions could benefit from discounts and economies of scale in bulk orders.

Time eaters in the inspection process

Another application example is the detection of bottlenecks in the inspection process. These can be varied and are very individual for each company. One case that occurs frequently, however, concerns constant media breaks in the workflow. These result, for example, in things being printed out several times and then digitized. Ideally, the inspection should run as digitally as possible via a networked infrastructure.

Learn more about an AI-ready NDT infrastructure here.

Critical measuring points / infrastructure / equipment

In addition to the procedures in the testing process, AI and data analytics can also be applied to test objects and other items. For example, pattern recognition can reveal that a certain piece of test equipment lasts twice as long on average as another. Or that a certain measurement configuration in the system constantly needs to be repaired. Or that a certain system, although it is much smaller than others, still takes much longer to test.

Use Case 2: Digital Twin

The use of digital twins is a core technology of NDT 4.0. On the basis of collected data, simulations or models can be used, for example, to make recommendations for action and predict decisions or events. The digital twin is then a digital entity that has exactly the same properties as the mapped element. As a result, the behavior and condition of the real object can be precisely viewed and predicted.

Artificial intelligence can be used as the basis for such models. Examples can be

Simulation of infrastructure, e.g. corrosion over time
Digital prototypes and testing in production, e.g. cast parts
Project planning, e.g. of pipelines
Prediction of maintenance (predictive maintenance)

Use case 3: In the evaluation / inspection process

The direct use of artificial intelligence in the evaluation and testing process offers some potential for improving NDT equipment and methods. Some examples of image-based testing (e.g. radiographic testing) are presented below, as these are easy to understand. However, there are also applications for other inspection methods such as ultrasonic testing. Many aspects of the approaches presented are transferable to other methods.

Critical Item / Image Detection (CID)

In NDT, there is often not just one test object that needs to be examined. Evaluators are often involved with a large number of images, e.g. in radiographic testing of weld seams or corrosion testing. Many of the images are without complaints or indications and only a fraction is particularly critical.

With the help of artificial intelligence and the parallel processing of several recordings (so-called “batch processing“), recordings can be pre-sorted for evaluation.

Critical item detection can be used for non-image-based inspections, e.g. for classic ultrasound or other sensor data, in order to process these in parallel.

Critical image Detection, on the other hand, offers the option of processing image-based inspection such as radiographic testing or image recordings from visual or (dye) penetrant testing as a batch.

The aim is to determine whether a recording is critical or not.

This involves finding recordings that, for example

Have poor quality
contain film errors or blemishes
Duplicates or manipulated recordings
Missing IQIs or identifications
Show indications of possible defects

Single Item / Image Defect Analysis (SIDA)

In addition to Critical Item Detection or Critical Image Detection (CID – batch processing), individual images can also be processed in detail with AI. The aim here is not only to determine whether an image is critical, but also why and where.

In single item detail analysis or single image detail analysis (SIDA), the image is shown to one or more AI models and the results are processed visually, e.g. in the image, and stored in the data container, e.g. DICONDE.

Single item detail analysis can be used for non-image-based inspections, e.g. for classic ultrasound or other sensor data, to evaluate a measuring point in detail.

Single Image Detail Analysis, on the other hand, offers the possibility of checking image-based inspections such as radiographic testing or image recordings of visual or dye penetrant inspections in detail in a similar way to a human being.

Entire test sequences and work steps can be taken into account. The combination of different AI models and algorithms helps to make evaluation easier for the evaluator.

Images are examined for the following parameters, for example:

Film defects or blemishes in the area to be evaluated
Text recognition for identification
IQI evaluation
Comparison with test parameters
Wall thickness measurements and wall profiles
Defect detection of e.g. weld seams (ADR)
Standard validation, e.g. for defect sizes
Recognition of test configurations
Application of further algorithms, e.g. denoising, superresolution, data & image enhancement

Denoising, Superresolution, Data & Image Enhancement

Until an image is displayed on a monitor or a film, several environmental factors can lead to a distortion of certain parts or a general noise overlaying the whole image. There are many causes for noise, which does not allow for rule-based enhancement. One example of noise is salt and pepper noise, where singular pixels are either fully black or white. Another example is gaussian noise, which is distributed uniformly over the picture. The standard method for estimating noise is the signal-to-noise ratio (SNR), which is also mentioned in several standards for weld image inspection.

The conventional approach for denoising is to use static filters (like low/high pass), which are applied to the image as a convolution. The main disadvantage of these filters is the missing spatial resolution since they treat all parts of the image in the same way. In this regard, modern AI algorithms can interpret information locally and create and combine several hundreds to thousands of filters, where each filter can concentrate on a different feature in an image. Doing this manually would mean necessitate dividing the image into multiple patches (e.g., 16×16 pixels), and optimizing each view based on your knowledge and the position of the patch in the image and putting everything back together.

These types of AI models are used heavily in photography to enhance low resolution images, correct lightning or when applying filters. A topic we focus on it the development of these AI algorithms for application in digital radiography/radioscopy.

For example, noise in weld images is a major factor when using national standards. Therefore, a denoising is useful to provide a clear representation of the defects. This improves the detection rate of a human observers, as well as time spent on one image. An example can be seen in Figure 2, where a noisy weld image (top) with several artifacts and different noise granularity is shown. If an AI could reconstruct the image and react on different noise types automatically it could transform it to a less noisy image (bottom) and keep major image details. The advantage of such an AI algorithm is that local features (e.g., defects, weld edges, background, etc.) can be treated differently based on their spatial occurrence.

For training such an AI you would need to have data with little noise, showing the algorithm examples of good images – e.g. by performing some shots with long exposure time. Afterwards you can put noisy images into the model and it will output the denoised images. (Disclaimer: Figure 2 is an exaggerated example, because latest results must not be shown publicly due to non-disclosure agreements. It does not represent the current state of the technology. Nevertheless, it shall visualize the potential of an AI algorithm.)

Other than denoising, Figure 3 shows an example of image enhancement, where a crack in a weld can be observed. Depending on the value range that is presented to the human observer, different depths can be seen. Thus, an observer often needs to take two or three views to fully see all defects in all depths. An AI algorithm can help to enhance every single defect in a way that a human observer can see all defects in one image. This saves time and improves the detection rate or POD of a human observer. For practical usage this also means that you can use images that are taken with imperfect conditions such as low exposure time, while maintaining valid image quality with post processing. (Disclaimer: Figure 3 is an exaggerated example, because latest results must not be shown publicly due to non-disclosure agreements. It does not represent the current state of the technology. Nevertheless, it shall visualize the potential of an AI algorithm.)

