☁️🧠 Automated Cloud-to-Edge Deployment of Industrial AI Models with Siemens Industrial Edge
Due to the sensitive nature of OT systems, a cloud-to-edge deployment can become a challenge. Specialized hardware devices are required, strict network protection is applied, and security policies are in place. Data can only be pulled by an intermediate factory IT system from where it can be deployed to the OT systems through highly controlled processes.
The following solution describes the “pull” deployment mechanism by using AWS services and Siemens Industrial AI software portfolio. The deployment process is enabled by three main components, the first of which is the Siemens AI Software Development Kit (AI SDK). After a model is created by a data scientist on Amazon SageMaker and stored in the SageMaker model registry, this SDK allows users to package a model in a format suitable for edge deployment using Siemens Industrial Edge. The second component, and the central connection between cloud and edge, is the Siemens AI Model Manager (AI MM). The third component is the Siemens AI Inference Server (AIIS), a specialized and hardened AI runtime environment running as a container on Siemens IEDs deployed on the shopfloor. The AIIS receives the packaged model from AI MM and is responsible to load, execute, and monitor ML models close to the production lines.
Machine Learning Platform at Walmart
Walmart is the world’s largest retailer, and it handles a huge volume of products, distribution, and transactions through its physical stores and online stores. Walmart has a highly optimized supply chain that runs at scale to offer its customers shopping at lowest price. In the process, Walmart accumulates a huge amount of valuable information from its everyday operations. This data is used to build Artificial Intelligence (AI) solutions to optimize and increase efficiencies of operations and customer experience atWalmart. In this paper, we provide an overview of the guiding principles, technology architecture, and integration of various tools within Walmart and from the open-source committee in building the Machine Learning (ML) Platform. We present multiple ML use cases at Walmart and show how their solutions leverage this ML Platform. We then discuss the business impact of having a scalable ML platform and infrastructure, reflect on lessons learnt building and operating an ML platform and future work for it at Walmart.
IBM and AWS partnering to transform industrial welding with AI and machine learning
IBM Smart Edge for Welding on AWS utilizes audio and visual capturing technology developed in collaboration with IBM Research. Using visual and audio recordings taken at the time of the weld, state-of-the-art artificial intelligence and machine learning models analyze the quality of the weld. If the quality does not meet standards, alerts are sent, and remediation action can take place without delay.
The solution substantially reduces the time between detection and remediation of defects, as well as the number of defects on the manufacturing line. By leveraging a combination of optical, thermal, and acoustic insights during the weld inspection process, two key manufacturing personas can better determine whether a welding discontinuity may result in a defect that will cost time and money: weld technician and process engineer.
A simpler method for learning to control a robot
Researchers from MIT and Stanford University have devised a new machine-learning approach that could be used to control a robot, such as a drone or autonomous vehicle, more effectively and efficiently in dynamic environments where conditions can change rapidly.
The researchers’ approach incorporates certain structure from control theory into the process for learning a model in such a way that leads to an effective method of controlling complex dynamics, such as those caused by impacts of wind on the trajectory of a flying vehicle. With this structure, they can extract a controller directly from the dynamics model, rather than using data to learn an entirely separate model for the controller.
The researchers also found that their method was data-efficient, which means it achieved high performance even with few data. For instance, it could effectively model a highly dynamic rotor-driven vehicle using only 100 data points. Methods that used multiple learned components saw their performance drop much faster with smaller datasets.
Using ML For Improved Fab Scheduling
The exact number of available tools for each step varies as tools are taken offline for maintenance or repairs. Some steps, like diffusion furnaces, consolidate multiple lots into large batches. Some sequences, like photoresist processing, must adhere to stringent time constraints. Lithography cells must match wafers with the appropriate reticles. Lot priorities change continuously. Even the time needed for an individual process step may change, as run-to-run control systems adjust recipe times for optimal results.
At the fab level, machine learning can support improved cycle time prediction and capacity planning. At the process cell or cluster tool level, it can inform WIP scheduling decisions. In between, it can facilitate better load balancing and order dispatching. As a first step, though, all of these applications need accurate models of the fab environment, which is a difficult problem.
The GlobalFoundries group demonstrated the effectiveness of neural network methods for time constraint tunnel dispatching. The relationship between input parameters and cycle time is complex and non-linear. As discussed above, machine learning methods are especially useful in situations like this, where statistical data is available but exact modeling is difficult.
