{"id":2007,"date":"2025-01-15T07:48:15","date_gmt":"2025-01-15T15:48:15","guid":{"rendered":"https:\/\/superwomansolutions.com\/engtechnica\/?p=2007"},"modified":"2025-01-16T12:54:28","modified_gmt":"2025-01-16T20:54:28","slug":"in-line-ct-scan-delivers-real-time-quality-feedback-on-production-line","status":"publish","type":"post","link":"https:\/\/superwomansolutions.com\/engtechnica\/in-line-ct-scan-delivers-real-time-quality-feedback-on-production-line\/","title":{"rendered":"Heitec Uses AI to Check for Flaws on Auto Assembly Line"},"content":{"rendered":"\n<p>The moving assembly line has certainly proven its worth over the 100-plus years since Henry Ford introduced the industrial time-and-motion breakthrough in his Michigan automobile factory in 1913. The addition of robots (by GM in 1962) alongside human workers has further increased efficiency.<\/p>\n\n\n\n<figure class=\"wp-block-image aligncenter size-full\"><img loading=\"lazy\" decoding=\"async\" width=\"909\" height=\"511\" src=\"https:\/\/superwomansolutions.com\/engtechnica\/wp-content\/uploads\/2025\/01\/engineer1.jpg\" alt=\"\" class=\"wp-image-2008\" srcset=\"https:\/\/superwomansolutions.com\/engtechnica\/wp-content\/uploads\/2025\/01\/engineer1.jpg 909w, https:\/\/superwomansolutions.com\/engtechnica\/wp-content\/uploads\/2025\/01\/engineer1-300x169.jpg 300w, https:\/\/superwomansolutions.com\/engtechnica\/wp-content\/uploads\/2025\/01\/engineer1-768x432.jpg 768w\" sizes=\"auto, (max-width: 909px) 100vw, 909px\" \/><figcaption class=\"wp-element-caption\"><em>A Heitec engineer monitoring a robotic CT scan and analysis system as it inspects a manufactured component.<\/em><\/figcaption><\/figure>\n\n\n\n<p>The success of the assembly line is tied to the quality of the components fed to it from production. As the complexity of modern vehicles has risen dramatically in recent decades, particularly in the case of electric vehicles (EVs), the value of pre-assembly quality checks at key stages during active production has become increasingly apparent. In addition to various methods of surface scanning, some of the leading automotive manufacturers (along with others in aerospace, energy and electronics) are using computed tomography (CT) to inspect inside components. <\/p>\n\n\n\n<figure class=\"wp-block-image aligncenter size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"577\" src=\"https:\/\/superwomansolutions.com\/engtechnica\/wp-content\/uploads\/2025\/01\/F1scan-1024x577.jpg\" alt=\"\" class=\"wp-image-2009\" srcset=\"https:\/\/superwomansolutions.com\/engtechnica\/wp-content\/uploads\/2025\/01\/F1scan-1024x577.jpg 1024w, https:\/\/superwomansolutions.com\/engtechnica\/wp-content\/uploads\/2025\/01\/F1scan-300x169.jpg 300w, https:\/\/superwomansolutions.com\/engtechnica\/wp-content\/uploads\/2025\/01\/F1scan-768x432.jpg 768w, https:\/\/superwomansolutions.com\/engtechnica\/wp-content\/uploads\/2025\/01\/F1scan.jpg 1055w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><figcaption class=\"wp-element-caption\"><em>Examples of CT-scan-data analysis of an automotive part.<\/em><\/figcaption><\/figure>\n\n\n\n<p>Germany&#8217;s luxury automaker, BMW, is doing CT scans of every engine housing in their iX EVs.<\/p>\n\n\n\n<p>The iX is based on an independent platform, unlike the company\u2019s other electric vehicles, However, it it uses the same drive unit (motor, inverter, transmission) known as HEAT (Highly-Integrated Electric Drive Train) that is employed in the other BMW models. Ensuring the consistency and quality of this valuable drivetrain is clearly a priority for the automaker.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">How In-line CT Quality Inspection Works<\/h3>\n\n\n\n<p>During quality testing with&nbsp;<a href=\"https:\/\/mail.google.com\/mail\/u\/0\/#m_-1050131206862963638_\">in-line computed tomography (CT<\/a>), an X-ray source emits radiation that penetrates objects and is attenuated depending on the material and geometry of the workpiece to be tested. A detector attached at the opposite side of the source creates a shadow image of the workpiece from the remaining radiation; the varying grey tones of the shadow images represent the differences in the nature of the test specimen. Control software then generates a 3D model of the object from the 2D X-ray image stacks. Finally, holistic analysis and visualization software is used to view and analyze the digital model from a variety of specific perspectives to either detect potential faults or to certify acceptable quality.