:::

【2020 Application Example】 AOI fabric inspector lowers the false negative rate, and reduced the re-inspection volume by 70%

Low detection rate, slow speed, difficult recruitment and high personnel costs

The textile industry has always been a labor-intensive industry. At present, almost all textile companies worldwide still inspect fabrics manually. There are three major pain points in manual fabric inspection: Low detection rate, slow speed, difficulty in recruiting workers, and high personnel costs On average, a fabric inspector can find up to 200 defects in one hour with a defect detection rate of about 70%.

However, inspectors are only able to maintain their concentration for 20 to 30 minutes at most, and their fabric inspection speed is generally limited to 20 to 30cm/s. Fabric inspectors become fatigued if they exceed this time and speed.

Domestic and foreign AOI fabric inspection machines purchased by textile manufacturers have not yet been officially integrated into the production line. At the beginning, 10,000 suspected defects could be detected in one roll of fabric. The detection rate was high but the accuracy [screening] was low. The number of suspected defects has been reduced to 7000, but is still not at the level of experienced inspectors.

High-speed cameras capture defects and record defect locations

▲High-speed cameras capture defects and record their locations

The rule-based defect identification method currently used by manufacturers requires a lot of adjustment time (about 1 to 3 months) before the manufacturers (site) actually uses it, and there is currently no solution to automatically correct the identification model after use. As a result, manufacturers need to spend extra time to adjust parameters. Therefore, it requires considerable cost for both manufacturers and clients (sites).

Current manufacturers' grey cloth inspection process

▲Current grew fabric inspection process of manufacturers

The specific method used by the guidance team and cooperating manufacturers to implement AI identification technology and learning framework (for model retraining) into the defect inspection process is described below

1. AI-based defect identification model:

Utilizes the large amount of image data collected (including fabrics with and without defects) to construct the defect detection model through machine learning, such as SVM, or deep learning object detection methods, such as SSD or YOLOv3. This model is used to determine the condition of the surface of grey fabric and determine if it is a normal product or a defective product, thereby achieving defect identification.

2. Identification model retraining framework:

If there is an error in the judgment of the visual inspector, the image will be marked and the data will be used in the dataset for re-training. After a certain number of misjudged data is accumulated, the system will automatically start the identification model retraining function, and the new model that is generated will automatically replace the old recognition model, thereby achieving the purpose of model update.

Post-project implementation grey cloth defect inspection process

▲Grey fabric defect inspection process after the implementation of this project

Low false negative rate and solves the challenges of labor shortage and higher quality requirements in the industry

This project uses a deep learning network architecture to reclassify defects that are detected, including real defects and false defects, and can further classify real defects and false defects to lower the false negative rate of traditional AOI solutions. This is expected to reduce re-inspection volume by 70% and above for fabric inspectors, eliminate concerns about implementation in the current production line, accelerate the application of AI-based AOI solutions by textile manufacturers, and solve the challenges of labor shortage and higher quality requirements in the industry.

Recommend Cases

這是一張圖片。 This is a picture.
[2023 Case Study] AI Steps into Philanthropy: Stylish Tech at Food Banks