Digital twins for the rapid startup of manufacturing processes: a case study in PVC tube extrusion
In this work, a soft sensor–based digital twin (DT) was developed to reduce the startup time in manufacturing plastic tubes and enable real-time product quality monitoring, i.e., the weight per unit length and the inner and outer diameters of the tube. An experimental campaign was conducted on a real tube extrusion line using three polyvinyl chloride (PVC) compounds and different process conditions, and machine learning regression algorithms were trained and tested to create the models of the extruder and the extrusion die the DT is based on. The characterization of the considered material, whose properties were given as input to the digital models, was carried out according to a procedure based only on the data collected by the production line. The DT was tested for the startup of the production of a single-layer tube and allowed to achieve the specified customer requirements (thickness and weight) in a few minutes. The proposed solution thus proved to be a valuable tool for reducing the setup time, thus increasing the efficiency of the process.
🖨️ Visual quality control in additive manufacturing: Building a complete pipeline
In this article, we share a reference implementation of a VQC pipeline for additive manufacturing that detects defects and anomalies on the surface of printed objects using depth-sensing cameras. We show how we developed an innovative solution to synthetically generate point clouds representing variations on 3D objects, and propose multiple machine learning models for detecting defects of different sizes. We also provide a comprehensive comparison of different architectures and experimental setups. The complete reference implementation is available in our git repository.
The main objective of this solution is to develop an architecture that can effectively learn from a sparse dataset, and is able to detect defects on a printed object by controlling the surface of the printed object each time a new layer is added. To address the challenge of acquiring a sufficient quantity of defect anomalies data for accurate ML model training, the proposed approach leverages a synthetic data generation approach. The controlled nature of the additive manufacturing process reduces the presence of unaccounted exogenous variables, making synthetic data a valuable resource for initial model training. In addition to this, we hypothesize that by deliberately inducing overfitting of the model on good examples, the model will become more accurate in predicting the presence of anomalies/defects. To achieve this, we generate a number of normal examples with introduced noise in a ratio that balances the defects occurrence expected during the manufacturing process. For instance, if the fault ratio is 10 to 1, we generate 10 similar normal examples for every 1 defect example. Hence, the pipeline for initial training consists of two modules: the synthetic generation module and the module for training anomaly detection models.
The right tool for the right job – ML and Design of Experiments
Typical statistical DOE software assumes that the response of experimental outputs to inputs is linear, or at best quadratic. ML makes no such assumption. Its models learn from the data provided even when that data contains complex, non-linear relationships. So ML can model difficult multi-component systems where cross-correlations would not be accounted for by other DOE approaches.
Standard DOE methods usually require you to vary only a limited number of inputs at any one time in your experimental design. With ML, you don’t have to identify which inputs are most important (thus potentially building bias into your design). You can ask the ML to explore all of the inputs simultaneously and it will find those that are most significant.
The Impact Of Machine Learning On Chip Design
🧠 Data-Driven Wind Farm Control via Multiplayer Deep Reinforcement Learning
This brief proposes a novel data-driven control scheme to maximize the total power output of wind farms subject to strong aerodynamic interactions among wind turbines. The proposed method is model-free and has strong robustness, adaptability, and applicability. Particularly, distinct from the state-of-the-art data-driven wind farm control methods that commonly use the steady-state or time-averaged data (such as turbines’ power outputs under steady wind conditions or from steady-state models) to carry out learning, the proposed method directly mines in-depth the time-series data measured at turbine rotors under time-varying wind conditions to achieve farm-level power maximization. The control scheme is built on a novel multiplayer deep reinforcement learning method (MPDRL), in which a special critic–actor–distractor structure, along with deep neural networks (DNNs), is designed to handle the stochastic feature of wind speeds and learn optimal control policies subject to a user-defined performance metric. The effectiveness, robustness, and scalability of the proposed MPDRL-based wind farm control method are tested by prototypical case studies with a dynamic wind farm simulator (WFSim). Compared with the commonly used greedy strategy, the proposed method leads to clear increases in farm-level power generation in case studies.
This AI Hunts for Hidden Hoards of Battery Metals
The mining industry’s rate of successful exploration—meaning the number of big deposit discoveries found per dollar invested—has been declining for decades. At KoBold, we sometimes talk about “Eroom’s law of mining.” As its reversed name suggests, it’s like the opposite of Moore’s law. In accordance with Eroom’s law of mining, the number of ore deposits discovered per dollar of capital invested has decreased by a factor of 8 over the last 30 years. (The original Eroom’s law refers to a similar trend in the cost of new pharmaceutical discoveries.)