(Automatic) Defect Recognition - ADR

In addition to the areas of application presented above, AI is also capable of accurately detecting errors or defects in an image. Although this detection is not yet the standard in NDT, it is another tool that helps human evaluators to recognize faster and more precisely. In addition, the AI is stable and is not affected by factors such as fatigue.

For defect recognition, a supervised AI is used. This AI is trained based on given examples of images with its correct labels. In supervised vision, three major types of AI exist. They are image classification, object detection and instance segmentation. The first one returns one label per image (defect or no defect). This type of AI is only suitable for use cases where only one object is present per image and if the occurrence of this defect implies a rejection of the entire sample (cracks in most standards).

Object detection draws a bounding box around a defect. This is useful, when several different defects can occur, but a human observer needs to decide whether the part is scrap or not. The last one, instance segmentation, is a prediction of the exact polygon of a defect. Although this algorithm needs the most data, it is the perfect for quantifying defect. E.g. a lot of standards for weld images need you to measure defects and ensure that these are below a given threshold in size. Those geometric coherences can be calculated for detected defects and autonomously reject samples.

Figure 4 shows an example of a weld image with porosity. Using a different AI algorithm, that can detect defects or discontinuities, it can be displayed and highlighted like the right part of the image. Although defect recognition detects defects directly, image denoising and image enhancement are often useful as a pre-processing step to make an algorithm more robust and reduce the number of images needed for AI Model training. Thus, both approaches for AI work well together to help the evaluator do their best work.

Even though training of those algorithms is time consuming, the execution of a trained algorithm is extremely fast. For images with a normal size (e.g., 1,200 x 1,800 pixels), the evaluation times of under 50 milliseconds are to be expected – depending on the hardware used.

What is the current performance of NDT (without AI)?

In Data Science and Artificial Intelligence, metrics are important when evaluating the result of a given method. For a binary decision problem, the most prominent metric is the receiver-operator characteristic. It describes four possible states of a prediction. For example, an algorithm or procedure should detect whether a component has defects or not.

This is a binary decision because it either rejects a sample or not. If the algorithm correctly rejects a sample, it is called true positive and if it is does not reject the sample and the sample has no defects it is called true negative. However, if the algorithm rejects the sample although the component has no defects, it is called false positive. The opposite (component has flaws, but the algorithm does not reject the sample) is called false negative.

In NDT the performance/accuracy for (true-positives and true-negatives) of the the best NDT evaluators is estimated to reach max. 85,2 % POD with 2,1 % PFA, “[…] but the overall results are worse with high false alarm rate.” [3]

What is the Probability of Detection (POD)?

Based on the above classifications we can carry out a test where we give the algorithm, examiner or procedure several samples (e.g. 1000) that are known to have defects or are defect free. The samples are then evaluated and it is decided if they have defects or not. Since we know how the examiner or procedure should perform with this sample set, we can determine how they performed according to the four performance classifications above.

With this method we can analyze data in many ways. A common metric for NDT is the Probability of Detection (POD), which is simply defined by:

POD = True Positive / (True Positive + False Positive)

Based on POD, an algorithm can be evaluated based on a different factor. For example, the figure below shows POD values in context of the flaw size for a given component. The probability runs asymptotically against 100%. At the marked point the POD is 90% for a flaw size of 22mm. In the NDT field POD is often interpreted as POD in dependence of flaw size.

If you want to learn more about this check out our article:

What is the POD – Probability Of Detection in NDT?

How can Artificial Intelligence reduce human factors in NDT?

Industry 4.0 is adopting a wide array of emerging technologies that enhance current NDT methods, resulting in NDT 4.0. Cloud Computing and Storage, Artificial Intelligence, Robotics, Big Data, Blockchain as well as Augmented Reality are among the emerging technologies of the fourth industrial revolution.

Defects are bound to happen in any manufacturing process. Manual part inspection and complex statistical methods (six-sigma) are currently required to have a sufficiently high probability of detection (POD) of known defects. This is especially time intensive, when testing welds e.g. in chemical plants, where weld seams are manually inspected using X-Rays. AI can be used to minimize human errors and helps inspectors find defects.

People looking at the same information may form different opinions. Even the opinions that one person forms may be different based on human factors such as fatigue. NDT is a task that requires a high level of attention. The inspector’s attention and the performance of the inspection system are factors that should be considered when determining the POD. Personality, past work experience, training and fatigue of the evaluator performing this task are among the most important human factors affecting the examination process. The inspector’s evaluation also depends on their perception and their ability to process and remember information. Additional problems are caused by the insufficient quality of the signals or images used. An incorrect classification may result in rejection of a part in without defects (false negative) or acceptance of a defective part that exceeds the limit defined by the relevant standards (false positive).

Figure 5 is showing an example of improved PODs. The intrinsic capability is the “best” performance a system can achieve with a small defect size that can be detected with a high probability (90%). The performance gap is the difference between what your system could possibly detect and what it really detects. If an AI can help to detect things earlier/better (smaller defects) the curve will move from the right closer to the best performance reducing the performance gap. [2]

When we are talking about improving the POD we want to detect smaller defects with a higher probability. We can achieve this by reducing disturbing factors. The next lines focus on factors that can be reduced after image acquisition so that companies can use existing equipment/systems just upgrading the image processing with a software.

Example 1: Improving Reliability & Perception with ADR

For a NDT process to be reliable, the entire system of equipment, procedures, and inspectors must be reliable. A technology that can boost perception and reduce the human factors affecting POD improves reliability. That is where Artificial Intelligence can help. AI improves the perception of the inspector by automatically labeling possible defects using Automatic Defect Recognition (ADR). This decreases the human factors affecting the POD like fatigue, motivation or attentiveness. AI also provides an objective second opinion, which may help less experienced inspectors and helps to ensure that inspectors are classifying flaws consistently.

Example 2: Improving Perception & Attentiveness with Image Enhancement.

Another problem affecting the perception and motivation of inspectors and therefore the POD is the image quality. If you can optimize the contrasts and reduce the noise in images in a sophisticated way, then the inspector does not have to choose filters manually and locally, they can focus on the actual interpretation process. If the whole image displays all information in an optimized way, choosing the right enhancement for the right spot, the inspector does not have to choose the right filters manually. This can increase the attentiveness and may help less experienced inspectors observe every detail of the image – finding smaller defects. For the actual image acquisition, AI may also help. Images can be taken faster or with less voltage because the image quality can be increased in detail afterwards overcoming capabilities of classical high-pass filters.

Free checklist
with 24 questions.

Is your company ready for AI & Machine Vision projects?

What sentin does with AI in NDT?