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Robots and Scanners and Software\u2014On the Line<\/h3>\n\n\n\n<p>At its Landshut, Germany, facility, BMW is using a combination of in-line robotics, CT scanning and detection, and analysis and visualization software, to inspect their HEAT e-engine housing during active production. The entire system was installed and integrated into their production line by Heitec, a global engineering solutions provider for a broad range of industries.&nbsp;<\/p>\n\n\n\n<p>Heitec\u2019s installation delivers a continuous-flow process in which a robot plucks an aluminum housing off the production line, deposits it on a platform that feeds it into a HeiDetect Inline CT scanner, then shifts to the other side of the scanner to retrieve an already-scanned component and moves that one back to the line. Each scan takes place in 50 seconds.<\/p>\n\n\n\n<figure class=\"wp-block-embed is-type-video is-provider-youtube wp-block-embed-youtube wp-embed-aspect-16-9 wp-has-aspect-ratio\"><div class=\"wp-block-embed__wrapper\">\n<iframe loading=\"lazy\" title=\"HeiDetect FX Inline CT (HDFX)\" width=\"500\" height=\"281\" src=\"https:\/\/www.youtube.com\/embed\/0yjRk5o7ORg?feature=oembed\" frameborder=\"0\" allow=\"accelerometer; autoplay; clipboard-write; encrypted-media; gyroscope; picture-in-picture; web-share\" referrerpolicy=\"strict-origin-when-cross-origin\" allowfullscreen><\/iframe>\n<\/div><\/figure>\n\n\n\n<p>Post-scan, Heitec\u2019s proprietary software quickly generates a digital model of each housing from the CT data, then transfers the model to&nbsp;Hexagon\u2019s VGinLINE software&nbsp;for rapid segmentation and analysis. Viewable onscreen by engineers, graphic-and-colored alerts are automatically generated to call out any significant porosity, cracks, geometric inconsistencies, or other production or material flaws that deviate from BMW\u2019s quality specifications. <\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Delivering Production Insights\u2014at Speed<\/h3>\n\n\n\n<figure class=\"wp-block-image aligncenter size-full\"><img loading=\"lazy\" decoding=\"async\" width=\"932\" height=\"687\" src=\"https:\/\/superwomansolutions.com\/engtechnica\/wp-content\/uploads\/2025\/01\/pic.jpg\" alt=\"\" class=\"wp-image-2011\" srcset=\"https:\/\/superwomansolutions.com\/engtechnica\/wp-content\/uploads\/2025\/01\/pic.jpg 932w, https:\/\/superwomansolutions.com\/engtechnica\/wp-content\/uploads\/2025\/01\/pic-300x221.jpg 300w, https:\/\/superwomansolutions.com\/engtechnica\/wp-content\/uploads\/2025\/01\/pic-768x566.jpg 768w\" sizes=\"auto, (max-width: 932px) 100vw, 932px\" \/><figcaption class=\"wp-element-caption\"><em>Christian Abt of Heitec<\/em><\/figcaption><\/figure>\n\n\n\n<p>\u201cDoing all this in the cycle time of the production line is the challenge,\u201d says Heitec co-founder Christian Abt. \u201cYou have to speed everything up to stay within that time\u2014or even several seconds less than that.\u201d<\/p>\n\n\n\n<p>Achieving this kind of fine-tuned industrial automation has become second nature to Heitec, which has been installing increasingly \u201csmart\u201d factory systems worldwide for more than 26 years. After completing his engineering degree, Abt first worked at the Fraunhofer Institute in Germany (\u201cfor a young engineer, a fantastic playground for robotics\u201d). He left to co-found Heitec as robots were becoming a commodity, deciding to strengthen his young company\u2019s automation offerings with the addition of CT scanning.<\/p>\n\n\n\n<p>\u201cPrior to that time, most CT was being used in the laboratory,\u201d he says. \u201cWe saw an opportunity to bring these machines into real-world production, supporting the guys in blue jackets on the factory floor with our x-ray technology, our robots, and our knowledge of automation.\u201d<\/p>\n\n\n\n<p>\u201cManufacturing companies that are now coming to us for in-line CT have usually checked out all other possibilities but found no other solution that provides the level of quality assurance they are looking for,\u201d says Abt. \u201cThey are demanding answers that work for them.\u201d<\/p>\n\n\n\n<p>Heitec customizes every installation to the specific industrial customer, sets up the robot(s) and CT scanner(s)\u2014and then must deliver the promised rapid-inspection results. \u201cIf you\u2019re familiar with using industrial CT in the past, you may remember hours for scanning and gigabytes of data to sift through for defects,\u201d Abt says. \u201cThat\u2019s not the case here in production\u2014there\u2019s no time for that kind of delay.