Our exploration program in northern Quebec provides a good case study. We began by using machine learning to predict where we were most likely to find nickel in concentrations significant enough to be worth mining. We train our models using any available data on a region’s underlying physics and geology, and supplement the results with expert insights from our geologists. In Quebec, the models pointed us to land less than 20 km from currently operating mines.
Over the course of the summer in Quebec, we drilled 10 exploration holes, each more than a kilometer away from the last. Each drilling location was determined by combining the results from our predictive models with the expert judgment of our geologists. In each instance, the collected data indicated we’d find conductive bodies in the right geologic setting—possible minable ore deposits, in other words—below the surface. Ultimately, we hit nickel-sulfide mineralization in 8 of the 10 drill holes, which equates to easily 10 times better than the industry average for similarly isolated drill holes.
HAYAT HOLDING uses Amazon SageMaker to increase product quality and optimize manufacturing output, saving $300,000 annually
In this post, we share how HAYAT HOLDING—a global player with 41 companies operating in different industries, including HAYAT, the world’s fourth-largest branded diaper manufacturer, and KEAS, the world’s fifth-largest wood-based panel manufacturer—collaborated with AWS to build a solution that uses Amazon SageMaker Model Training, Amazon SageMaker Automatic Model Tuning, and Amazon SageMaker Model Deployment to continuously improve operational performance, increase product quality, and optimize manufacturing output of medium-density fiberboard (MDF) wood panels.
Quality prediction using ML is powerful but requires effort and skill to design, integrate with the manufacturing process, and maintain. With the support of AWS Prototyping specialists, and AWS Partner Deloitte, HAYAT HOLDING built an end-to-end pipeline. Product quality prediction and adhesive consumption recommendation results can be observed by field experts through dashboards in near-real time, resulting in a faster feedback loop. Laboratory results indicate a significant impact equating to savings of $300,000 annually, reducing their carbon footprint in production by preventing unnecessary chemical waste.
Better spinach through AI: Tokyo startup automates seedling selection
A Japanese agricultural startup has developed technology that uses artificial intelligence to assess the growth and potential of spinach seedlings, aiming to reduce food loss by increasing yields and efficiency.
The AI system has two parts. The first uses photographs to estimate the height, width and weight of seedlings grown in plant factories. The other predicts future growth using an index developed by Farmship. The first eliminates seedlings that are obviously not growing well, and the other narrows the remaining seedlings to only superior ones, making harvesting easier. In trials, the ratio of seedlings that grew properly increased to 80%, from 54% using standard methods. This corresponds to a 17% harvest increase.
Industrial defect detection at the edge
CAD-based data augmentation and transfer learning empowers part classification in manufacturing
Especially in manufacturing systems with small batches or customized products, as well as in remanufacturing and recycling facilities, there is a wide variety of part types that may be previously unseen. It is crucial to accurately identify these parts based on their type for traceability or sorting purposes. One approach that has shown promising results for this task is deep learning–based image classification, which can classify a part based on its visual appearance in camera images. However, this approach relies on large labeled datasets of real-world images, which can be challenging to obtain, especially for parts manufactured for the first time or whose appearance is unknown. To overcome this challenge, we propose generating highly realistic synthetic images based on photo-realistically rendered computer-aided design (CAD) data. Using this commonly available source, we aim to reduce the manual effort required for data generation and preparation and improve the classification performance of deep learning models using transfer learning. In this approach, we demonstrate the creation of a parametric rendering pipeline and show how it can be used to train models for a 30-class classification problem with typical engineering parts in an industrial use case. We also demonstrate how our method’s entropy gain improves the classification performance in various deep image classification models.
AI: how it’s delivering sharper route planning
Creating a route requires a dispatcher to answer a host of questions such as: “What is the wind today?”, “What is the best altitude for this flight?” and “Is there any military training?” Before the Flyways software, the 100 or so dispatchers at the NOC had to find answers to these questions by visiting multiple websites. These included FAA websites designed specifically for dispatchers, but that information was available only as strings of text that were hard to read.
Having decided to focus on the aviation industry, the team started spending an obscene amount of time at the NOC in an effort to understand how dispatching works and to create a user-friendly product — one that a real dispatcher could seamlessly operate when under pressure. Alaska Airlines’ employees would joke that the team was basically camping in their operations center with sleeping bags, Buckendorf says.
Flyways improves itself further by learning from a human dispatcher’s acceptance or rejection of its recommendations. When the dispatcher dismisses a suggestion, Flyways asks why: Was it because of the weather? Was the route putting an airplane uncomfortably close to somewhere it shouldn’t be? The idea is that Flyways learns from those decisions and evolves — though certain data points need to be filtered out so that the software does not simply emulate human dispatchers’ choices, stifling innovation.