Taiwan Food Bank AssociationHereinafter referred to as 'the Association'With the mission of providing food aid, poverty relief, reducing food waste, and building a hunger-free network, there are locations across Taiwan that gather donations from wholesalers, intermediaries, retailers, manufacturers, and even generous individuals These sites also rescue food that would otherwise be discarded, properly allocate and distribute it to needy households, thus aiding local vulnerable families55Food banks at various locations collect daily donations from wholesale stores, intermediaries, retailers, manufacturers, and even benevolent individuals from all over Taiwan These places also rescue about-to-be-discarded edible materials, properly sort them, and distribute to needy households, assisting local vulnerable populations However, each location requires significant human and volunteer resources to manage daily operations using traditional methods of communication with non-profit organizations and donors After receiving donations, these resources are then allocated to needy families or individuals There is a potential issue of uneven distribution of resources due to a lack of digitalization and integrated information management in these processes Warehouse and Transportation Centers and Mini Food Banks Distributing Resources to the Disadvantaged The location under validation by the Kaohsiung Charitable Organizations Association,Hereinafter referred to as 'Kaohsiung Charity' In109year6month24Officially inaugurated Taiwan's first 'Food Bank-Warehouse and Transportation Center' at a location measuring200square meters, enhancing the efficiency of food resource redistribution, proper storage, and management So far, nearly two hundred tons of vegetables and fruits have been saved, serving over a hundred organizations and benefiting over5thousand vulnerable households, and continues to serve19mini food banks, with planned completion across multiple districts in Kaohsiung, distributing food resources to over10ten thousand vulnerable families Kaohsiung Charity 'Food Bank-Warehouse and Transportation Center' in the Dasha Community Photo Source Kaohsiung Charitable Organizations Association Challenges in Labor and Food Resource Management Facing the needs of a large number of economically disadvantaged families, the management of the 'Food Bank-Warehouse and Transportation Center' is particularly critical During procurement, tasks such as sorting, purging, and bookkeeping must be performed, while during shipment, food resource needs suggested by social workers must be followed These activities rely on manual judgment and accumulated experience Many volunteers involved are elderly and have limited physical strength, making warehouse tasks physically demanding and recruitment challenging If a large batch of food resources arrives, space and manpower are consumed in sorting and inventory management, raising concerns about the effective use of resources and turnover rate This highlights the challenge of scaling up food bank services while lacking corresponding labor and material management systems At the same time, food bank resources come from various donations, thus they vary greatly in type, shelf life, standards, and quantity Volunteers at mini food banks, mostly also elderly, must handle multiple responsibilities such as case services, food resource management,resource allocation, and resource development Sometimes they must also explain and accept immediate, large quantities of specific resources, such as adults receiving baby formula 'Food Bank-Warehouse and Transportation Center' Resource Inventory Relies Entirely on Manual Labor Mini Food Bank Volunteers Handle Multiple Responsibilities Photo Source Taiwan Food Bank Association Reducing Scrap Resources60 Increasing Speed of Resource Transfer80 To enhance resource management and ensure effective use of materials, and to address personnel shortages, this field validation case has introduced 'Food Bank Warehouse Resource CollectionAITo advance resource management, ensure effective use of resources, and solve manpower shortages, this validation site has implemented an 'Automated Early Warning Needs Assessment System' for the food bank's warehouse resource gathering The first part involves building a classification model, setting up and collecting warehouse information at the site, andAItraining the model Past sitewarehouse information is collected and stored in a database, allowingAIfor preprocessing, classification, and other tasks At the same time, depending on the dependency conditions of the types of goods as features, algorithms are introduced for computation and modeling, and the data collected is used for retraining, ultimately validating the field and organizing data for the five most common types of goods into training and test datasets as required The second part involves constructing the classification model using AI techniques further use of reinforcement learning constructs the management mechanism for the food bank's warehouse, perfecting the classification of donated goodsRNNTechnical construction of classification models further use of reinforcement learning constructs food bank warehouse management mechanisms, making the classification of donated goods perfectlike white rice, instant drinks, noodles, instant noodles, and canned goodscan then be automatically assigned storage based on storage assignment principles AI Service System Process and Description Source Taiwan Food Bank Association AtAIUnder forecasts, it can optimize the speed of resource transfer and allocation, effectively and accurately match resource donations reducing the loss in the donation process, increase the accuracy of resource distribution, and improve the service rate—the successful donation rate—reducing the waste of resources due to incorrect items, and enabling instant monitoring of food resource stock, ensuring operators can respond quickly to needs, effectively providing resource assistance WithAIthe system's introduction and the establishment of data intelligence, it helps the operations of the warehouse and transportation center, allowing more time for the allocation of donated goods The introduction aims to accelerate the digital service rollout for social welfare organizations, thoroughly addressing the needs of the overall vulnerable segments of society Using the system for resource allocation and dispatching Photo Source Kaohsiung Charitable Organizations Association Following this field validation, it is possible to expand the system to other food bank service pointsAIThe system can also collaborate with more non-profit organizations, public welfare groups, and charitable organizations, expanding 'Food Bank Warehouse Resource CollectionAIAutomated Early Warning Demand Assessment System' application range such as medical supply distribution, helping more organizations manage and distribute more intelligently, reducing resource wastage, and enhancing social welfare 「Translated content is generated by ChatGPT and is for reference only Translation date:2024-12-12」