The different AI use cases like data analysis and digital twin, as well as CID, SIDA and ADR and Image Enhancement therefore help to evaluate images & discontinuities as quickly and accurately as possible in a cost-effective way. It improves the detection and classification process of defects, ensuring a high POD, good product quality and application reliability. At the moment the ADR technology is mostly used as an assistance system including a user interface – to meet the standards for human interpretation.

In order to save time and nerves during examination and evaluation, having the right tool for the job is important. We are always working on the integration of the technologies mentioned above into our software, the sentin EXPLORER. Using existing data and tools, you could get a customized software for your evaluation processes without having to be an expert in AI.

In our case the software is based on modern operating concepts from the field of web technologies. An important advantage here is flexible integration into all operating systems. This technology can be integrated via software as a service either as a standalone product for evaluating image files or in digital NDT image management software.

The software-based evaluation of the images decouples the process of recording and evaluation. Using ADR and Image Enhancement/Denoising it supports the examiner in localizing and classifying errors. In time critical applications, presorting can also be carried out so that images
with the most defects can be analyzed first, so that repair work can be initiated with sufficient lead time.

Learn more about an AI-ready NDT infrastructure here.

We want to give you the opportunity to adjust your recognition or image enhancement models as precisely as you want, to get the best results from evaluations and boost your POD. We want your testing to be fast, objective and precise.

References

[1] Vrana, J., Singh, R. (2020). – The NDE 4.0: Key Challenges, Use Cases, and Adaption. DOI: 10.1007/s10921-020-00735-9

[2] Vrana, J., Singh, R. (2020). – The WHY of NDE 4.0 @ Ripi Singh

[3] Fücsöck, F., Müller, C., Scharmach, M. (2005). Measuring of Reliability of NDE

[4] Bertovic, M. (2015) Human Factors in Non-Destructive Testing (NDT): Risks and Challenges of Mechanised NDT DOI: 10.14279/depositonce-4685

Der Beitrag How does Artificial Intelligence (AI) in NDT work? erschien zuerst auf sentin.

One Day as A NDT-Inspector – Work and Life in NDT

Fri, 29 Oct 2021 07:36:00 +0000

Working as a non-destructive test (NDT) inspector is both a rewarding and challenging career. The inspector needs to be highly attentive to detail and fairly technologically savvy. Besides, considering that NDT is a highly dynamic field, thanks to emerging technological advancements, the technician needs to stay in the loop with the emerging trends and information to stay relevant and marketable.

Anyone aspiring to be an NDT inspector would be interested in having a sneak peek of the kind of life they’re likely to have when the career dream comes true. They would like to know how an ordinary day of the life of a non-destructive test inspector looks like.

If you want to learn more about non-destructive testing, be sure to check out our other articles:

The Daily Tasks of an NDT-Inspector

The main work of a non-destructive test technician is to perform non-destructive tests on materials, systems and/or components to detect defects that are likely to have adverse effects on that the functionality and safety of the item being assessed.

These tests depend on the test method, standard and qualification level, as not all inspectors are allowed to do the evaluation of images taken, for example.

On that note, here are some of the key tasks that the inspector could be engaged in on a daily basis:

Examining systems, materials or structures such as bridges, pipelines, aircrafts and plant machines using NDT techniques like acoustic emission, radiographic, electromagnetic, liquid penetrant, magnetic particle, neutron radiographic, thermal, ultrasonic or vibration analysis.
Collecting Data for flaw detection in materials and objects.
Choosing, calibrating or operating NDT equipment.

Equipment maintenance. After using NDT equipment or machinery, the technician is supposed to ensure that the it is cleaned up and ready for use the following day, or whenever necessary thereafter.
Evaluating the results of non-destructive tests with adherence to the relevant standards, guidelines or specifications.

Interpreting NDT results according to the applicable standards, guidelines or specifications. The interpretation could involve examining film produced by X-ray imaging, translating raw test data to easy-to-understand information to establish if it’s necessary to carry out further tests or whether the structure, object or material being tested ought to be done away with immediately.
Preparing test reports. The technician is expected to record the activities and findings of the tests they perform on a daily basis in the form of a report. In the report, the NDT expert should outline the procedures they followed during an NDT test, the test results, and what they recommend the engineers or other people who are responsible for the integrity of the tested item or structure to do.
Report presentation. After compiling the test reports, they are supposed to submit or present them to the relevant experts or management for further action or to ascertain that all is well with the tested object or material.

Communication

As earlier mentioned, the inspector is expected to write reports and make verbal communications effectively. As such, excellent written and verbal communication skills are an integral part of the work of the technician.

Teamwork

In most cases, non-destructive test activities and procedures require teamwork, as different aspects of the work are under different individuals who have to work together to achieve a common goal. No wonder, according to OwlGuru, approximately 75% of NDT inspectors to have group discussions at their workplace every day.

Work Hours and Compensation

According to Zip Recruiter, the hourly wages for an NDT inspector can range from $11.78 to $48.80, depending on the location of the technician. Considering the technician can work for eight hours a day, they can earn up to ~$390 dollars per day.

Using Artificial Intelligence to support Inspectors

Such a working day of an inspector, described above, tugs very hard at a person’s concentration as well as performance. This means that in the course of a shift, the concentration decreases considerably and thus the interpretation becomes more challenging. To prevent this, the inspector can be supported with software that simplifies and accompanies the entire process. Such a software is the sentin EXOLRER.

The sentin EXPLORER is a tool to automate such inspection or image evaluation tasks. Our customers sometimes produce thousands of images per week, all of which have to be reliably checked. Therefore we give you the power to perform evaluations with the help of artificial intelligence and to find defects more reliably or gain new insights.

Der Beitrag One Day as A NDT-Inspector – Work and Life in NDT erschien zuerst auf sentin.

History of NDT – 5 Interesting Facts

Fri, 08 Oct 2021 11:00:00 +0000

Non-destructive testing (NDT) has been an integral part of modern science and technology. It has provided an array of techniques to detect and analyze defects in components, materials and systems or without causing further damages. The testing method, which boasts a market size value of USD 14.6 billion, has gone through a great deal of evolution over the centuries. Therefore, the history of NDT is long and can occupy thousands of pages if it was to be written down entirely. To give a quick glance into the history, this article handpicks the five most interesting facts of the history.

If you want to learn more about non-destructive testing, be sure to check out our other articles:

People begin to recognize NDT as a technology

1920s and before

By the time the expression NDT was invented in 1920s, the ancient people had already been using techniques that followed the principles of nondestructive testing for centuries. For instance, they could analyze objects with their naked eyes to identify the presence of visible surface flaws without causing damages to the objects. Blacksmiths used to listen to different metals when shaping metal to ensure that they applied the right force to the right metal to produce the desired shapes.