\u201d<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">The Secret\u2019s in the Scan\u2014and the Software<\/h3>\n\n\n\n<p>The solution to the time crunch? Employing lower-resolution industrial CT scans that need significantly less time to capture an image\u2014plus Heitec\u2019s scanner-integrated software designed to process this \u201cless-heavy\u201d information and reconstruct the component\u2019s volume data into a 3D computer model. This data is then loaded into holistic analysis and visualization software that queries the digital model, employing sophisticated algorithms and, most recently, deep learning\u2014to reliably identify, interpret and report potential quality issues despite the noisy images\u2014within a much shorter time window.<\/p>\n\n\n\n<p>\u201cFor the kind of advanced data processing software that would deliver the metrology results our customers were asking for, we found our way to Hexagon\u2019s VGSTUDIO MAX and VGinLINE,\u201d says Abt. \u201cWe had evolved from software developed for research to the situation where industry was asking for more standard and pre-qualified digital tools. Everyone was aware of VG software; it was, and is, the standard in CT software.<\/p>\n\n\n\n<p>\u201cThe complete infrastructure of VGinLINE is a ready-to-use software package that allows us to easily integrate our own modules and functions, enabling us to combine accepted standards with flexibility. We just drop out a volume and VG starts to analyze it. This process is also easy to parallelize to more than one computer to speed things up in case the time for analyzing the data is slightly longer than the time to scan the part.\u201d<\/p>\n\n\n\n<p>The VG CT-data analysis and visualization software has undergone over a quarter century of evolution and development, with process automation and, most recently, deep- and machine-learning, increasingly being integrated into its offerings. Says Hexagon\u2019s general manager for VG products, Dr. Daniela Handl, \u201cThe huge quantities of design and engineering data that can be captured by CT needs a holistic approach, coupled with robust computational capabilities, to be most valuable. Rapid results delivery is critical for engineers looking for real-time insights into production-line quality as well as how manufacturing-parameter variations might be affecting outcomes.\u201d<\/p>\n\n\n\n<p>Key to determining which data is significant in the scan of a component is the process of segmentation\u2014the extraction of regions of interest (ROIs) from the 3D image data. Abrupt discontinuities in voxel grey-values typically indicate edges that define a region. By partitioning a scan into discrete regions, the software can more rapidly focus on and catalogue the characteristics of the voxels in each segment. Segmentation thus enables the processing of only the important portions of a dataset\u2014which takes a lot less time.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\">Customizing Proprietary Solutions with Machine Learning<\/h3>\n\n\n\n<p>But how do you determine which segments are the important ones? Especially since, in a lower-resolution CT scan obtained on a production line, the less-crisp image slices can be more challenging to understand. This is where deep- and machine-learning, forms of artificial intelligence (AI), comes into the picture: using trainable algorithms, AI can be harnessed to process and make decisions based on noisy CT data&nbsp;as accurately as on hi-resolution data.<\/p>\n\n\n\n<figure class=\"wp-block-image aligncenter size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"1017\" src=\"https:\/\/superwomansolutions.com\/engtechnica\/wp-content\/uploads\/2025\/01\/CT-analysis-1024x1017.jpg\" alt=\"\" class=\"wp-image-2013\" srcset=\"https:\/\/superwomansolutions.com\/engtechnica\/wp-content\/uploads\/2025\/01\/CT-analysis-1024x1017.jpg 1024w, https:\/\/superwomansolutions.com\/engtechnica\/wp-content\/uploads\/2025\/01\/CT-analysis-300x298.jpg 300w, https:\/\/superwomansolutions.com\/engtechnica\/wp-content\/uploads\/2025\/01\/CT-analysis-150x150.jpg 150w, https:\/\/superwomansolutions.com\/engtechnica\/wp-content\/uploads\/2025\/01\/CT-analysis-768x763.jpg 768w, https:\/\/superwomansolutions.com\/engtechnica\/wp-content\/uploads\/2025\/01\/CT-analysis.jpg 1200w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><figcaption class=\"wp-element-caption\"><em>Advanced CT analysis software enables grey-value-based, shape-based, or machine-learning-based segmentation of 3D data. These defined regions of interest (ROIs) can be used to perform further analyses.