這是一張圖片。 This is a picture.
Realizing the dream of unmanned stores, Magpie Life is building the future of the smartphone industry

"The DNA of Magpie Life is not limited to vending machines We believe that vending machines combine technology, access, and humanities to bring us exciting results" This is a sentence on the official website of Magpie Life Let the vending machines bring To live a pleasant life and build a considerate, technological and sustainable future for the smartphone industry is also the original intention of Magpie Life Founded in 2018, Magpie Life launched Taiwan’s first private-brand mobile payment scan code sensor 4 months after its establishment, completing the consumption experience through screen touch The Magpie U1 smart vending machine manages the POS system and gathers data in the background, allowing consumers to synchronize with the world's new retail pace and experience a new retail consumption experience of purchasing convenience, checkout security, visual entertainment, and improved logistics replenishment efficiency Traditional vending machines lack information visibility and AI technology assists in information transparencyThis time, the Magpie smart vending machine is also equipped with AI technology to provide adjustable shelf space , a vending machine equipped with an industrial computer and a large-size touch display screen to achieve the purpose of a store-less store Magpie Life stated that the biggest problem with traditional vending machines is the lack of information visibility To check inventory, replenishment personnel must physically inspect each machine, which is time-consuming and costly When a machine breaks down, it will generally be unable to operate for a long time Most failures go unreported and are not discovered until the next restocking crew arrives to replenish supplies Then you have to wait for a service technician to be scheduled, which can take weeks Traditional vending machines lack real-time interactivity When consumers encounter problems after inserting coins, manufacturers cannot handle them immediately In addition, traditional vending machines are less flexible and cannot adapt to changes in consumer preferences Traditional vending machines have shortcomings such as limited change shopping, single payment tools, limited number of products, and few choices Affected by the COVID-19 epidemic, consumption habits have shifted to contactless methods, causing the unmanned store market to heat up Generally, vending machines can only place relatively simple products such as drinks, food, etc The properties available for sale are limited The patented vending machine developed by Magpie can adjust the shelf space and is equipped with a lifting cargo elevator, which is suitable for various types of goods In addition, the machine is equipped with an industrial computer and a large-size touch display screen, which can meet the needs of advertising support at the same time It is expected to move towards a storeless store According to Magpie Life Observation, the consumer market trend in the past two years is that consumers demand convenient life, food consumption patterns value dining experiencesimple and fast, and are equipped with mobile phone-connected ordering models, and hot drinks and Fresh food delivery is the focus of two major trends The location, items sold, consumption methods and multiple payment methods are the focus of market growth for smart vending machines In terms of convenience, Taiwanese consumers still prefer to purchase vending machine food near stations, airports, schools, and businesses in business districts Various payment methods are also gaining more support from consumers, indicating that in the future, automatic Vending machines can be developed in two directions diversified items and diversified payment methods AI sales forecast technology integrates back-end management to achieve precise marketing purposesDue to the wide variety of products, it is difficult to know the performance of products under different factors such as season, market conditions , promotional activities, etc, it is easy to cause out-of-stock