World War II - Catalyst for NDT evolution

1930s and 1940s

The global war, that occurred between 1939 to 1945 led to the invention of the NDT technologies and devices that are still applicable in modern day science and technology.

For instance, the aircrafts that were used during the war were made of light non-magnetic metals, necessitating a different testing approach to replace magnetic testing. The need resulted in penetrant testing using fluorescents and dye penetrants. 

Penetrant Testing - application of the contraster

Karl Deutsch Prüf- und Messgerätebau GmbH + Co KG https://www.karldeutsch.de
Licenced under Creative Commons Attribution 3.0 Germany

Radiographic Testing

1940s and after

In 1895, Wilhelm Conrad Röntgen, the inventor of the x-ray technique, discovered “An Unknown Kind of Radiation” that could help in flaw detection. This discovery led to the development of an x-ray equipment that was majorly used in the medical field. However, ray had not foreseen the possible grave adverse effect that X-rays would have on human health. As a result, a lot of people who got exposed to the electromagnetic waves died.

To curb the deaths and other related health complications radiation protection clothing has been invented. Later, the U.S. X-ray and Radium Protection Advisory Committee introduced the first formal standard aimed at protecting humans from X-ray sources.

In the 1930s and 1940s early radiographic testing took place. Until today it is a common solution to find defects inside of materials and products.

Foundation of Eddy Current Testing - Hans Christian and Michael Faraday

1948 to 2012

Although Leon Foucault, a French physicist, discovered the phenomenon of eddy currents in 1851, this discovery was inspired by the discovery of electromagnetic induction by Michael Faraday, an English scientist in 1831, and electromagnetism discovery by Hans Christian in 1820.

During World War II the NDT method relying on Eddy Current has been further developed. In 1948 Friedrich Foerster in Germany founded a company to develope equipment based on this method.

Until today the market for Eddy Current Testing is growing and was estimated to have a size of 220$ million in 2012.

Eddy current testing is an NDT technique that uses principles of electromagnetic induction to detect surface and sub-surface flaws.

NDT applications keep on increasing

2010s and after

The global automation and robotics sector is growing 24/7, creating more demand for reciprocating processors to speed up workflows and cut down on labour costs. For instance, NDT sensors are integrated with industrial robots and cobots for damage-free automated inspection systems that are popularly used in production inspection as well as research and development. With a recent Robotics Business Review report showing that a total investment of US$1.16 billion were injected into the robotics sector in 2020, it’ll be reasonable to expect NDT applications to become even more popular and helpful going forward.

Non-destructive testing techniques have evolved for hundreds of years, and they are still getting better in terms of efficiency, safety and accuracy with time. The above five are among the interesting facts that you need to need to know about the history of NDT.

NDT 4.0 - Combines new technologies

2020s and after

Nowadays, NDT has developed strongly. It also could not avoid digitalization and software development. One of the most interesting points of contact is with Artificial Intelligence. An example is image analysis software that supports the inspectors or even automates the processes completely. One such software is the sentin EXPLORER.

This new field of NDT is called NDT 4.0 and uses several modern technologies to boost non-destructive testing. You can read more about it here:

Definition: What are NDT 4.0 & NDE 4.0?

Der Beitrag History of NDT – 5 Interesting Facts erschien zuerst auf sentin.

3 Facts about Digital Twins for Pipelines

Fri, 24 Sep 2021 08:33:00 +0000

The concept of the digital twin dates back to the 1950s, when NASA, GE and other industrial manufacturers began creating abstract digital models of assets to model their performance in simulations and maintain a record of the asset throughout its life. Over the years, more and more industries have adopted the digital twin paradigm to improve traceability, maintenance, and analysis, enabling better maintenance of the plant or equipment while reducing various risks identified during lifecycle management.

The following article is a guest article by the systems engineer James Wardrop. He is the founder of PipelineSentry, which enables the automation of technical processes by organizing, linking and providing context to technical data using digital twins.

These are the 3 Facts:

Not using all your data costs, you time and money – one example.
The key to a digital twin is in the linking data to an object at a location in time – one benefit is the ability to provide context.
It is as important to make a digital twin navigable by a human as it is by a machine – presenting data visually allows the human brain to digest and utilise it in decision making.

Here are some articles about:

1. Benefit

Reducing data costs, time and money

Physical properties for pipe joints vary from joint to joint (this is also true for the connections between joints). This is due to the manufacturing processes. Every pipe joint must exceed the minimum requirements if it is to become part of a pipeline. Material strengths for example can be up to 20% greater than the minimum requirement.

A digital twin allows you to utilise this extra strength in determining asset integrity. This is the same for all pipeline physical properties, by assuming the worst for every variable in every pipe joint it causes a huge stack up of incorrect assumptions. They say a chain is only as strong as its weakest link and this rings true for pipelines. A digital twin allows you to know where the weakest link is and act appropriately. One bad apple doesn’t have to spoil the bunch!

2. Benefit

The ability to provide context.

A digital twin for pipelines digitises and links data. Each data set is linked to an object at a location in time. By doing this you can automate any engineering integrity assessment you need and feed the results back into your model. By building a data schema with the integrity of a pipeline as the guiding force you can identify the relevant variables and eliminate the noise.

If you know the variables you can thread them through engineering equations automatically and create higher level data than an individual data point on its own. This allows an engineer to focus on areas of interest and perform detailed engineering rather than perform repetitive tasks. By organising data, you make integrity engineering hyper scalable. When you can do it for 1 case you can do it for “n” cases. (n = any number).

3. Benefit

Improved Decisions.

By Visualizing data.

It is as important to make a digital twin navigable by a human as it is by a machine – presenting data visually allows the human brain to digest and utilise it in decision making. By structuring and linking data in database using and entity relationship diagram not only do you make it easy for a machine to extract and manipulate data, but you can generate immersive 4D environments.

A digital twin links data to a physical location in time this is how the human brain has evolved over the millennia. Like hanging your car keys on a hook, you will always know where to find the right data. More importantly you will know where to put any new data… the digital twin becomes an attractor for more and more data. More importantly with data easily accessible it will enhance our decision making and make our assets operate safer and longer.

About James Wardrop and PipelineSentry

James Wardrop is an experienced System Engineer with a demonstrated history of working in the oil & energy industry. Skilled in Petroleum, Gas, Pipelines, Front End Engineering Design (FEED), Subsea Equipment Design and Engineering. He is a strong engineering professional with a Commerce and Engineering Degree (Materials Engineering, Corporate Finance) from The University of Western Australia. James Wardrop is the founder of Engtech Analytics, a company creating digital twins for pipelines. He is Leading technical and business development for a cloud based pipeline digital twin technology at PipelineSentry.