<\/em><\/figcaption><\/figure>\n\n\n\n<p>\u201cTrainable\u201d is the key word here. The algorithms compare what they see against an existing database of all known identifiable defects that have been customized to each manufacturer\u2019s product geometries (which is where the \u201ctraining\u201d comes in). The algorithms \u201clearn\u201d to identify specific manufacturing flaws by interpreting what the less-crisp voxels show.<\/p>\n\n\n\n<p>Over time, deep learning gets even better at what it does; by comparing what it sees against real-world product data for which the solution is known, it learns to recognize patterns and features and call out deviations from the norm. In this way, it can provide highly accurate snapshots of what\u2019s actively happening on a particular production line\u2014supporting confident decision-making about whether to accept or reject a part. This in turn informs production-variable changes, the effects of which can be captured, collated, and statistically examined. EV batteries are another area in which machine learning can be used in this way.<\/p>\n\n\n\n<p>How to optimize this kind of AI methodology is what BMW is now exploring\u2014as are forward-thinking manufacturers in other industries. Working closely under NDAs, with innovators such as Heitec and Hexagon, these companies are creating and curating their own internal data sets with which to train in-house deep learning systems.<\/p>\n\n\n\n<p>\u201cTraining a deep learning system to one\u2019s proprietary data understandably takes some time,\u201d says Handl. \u201cBut it pays off by saving highly valuable time and resources on the production line.\u201d Actionable insights into what\u2019s happening on one\u2019s own factory floor are contributing to the further evolution of smart manufacturing, enabling companies across many industries to identify ways to improve quality while making their products more competitive.<\/p>\n\n\n\n<hr class=\"wp-block-separator has-alpha-channel-opacity\"\/>\n\n\n\n<p class=\"has-text-align-center\">Sponsored content, source: Heitec. <br>See <a href=\"https:\/\/superwomansolutions.com\/engtechnica\/sponsorship\/\">Sponsor<\/a> to get your article or news posted on ENGtechnica. <\/p>\n","protected":false},"excerpt":{"rendered":"<p>Heitec does low-resolution CT scans so it can keep up with the assembly line and depends on AI to detect flaws such as cracks in aluminum castings. <\/p>\n","protected":false},"author":6,"featured_media":2008,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[23,16],"tags":[267,269,266,268,265,261,263,264,262],"class_list":["post-2007","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-manufacturing","category-news","tag-automotivetechnology","tag-ctscan","tag-ctscanningtech","tag-deeplearningai","tag-evproduction","tag-inlinect","tag-productionlineinnovation","tag-qualitycontroltech","tag-smartmanufacturing"],"_links":{"self":[{"href":"https:\/\/superwomansolutions.com\/engtechnica\/wp-json\/wp\/v2\/posts\/2007","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/superwomansolutions.com\/engtechnica\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/superwomansolutions.com\/engtechnica\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/superwomansolutions.com\/engtechnica\/wp-json\/wp\/v2\/users\/6"}],"replies":[{"embeddable":true,"href":"https:\/\/superwomansolutions.com\/engtechnica\/wp-json\/wp\/v2\/comments?post=2007"}],"version-history":[{"count":9,"href":"https:\/\/superwomansolutions.com\/engtechnica\/wp-json\/wp\/v2\/posts\/2007\/revisions"}],"predecessor-version":[{"id":2071,"href":"https:\/\/superwomansolutions.com\/engtechnica\/wp-json\/wp\/v2\/posts\/2007\/revisions\/2071"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/superwomansolutions.com\/engtechnica\/wp-json\/wp\/v2\/media\/2008"}],"wp:attachment":[{"href":"https:\/\/superwomansolutions.com\/engtechnica\/wp-json\/wp\/v2\/media?parent=2007"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/superwomansolutions.com\/engtechnica\/wp-json\/wp\/v2\/categories?post=2007"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/superwomansolutions.com\/engtechnica\/wp-json\/wp\/v2\/tags?post=2007"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}