or over-inventory situations Magpie Life has specially developed "AI sales forecasting technology" and integrated it into the back-end management system, hoping to lock in customer purchasing preferences and intentions through data analysis In order to achieve the purpose of precise marketing, make accurate business decisions and effectively allocate limited resources The introduction of AI systems can achieve the three major goals of precise marketing, inventory management and supply chain management This system is a replenishment decision-making aid designed specifically for supply chain managers It uses AI to predict future sales demand, helping companies effectively optimize production capacity, inventory and distribution strategies Its overall system architecture includes1 Data exploratory analysis function Provides automatic value filling, automatic coding and automatic feature screening functions for missing values in the data 2 Modeling function 1 Provides model training functions for two types of prediction problems regression Regression and time series Time Series Forecast nbsp2 Supports Auto ML automatic modeling, and the best model is recommended by the system Integrated models can also be established to improve model accuracy nbsp3 Supports multiple algorithm types Random Forest, XGBoost, GBM and other algorithms nbsp4 Supports a variety of time series models exponential smoothing, ARIMA, ARIMAX, intermittent demand, dynamic multiple regression and other models nbsp5 Supports a variety of model evaluation indicators R, MAE, MSE, RMSE, Deviance, AUC, Lift top 1, Misclassification and other indicators nbsp6 Supports automatic cutting of training data sets and Holdout verification data sets, and can manually adjust the ratio nbsp7 Supports automatic model ensemble learning Stacked Ensemble, balancing function learning Balancing Classes, and Early Stopping nbsp8 Supports the creation of multiple models at the same time The system will allocate resources according to modeling needs, so that modeling, prediction and other tasks have independent computing resources and do not affect each other In the overall server space With an upper limit, computing resources can be used efficiently nbsp9 It has in-memory computing function, which can use large-capacity and high-speed memory to perform calculations to avoid reading and writing a large number of files from the hard disk and improve computing performance 3 Data concatenation function Using API grafting and complete data concatenation automation, there is no need to manually import data, improving user experience 4 Chart analysis function Provides visual charts and basic statistical values for product sales AI data analysis solutions have two major advantages 1 Entrepreneurship machines can be rented and sold at low cost to open unmanned physical stores and cooperate with the chain retail industry Through smart machines, entrepreneurs can rent and sell them at a lower cost than the store rent Cost of running a retail business Two cooperation models, machine sales and leasing, are provided, and the choice is based on the evaluation of the industry 2 Various types of products are put on the shelves Products are sold anytime and anywhere 24 hours a day Up to 60 kinds of diversified products can be put on the shelves Large transparent windows enhance the visibility of products Regular replenishment and tracking of product sales status are available, and product types can be adjusted according to needs Recently, the line between the Internet and the physical world has blurred, the way customers interact has changed significantly, and consumer demand is changing and personalized The retail industry is facing unprecedented challenges and uncertainties, and mastering data has become key AI data analysis solutions can help the retail industry quickly activate large amounts of data, create seamless personalized experiences, optimize the operational value chain and improve efficiency, and strengthen the core competitiveness of enterprises 「Translated content is generated by ChatGPT and is for reference only Translation date:2024-05-19」