PipelineSentry creates digital twins for pipelines. Their team is passionate about increasing the efficiency of engineering work by organizing engineering, asset, operational and environmental data. With data organized efficiently they automate engineering processes which are otherwise labor intensive and time consuming, essentially becoming a force multiplier for an engineer.

Digital Twins in connection with KI

What is the sentin EXPLORER?

If you have a digital twin of your projects, an inspector has to evaluate these and find any problems and reasons for possible upcoming disasters. Therefore you can use supporting, or automatic softwares which are based on artificial intelligence. They evaluate the images and find any mistakes in the buildings. One AI based software is the sentin EXPLORER.

If you want to learn more: The 5 Types – What is Non-Destructive Testing (NDT)?

Der Beitrag 3 Facts about Digital Twins for Pipelines erschien zuerst auf sentin.

3 Failures and Catastrophes in NDT

Mon, 14 Jun 2021 07:06:22 +0000

Non-destructive testing (NDT) plays a significant role in ensuring that engineering systems such as bridges, nuclear plants, and airplanes, are in good shape and safe to use. This great input is attributed to the test methodology’s capability to detect and analyze welding defects and discontinuities in an engineering system or component without inflicting further damage to the original part or system. However, if the right procedures aren’t followed, NDT can lead to various failures and catastrophes. This article focuses on the hiccups and the potential subsequent impacts.

These are 3 types of failures and catastrophes in NDT:

NDT Unreliability
Failure Due to Incorrect Interpretation
The catastrophic impact of NDT Failures

Here are some articles about NDT and welding:

– Top 3 incredibl e Places for NDT (Non-Destructive Testing)

– Definition: What are NDT 4.0 & NDE 4.0?

– The 10 Common Weld Defects You Should Kn ow .

1. Danger | Using the wrong NDT Method

In some cases, a non-destructive test process may fail to detect some critical flaws in crucial engineering components. In some instances, human factors are to blame for the failures. For instance, lack of the necessary skills/experience could lead to poor inspection, whereby unacceptable defects go unnoticed.

The reliability of an NDT method is the test’s ability to detect critical defects in a system. When a test fails to detect critical defects with significant impact on the integrity of a structure, the undetected problems might escalate to cause structural failure. Consequently, the compromised structure might turn out to pose a safety and cause negative economic, environmental and reputation impact.

Axial Flaws in North Anna Steam Generator

For instance, in March 2012, an ultrasonic inspection, one of the NDT approaches, missed five significant axial flaws in North Anna steam generator. The test involved a site-specific mock-up for the qualification of the inspection procedure. An encoded passed array technique was used to scan for circumferential defects. However, even with all those keen approaches in place, some flaws were still missed.

Earthquakes or other environmental disasters can cause damage to a nuclear power plant or steam generator. These can lead to a serious disaster in the short or long term! Explosions or leaks then lead to widespread damage or even contaminated regions.

2. Danger | Failure Due to Incorrect Interpretation

In some cases, the NDT equipment/system might detect the flaws, but the failure arises when it comes to defect sentencing. Improper defect sentencing/interpretation simply means that the inspector failed to clarify the accurate picture of the significance of the flaw. For instance, an inspector can downplay the seriousness of a defect, terming it as so minute to cause any remarkable structural failure, thus not requiring any intervention. In most cases, the misinterpretation occurs when the assessor lacks sufficient levels of the necessary skills and experience. Additionally, external factors such as physical and social environment as well as motivation factors have a remarkable role to play for NDT accuracy.

Explosions

Explosions are also possible catastrophes of the non-destructive testing. Liquid Penetrant Testing employs a form of kerosene called Naphtha. The high-grade cleaner evaporates quickly and is very effective for pre-cleaning. However, high-grade Naphtha is flammable. Therefore, if not handled carefully, especially in confined spaces, the cleaning agent can cause an explosion. This can lead to massive property damages, physical injuries on people and loss of lives. Besides, prolonged exposure to Naphtha can result in respiratory issues.

Eddy current testing, another form of NDT, also has its fair share of potential dangers that can be catastrophic. The people involved running the test fail to shut down the system correctly when changing probes, they might experience serious electrical shock from the probe connection points. The mistake could also trigger sparks that might cause an explosion.

Nondestructive tests are instrumental in the maintenance of the integrity of various crucial mechanical systems and components. Nevertheless, failures resulting from human factors or inherent errors can occur if particular things aren’t well taken care of. Therefore, it’s important that inspectors apply various testing technologies and complementary techniques to ensure that no significant defect in a mechanical system or unit goes unnoticed on time. Besides, companies and individuals looking for NDT inspectors should ensure that they confirm that the experts they engage are sufficiently skilled and experienced. This would go a long way to preventing some failures that can be catastrophic. It’s also crucial to ensure that the testing process is done with utmost care to prevent exposure that can harm people and the external environment at large.

3. Danger | Wrong procedures or instructions.

NDT failures as a result of error of omission or misinterpretation of results can be catastrophic. For instance, if serious defects in a nuclear plant are overlooked in a non-destructive test process, highly radioactive nuclear waste can find its way to the environment. Plants and humans would be the key victims of the pollution, some losing their lives or sustaining long-term damages.

Potential Radiographic Testing (RT) Hazard

Radiographic testing (RT) is a component of non-destructive testing that utilizes x-rays or gamma rays for internal structure examination. The testing procedure involves placing the test-part between the gamma or x-ray source and the film (detector) to scan for defects in the component.

Unfortunately, RT can pose hazards to human tissue if a person handling the process is overexposed to the film processing chemicals used in the procedure. This danger is mainly pronounced if the process takes place in an area that’s not well ventilated. Someone interacting with the process for an extended period in such an environment has high risks of various health issues as a result of the inhalation of the fumes. Apart from direct health concerns, RT has waste effluent that can cause environmental pollution. The waste resulting from film processing contains residual silver and other possibly harmful deposits can have a serious adverse effect on waterways, wildlife and plants if not disposed of appropriately.

Where X-Ray Tube is used for NDT testing, the tube head is filled in using an inert gas such as Sulfur-Hexafluoride. Unfortunately, overexposure to this gas can be a health hazard. If a person breathes in the gas in low quantities, the gas relaxes the vocal cords, making his/her voice drop. Inhaling larger volumes of the gas can be more dangerous, as it can lead to suffocation. In such a case, doctor’s intervention is required to vacuum out the gas before it becomes fatal.

Lung Ulcerations and Eye/Skin Tissue Damage

Liquid penetrant testing uses the developer agent for improved penetrant indication visibility. Unfortunately, the agent can also cause problems when the testing is conducted in confined spaces. Prolonged inhalation of the agent can cause respiratory issues as it enters the lungs and absorbs moisture from the walls of the organs. Consequently, ulcerations develop in the lungs.

Penetrant Testing - application of the contraster

Karl Deutsch Prüf- und Messgerätebau GmbH + Co KG https://www.karldeutsch.de
Licenced under Creative Commons Attribution 3.0 Germany

Furthermore, the black lights used in liquid penetrant inspection produce UV-A, UV-B and UV-C light wavelengths. Too much exposure to these rays can cause significant eye and skin tissue damage in humans over time.

How to work flawless?

What is the sentin EXPLORER?

All these catastrophes and failures are reasons for thousands of dead people, destroyed buildings and whole cities. To prevent such disasters, professional inspectors and supporting software should be used to evaluate images and data from constructs. Experienced inspectors often have concentration problems, resulting in errors that can cost lives and millions of dollars. One supporting software is the sentin EXPLORER. This evaluation software supports the examiner in the evaluation and can automate this process by an artificial intelligence.

If you want to learn more: The 5 Types – What is Non-Destructive Testing (NDT)?

Der Beitrag 3 Failures and Catastrophes in NDT erschien zuerst auf sentin.

Human Factors in NDT 4.0 | Interview: Marija Bertovic | Part 2

Mon, 31 May 2021 06:58:10 +0000

Marija Bertovic is a chairwoman of the DGZfP subcommittee “Human machine interaction” within the committee “NDE 4.0” and a secretary of the ICNDT specialist international group “NDT reliability”.

In 2017 she was awarded a certificate of merit for her overall contributions to the field of NDT and in 2018 the Scientific Award by the German Society for Non-destructive Testing (DGZfP) for this work

Nowadays Marija Bertovic is advising on how to include human factors into reliability assessments and simulations and how to make sure new technologies (NDT 4.0, industry 4.0, hydrogen) get accepted, are usable and used appropriately.

Marija Bertovic on Research Gate and LinkedIn

Here is a little overview about the questions Marija Bertovic and we talked about:

First Part of the interview

Marija Bertovic: Her way to becoming a human Factors Research Scientist in NDT
Chances for NDT 4.0 and human factors

Second Part of the interview

Certifications & POD for NDT 4.0
The international future of NDT

If you are interested in human factors of NDT/NDE, check out our article: “6 Human factors of NDT/NDE”

Experiences and thoughts

sentin: How important are standards and certifications for such a system? Is there any other way to establish trust in an artificial intelligence?

Marija Bertovic: Standards and certifications are for sure crucial for the newly developed systems to gain trust and wide acceptance. However, I see potential for raising acceptance not only by ensuring technical capability but also by ensuring that those technologies are easy to use. Therefore, I strongly advise NDT community to get familiar with the standards on human-centred design and usability.

Similar to reliability, that is often only a measure of an idealised performance in a lab, whereas the influences of other factors, such as the human factors are acknowledged but neglected; it is necessary to include those factors when designing or assessing the technology, as they will tell us how this system will actually function and be used in the practice, allowing us to detect potential problems before those systems are put In use.

sentin: Among other things, you are researching in the field of POD (Probability of Detection). In medicine, doctors are now supported by an AI when diagnosing images. Do you see parallels to NDT here?

Marija Bertovic: Of course. From the psychological perspective, there is not much difference between evaluation of xrays conducted by a radiologist or a material scientist: similar principles apply. The difference may only be in the fact that an error in medicine may cost one or several lives, whereas an error in the industry such as nuclear, chemical or aerospace can affect a significantly larger amount of people and have a disastrous effect on the environment. Therefore, reliability of NDT is especially important.

I read recently that one of the first big areas AI will find Its implementation will be to replace general practice doctors, as what they primarily do is diagnose based on the combination of given symptoms. I can see this happening and support the idea that AI may be much more efficient in this task than people. The aim of developing this technology is indeed to take advantage of AI to relieve people from tedious, monotonous tasks, requiring large amounts of memory and attention, i.e. tasks people are generally not good at and to allow them to focus on tasks people excel at: solving complex problems.

“In NDT, as well as in medicine, there is a large number of tasks that can be replaced by AI. However, there will always be tasks that technology will not replace, and a need for people to solve problems that technology is not able to, which is why people will always remain in the centre of the industry and of the AI technology.”

Thus, we need to be careful not to be overrun by our own aspirations to achieve what used to be unimaginable. There should always be an ON/OFF switch controlled by a qualified, skilled and experienced human expert.

sentin: What accuracies can examiners achieve in the near future with an assistance system?

Marija Bertovic: I believe the accuracy of the technical systems will for sure increase with time (at least that is what is promised to me by my fellow engineers). However, I believe people can achieve greater efficiency with those systems only if the systems are designed to be an aid and not a burden, in line with what I’ve already mentioned before.

National and international future in NDT

sentin: You are also active internationally. What are the national and international differences?

Marija Bertovic: It has been my experience that there has always been more openness towards understanding and addressing human factors and nowadays towards human-machine Interaction in NDE 4.0 on the international level, than it has been in Germany. In Germany, standards play a very important role and so far, to my knowledge, there are no standards (nationally or internationally) mentioning how to address the human factors (except in terms of training, qualification and certification). However, it is also Germany that is aiming currently to rectify this. Within a WIPANO project that we just started at the end of 2020, we are working towards developing a guideline for the assessment of the reliability of NDT including human factors, for which we hope will one day result in a standard. So, stay tuned.

sentin: If you think several years into the future. What does this look like with regard to the POD and the HMI?

Marija Bertovic:

“In their strategy on AI, the German Government has repeatedly stated that AI-technologies need to be developed with the people in the centre.”

Thus, I hope it will become unavoidable to pay attention to HMI. That means that technology will be developed with people and for people, tested in the lab until the HMI is efficient, reliable and safe. This means that the technology that will be put in the field will surely function well with humans using them. This is something that our up-to-date POD approaches have been missing, focusing only on the technical capability, and was often criticized by our group

sentin: Assuming you had one wish to advance the issue. What would you wish for?

Marija Bertovic: I would wish that user-centred design becomes a vital part of the development of NDT technologies and procedures. I often get asked what I mean by “developing with the users”. Imagine an iPhone developed by a team of experts and never tested on millions of users world-wide and reshaped according to their feedback. Would it be as intuitive as it is? Would it correspond to your expectations as well as mine? Would it be as easy to use for our parents as well as for our kids? I am not suggesting NDT equipment should be tested by millions of users, but it should be developed and continually tested with a large enough representative number of its potential users. That is, the potential users should be involved from the conception phase (e.g. which functions do they need?), over the prototyping (e.g. does it work as I expect it?) to the final design of the system. User experience design (UX) is no longer such an unknown term.