【導入案例】海量數位工程AOI機器智能手臂檢測系統 大幅提高瑕疵檢測精準度
Massive Digital Engineering AOI Intelligent Robotic Arm Inspection System Significantly Improves Defect Detection Accuracy

Taiwan is known as a manufacturing powerhouse, yet quality defect detection has always been a chronic sore point in production lines While AOI equipment is available to assist, most use fixed machinery which are limited by angles, resulting in less precise diagnostics and high false positive rates Massive Digital Engineering introduced an AOI intelligent robotic arm detection system that effectively reduces false positives and increases the accuracy of defect detection Generally, the yield rate of products affects the costs for enterprises and the return rate for customers The quality defect detection process in the manufacturing industry often necessitates a substantial amount of quality inspection labor Although there is AOI equipment to assist, these tools are mostly fixed detection machines Fixed cameras are easily limited by angles, resulting in less precise diagnostics and high false positive rates Thus, personnel need to re-screen and inspect afterwards, often manually visual inspection misses defects on average about 5, and can be as high as 20 Three major pain points in manufacturing quality detection Robotic Arm AOI with dynamic multi-angle inspection helps to solve these issues According to the practical understanding by Massive Digital Engineering, there are three main pain points in detecting product quality within the manufacturing industry Pain point one, manual inspection of product quality is prone to errors Currently, the manufacturing industry largely relies on human labor to inspect product appearance, but human judgment often entails errors, such as surface scratches, color differences, solder appearance, etc The error rate in defect judgment is high, and can only be inspected at the finished product stage, often leading to whole batch rejections and high costs in labor and production Pain point two, inability to quantify and record data from quality inspections Traditional manual inspections do not maintain inspection data, which makes it difficult to assign responsibility when quality disputes occur Moreover, high-end contract manufacturing orders from overseas brands often require traceability and corresponding defect records, which traditional human inspection methods struggle to meet Pain point three, limitations of traditional AOI visual inspection systems Current manufacturing uses AOI visual inspection systems, which due to the limitations of visual software technology, employ fixed cameras, fixed lighting, and single-angle operations This method may handle flat or linear-shaped products like rectangular or square items at a single inspection point However, it is more challenging to implement for products with complex shapes eg, irregular automotive parts, requiring multi-point and multi-degree inspections Massive Digital Engineering developed an AOI intelligent robotic arm detection system, effectively improving the accuracy of defect detection To address the pain points in quality inspection in manufacturing, Massive Digital Engineering initiated the concept of developing a multi-angle, movable inspection device, starting with the combination of two representative technologies in factory automation - robotic arms and machine vision By integrating robotic arms with AOI for dynamic multi-angle AI real-time quality inspection, the limitations of fixed inspection systems are addressed, and visual inspection techniques are enhanced by leveraging artificial intelligence, further elevating the sampling of images from flat to multi-dimensional and multi-angular Selected the automotive industry as the real-world testing ground to quickly respond to customer needs The AOI intelligent robotic arm detection system, utilizing AI technology including unsupervised learning, supervised learning, and semi-supervised learning, allows operators to use unsupervised deep learning techniques to learn about good products even when initial samples are incomplete or there are no defective samples, applying it in the visual inspection of automatic welding of car trusses This can solve issues of limited angles with fixed machinery before implementation, less precise diagnostics, and high false positive rates Automotive components are high in unit price and demand a stricter defect detection accuracy In industries that have adopted AI services, the automotive manufacturing sector was chosen as the real-world testing ground Massive Digital Engineering states that the automotive industry mainly consists of related component manufacturers and components typically have a higher unit price, hence requiring more in terms of quality inspection and yield rates, and demanding stricter accuracy Therefore, the automotive sector was chosen as the area for introduction By using a robotic arm combined with AI for dynamic multi-angle AOI visual real-time quality inspection, not only can the defect quality error rate of automotive components be improved, but the fixed-point AOI optical inspection can be enhanced to meet the measurement needs of most industries and finally, establishing a third-party system platform to build an integrated monitoring system platform, enabling immediate response and action when issues arise This system allows for recording and storing important data of products leaving the factory, serving as a basis for future digital production lines and virtual production At the same time, in the event of defects, it can immediately connect to Massive's MES monitoring system, quickly responding to the relevant manufacturing decision-making department, subsequently utilizing ERP systems for project management and reviews, effectively improving production efficiency and reducing production costs Helps to reduce communication costs and aims to become an industry standard In terms of industry integration, it provides a foundational standard for data continuity among upstream and downstream businesses, reducing communication costs within the supply chain Through certification of the contract manufacturers and brand owners, there is a chance to become the industry standard configuration Through the data database established by this project, operators can further optimize their supply chain management solutions using big data analysis Data Analysis, based on data, establish forecast planning, and utilizing technology to link upstream and downstream data of the supply chain, accurately controlling product quality In the future, when interfacing with European, American, and Japanese markets, which demand highly fine-tuned orders, operators can respond and integrate the industry supply chain Supply Chain more swiftly Ultimately, through the benchmark demonstration industry's field verification, such as with the automotive component manufacturing industry used as the benchmark demonstration field, by implementing the robotic arm combined with AI for dynamic multi-angle AOI visual real-time quality inspection system project, the supply chain connection between automotive contract manufacturers and OEMs can be optimized, becoming the industry standard Further seeking more AI teams to join the cross-industry development on the field collaboration platform, driving the overall ecosystem combining AI innovation with field application Self-driving vehicle developed by Massive Digital Engineering「Translated content is generated by ChatGPT and is for reference only Translation date:2024-05-19」