“I hope it will become more known in NDT.”

About Marija Bertovic

Marija Bertovic

Senior human factors research scientist at Federal Institute for Materials Research and Testing

DGZfP

Member of the German Society for Non-Destructive Testing
Science Award 2018

BAM

Coordinator of the NDT department WG “NDT-AI”

Marija Bertovic is an Awarded and experienced research scientist in the area of human factors in reliability of non-destructive testing (NDT). Actively pursuing projects related to human machine interaction (HMI) in industry 4.0 (artificial intelligence, NDT 4.0, machine learning, user experience, software development, augmented reality etc.).

Chairwoman of the subcommittee “Human Machine Interaction” of the DGZfP technical committee on NDT 4.0; Secretary of ICNDT Specialist international group “NDT reliability” and coordinator of the NDT department WG “NDT-AI” (BAM)

Skilled and experienced in usability testing, risk assessments, experimental method and coordination, human factors training and consulting.

She is a receiver of the Science Award 2018 by the German Society for Non-Destructive Testing (DGZfP) and repeatedly a keynote and plenary speaker.

Member of the German Society for Non-Destructive Testing (DGZfP), Academia NDT International and Human Factors and Ergonomic Society – European Chapter.

See how AI in NDT works

What is the sentin EXPLORER?

Marija Bertovic strongly advises NDT community to get familiar with the standards on human-centred design and usability. This is one way, to develop a user friendly software for the usage of AI in Non-Destructive Testing. Also the medicine works with softwares which are based on this development. Doctors are getting supported by them when diagnosing images. “An error in medicine may cost one or several lives, whereas an error in the industry such as nuclear, chemical or aerospace can affect a significantly larger amount of people and have a disastrous effect on the environment.” This quote describes the importance of a software which brings the AI in the industry. Lots of tasks can be replaces by an AI, so many people work on softwares to make these better and more efficient. This is why the “German Government has repeatedly stated that AI-technologies need to be developed with the people in the centre.” So user-centred design becomes a vital part of the development of NDT technologies and procedures.

If you want to learn more: The 5 Types – What is Non-Destructive Testing (NDT)?

First Part of the interview

Der Beitrag Human Factors in NDT 4.0 | Interview: Marija Bertovic | Part 2 erschien zuerst auf sentin.

Human Factors in NDT 4.0 | Interview: Marija Bertovic | Part 1

Mon, 17 May 2021 10:11:13 +0000

Marija Bertovic on Research Gate and LinkedIn

Here is a little overview about the questions Marija Bertovic and we talked about:

First Part of the interview

Marija Bertovic: Her way to becoming a human Factors Research Scientist in NDT
Chances for NDT 4.0 and human factors

Second Part of the interview

Certifications & POD for NDT 4.0
The international future of NDT

If you are interested in human factors of NDT/NDE, check out our article: “6 Human factors of NDT/NDE”

Marija Bertovic about her position in NDT

sentin: Hello, Ms Bertovic. We are pleased that you, as an expert, want to give us and our readers an understanding of the topic. It would be best if you briefly introduce yourself.

Marija Bertovic: – I studied general psychology at the University of Rijeka in Croatia and soon after in 2006 I became a PhD student at the BAM Federal Institute for Materials Research and Testing in Berlin. There I worked in a research group “Reliability of non-destructive diagnosis systems” led by Dr. Christina Müller and conducted research related to various human factors aspects of NDT, e.g. I looked at the usability and understandability of the inspection procedures, at how people work under time pressure, with other people or with automated systems, how people make single and group decisions and have conducted various studies to identify risks involving working with manual and automated (mechanised) NDT systems and to devise measures against those risks. In 2015 I obtained my doctorate degree on this topic. I also worked shortly for the DGZfP and one year as an independent consultant before returning to BAM to a new position in which I lead a work area “Probabilistic safety and reliability analyses/human-factors analyses”.

sentin: In your dissertation you deal with the topic of human factors in NDT. What are the most relevant aspects? How can these be eliminated?

Marija Bertovic: – In my dissertation I focused on the topic of automation in NDT.

“It is widely assumed that human factors can be eliminated with the Increase in automation in NDT. I postulated in my thesis that whereas some human errors can be eliminated, automation will not eradicate the error as it does not completely replace people but instead aids them in their work, thereby changing the task and potentially giving rise to new risks and human errors, that are unknown.“

Human factors will never be eliminated as long as people are involved in the task, whether as designers of the system or its end users. It should also never be our goal to eliminate human factors, as they refer to all the factors that affect people at work in a way that can affect safety, i.e. the individual, the team, the technology, the organisation (including working conditions) and the extra-organisational environment (e.g. the regulations) and their mutual interactions.

“Instead, our goal should be to reduce or eliminate those factors that give rise to human error.”

They include faulty or not understandable procedures, poorly designed human-machine interfaces, difficult working conditions, poor communication, unclear division of responsibilities and many more.

sentin: You are a researcher at the Federal Institute for Materials Research and Testing (BAM). What changes do you see in the coming years for the NDT 4.0 sector in Germany?

Marija Bertovic: I see an increasing need to keep up with the developments in the industry. NDT will need to keep up as It is a vital part of Industrial processes and because new technology developments offer ways of conducting NDT inspections in a more efficient and reliable manner. As a research scientist, I see BAM supporting those attempts by providing support in the development of those systems as well as by making sure those systems get implemented reliably and safely.

Experiences and thoughts

sentin: You do research in the field of Human Machine Interaction (HMI) and are the chair of this working group at the DGZfP. I have a few questions about this:

How well does the communication between people and machines work in NDT?

Marija Bertovic: I guess people were always able to communicate with NDT systems. Traditionally a system would be developed in a lab, put on the market, people were trained to use it and people and machines worked more or less good together. Thus, communication between people and machines can be forced and learned by training. However, the more complex technology gets, the more complex training gets and with that the demand on the people rises. In spite of extensive training,

“in situations of high stress, or due to distractions, people are likely to react intuitively and not according to the rules they have learned”

(i.e. a red light will be seen as the alarm, as danger, in spite of training that it is the blue light that signals an alarm). Thus, the more complex technologies get, the more important it becomes to give attention to the design of the technical systems as well as to the interaction between people and the systems to ensure that communication runs smoothly.

In my experience, the development and design of hardware and software in NDT still happens in a lab and gets tested only by a limited amount of people before the systems are used in the field. Rapid development of technologies in the era of Industry 4.0 will show that higher effectiveness, efficiency and satisfaction with the product, as well as a more reliable and effective human-machine interaction can only be achieved if the systems are developed together with the users following human-centred design principles (see ISO 9241-11:2018 and ISO 9241-210:2020).

sentin: How willing is one to work with a machine/computer? To what extent do people trust a hardware or software system?

Marija Bertovic: Trust is a very complex issue and It depends on a variety of factors. For example, trust in a piece of technology (hardware or software) will depend on its capability, reliability, as well as one’s own experience with that system. Moreover, trust can be shaped by individual differences in e.g. attitudes towards technology, one’s own abilities, self-confidence or the extent of risk-taking. According to the Technology Acceptance Model developed by Davis in 1986, the acceptance and, respectively, the use of the system, is affected by the perceived usefulness and the perceived ease of use (both of which can be affected by a variety of external variables, such as usability) that together with the attitude towards using affect the behavioural intention to use and, consequently, its actually use. In other words, the willingness of people to accept and work with a piece of technology is highly affected by its usability, emphasizing again the relevance of user-centred design.

sentin: What examples can you think of spontaneously where a machine or a computer is already helping the inspector? And how exactly can they support the examiner/evaluator?

Marija Bertovic: In my work I came across software solutions that were designed to automatically detect and sometimes even characterise defects. These kinds of software solutions were used as aids to the inspectors, wherein it was the role of the inspector to control and, if necessary, correct the results of the software. I believe this kind of technology will soon become omnipresent thanks to the advances in machine learning.

“With time, the software should be able to do this task independently and the role of people will change from supervising those systems Into solving complex problems that the technology is not yet able to.“

Another example is the use of augmented reality to visualise defects (by visualising signal amplitudes) that can be used in training of the inspectors. This kind of technology will for sure find wider application in the field by visualising not only present defects but also results from past inspections, examples from books, procedures, allowing communication with people off site, thereby aiding the inspector in making his judgment on the presence of a potentially critical defect.

About Marija Bertovic

Marija Bertovic

Senior human factors research scientist at Federal Institute for Materials Research and Testing

DGZfP

Member of the German Society for Non-Destructive Testing
Science Award 2018

BAM

Coordinator of the NDT department WG “NDT-AI”

Skilled and experienced in usability testing, risk assessments, experimental method and coordination, human factors training and consulting.

She is a receiver of the Science Award 2018 by the German Society for Non-Destructive Testing (DGZfP) and repeatedly a keynote and plenary speaker.

Member of the German Society for Non-Destructive Testing (DGZfP), Academia NDT International and Human Factors and Ergonomic Society – European Chapter.

See how AI in NDT works

What is the sentin EXPLORER?

As Marija Bertovic said, more and more people become trained to use AI in Non-Destructive Testing. “In situations of high stress, or due to distractions, people are likely to react intuitively and not according to the rules they have learned“ For these cases, there are softwares which support the human inspectors. She talks about higher effectiveness, efficiency and satisfaction, if the rapid development of technologies in the era of Industry 4.0 goes on. Trust in a piece of technology will depend on its capability, reliability, as well as one’s own experience with that system. This is a reason for the fact, that the willingness of people to accept and work with a piece of technology is highly affected by its usability, emphasizing again the relevance of user-centred design. Marija Bertovics thoughts about the future are that with time, the software should be able to do their tasks independently.

If you want to learn more: The 5 Types – What is Non-Destructive Testing (NDT)?

Second Part of the interview

Der Beitrag Human Factors in NDT 4.0 | Interview: Marija Bertovic | Part 1 erschien zuerst auf sentin.

sentin

The AI-ready NDT infrastructure Guide.

Inhalt

What is an AI-ready NDT infrastructure?

What are the components of an AI-ready infrastructure?

What are the benefits of an AI-ready NDT infrastructure?

What tasks does an AI-enabled infrastructure assist with?

What are the requirements of an AI-ready NDT infrastructure?

How to get an AI-ready NDT infrastructure?

Software Products:

Services:

AI & Automation Tools:

Hardware & Services with partners:

The NDT 4.0 Guide – Everything you need to know.

Inhalt

What is NDT 4.0?

How do industry and NDT compare?

What technologies does NDT 4.0 use?

What advantages does NDT 4.0 offer?

Advantage 1: Improved NDT merhods

Advantage 2: Improved Equipment

Advantage 3: More control & traceability

Advantage 4: Simplified data transfer

Advantage 5: Better data evaluation

Advantage 6: Improvement of the production process

What are the challenges for NDT 4.0?

Challenge 1: Technical hurdles, data and infrastructure

Challenge 2: Relevance of the skills learned

Challenge 3: Hesitant managers

Challenge 4: Insufficient understanding

What do experts say about NDT 4.0?

How can I use NDT 4.0 today?

Sources and further links

What is a PACS in NDT?

What does PACS stand for in NDT?

What are the advantages of a PACS?

What interfaces does a PACS have?

What is the DICOM or DICONDE format?

What are DICOM services?

What is DICOMWeb?

How do you set up a PACS?

Is a PACS future-proof?

Where can I buy a PACS (for NDT)?

Which viewer & evaluation softwares are available?

How does Artificial Intelligence (AI) in NDT work?

Inhalt

What is Artificial Intelligence (AI)?

Where can Artificial Intelligence (AI) be used for NDT 4.0?

Use case 1: Data Analytics

Use Case 2: Digital Twin

Use case 3: In the evaluation / inspection process

Critical Item / Image Detection (CID)

Single Item / Image Defect Analysis (SIDA)

Denoising, Superresolution, Data & Image Enhancement​

(Automatic) Defect Recognition - ADR

What is the current performance of NDT (without AI)?

What is the Probability of Detection (POD)?

How can Artificial Intelligence reduce human factors in NDT?

Example 1: Improving Reliability & Perception with ADR

Example 2: Improving Perception & Attentiveness with Image Enhancement.

What sentin does with AI in NDT?

One Day as A NDT-Inspector – Work and Life in NDT

Facts

The Daily Tasks of an NDT-Inspector

Communication

Teamwork

Work Hours and Compensation

Using Artificial Intelligence to support Inspectors

History of NDT – 5 Interesting Facts

Facts

People begin to recognize NDT as a technology

1920s and before

World War II - Catalyst for NDT evolution

1930s and 1940s

Radiographic Testing

1940s and after

Foundation of Eddy Current Testing - Hans Christian and Michael Faraday

1948 to 2012

NDT applications keep on increasing

2010s and after

Denoising, Superresolution, Data & Image Enhancement