📦 How AWS used ML to help Amazon fulfillment centers reduce downtime by 70%
The retail leader has announced it uses Amazon Monitron, an end-to-end machine learning (ML) system to detect abnormal behavior in industrial machinery — that launched in December 2020 — to provide predictive maintenance. As a result, Amazon has reduced unplanned downtime at the fulfillment centers by nearly 70%, which helps deliver more customer orders on time.
Monitron receives automatic temperature and vibration measurements every hour, detecting potential failures within hours, compared with 4 weeks for the previous manual techniques. In the year and a half since the fulfillment centers began using it, they have helped avoid about 7,300 confirmed issues across 88 fulfillment center sites across the world.
Closed-loop fully-automated frameworks for accelerating materials discovery
Our work shows that a fully-automated closed-loop framework driven by sequential learning can accelerate the discovery of materials by up to 10-25x (or a reduction in design time by 90-95%) when compared to traditional approaches. We show that such closed-loop frameworks can lead to enormous improvement in researcher productivity in addition to reducing overall project costs. Overall, these findings present a clear value proposition for investing in closed-loop frameworks and sequential learning in materials discovery and design enterprises.
UVA Research Team Detects Additive Manufacturing Defects in Real-Time
Introduced in the 1990s, laser powder bed fusion, or LPBF uses metal powder and lasers to 3-D print metal parts. But porosity defects remain a challenge for fatigue-sensitive applications like aircraft wings. Some porosity is associated with deep and narrow vapor depressions which are the keyholes.
“By integrating operando synchrotron x-ray imaging, near-infrared imaging, and machine learning, our approach can capture the unique thermal signature associated with keyhole pore generation with sub-millisecond temporal resolution and 100% prediction rate,” Sun said. In developing their real-time keyhole detection method, the researchers also advanced the way a state-of-the-art tool — operando synchrotron x-ray imaging — can be used. Utilizing machine learning, they additionally discovered two modes of keyhole oscillation.
AI farming tool from BASF finds fertile ground in Japan's rice country
Yamazaki Rice, based near Tokyo in Saitama prefecture, began using BASF’s Xarvio Field Manager system this year with five workers on about 100 hectares of land.
Xarvio provides real-time analysis informed by satellite and weather data. Automated maps customize the amount of fertilizer recommended for each section of the farm. The data is fed to GPS-equipped farm equipment. The AI gives daily suggestions that Yamazaki Rice’s president said helped improve yields by up to 25% in some fields. Xarvio’s machine learning covers more than 10 years of crop data as well as scientific papers worldwide.
How a universal model is helping one generation of Amazon robots train the next
In short, building a dataset big enough to train a demanding machine learning model requires time and resources, with no guarantee that the novel robotic process you are working toward will prove successful. This became a recurring issue for Amazon Robotics AI. So this year, work began in earnest to address the data scarcity problem. The solution: a “universal model” able to generalize to virtually any package segmentation task.
To develop the model, Meeker and her colleagues first used publicly available datasets to give their model basic classification skills — being able to distinguish boxes or packages from other things, for example. Next, they honed the model, teaching it to distinguish between many types of packaging in warehouse settings — from plastic bags to padded mailers to cardboard boxes of varying appearance — using a trove of training data compiled by the Robin program and half a dozen other Amazon teams over the last few years. This dataset comprised almost half a million annotated images.
The universal model now includes images of unpackaged items, too, allowing it to perform segmentation across a greater diversity of warehouse processes. Initiatives such as multimodal identification, which aims to visually identify items without needing to see a barcode, and the automated damage detection program are accruing product-specific data that could be fed into the universal model, as well as images taken on the fulfillment center floor by the autonomous robots that carry crates of products.
Automated Optical Inspection
Machine-Learning-Enhanced Simulation Could Reduce Energy Costs in Materials Production
Thanks to a new computational effort being pioneered by the U.S. Department of Energy’s (DOE) Argonne National Laboratory in conjunction with 3M and supported by the DOE’S High Performance Computing for Energy Innovation (HPC4EI) program, researchers are finding new ways to dramatically reduce the amount of energy required for melt blowing the materials needed in N95 masks and other applications.
Currently, the process used to create a nozzle to spin nonwoven materials produces a very high-quality product, but it is quite energy intensive. Approximately 300,000 tons of melt-blown materials are produced annually worldwide, requiring roughly 245 gigawatt-hours per year of energy, approximately the amount generated by a large solar farm. By using Argonne supercomputing resources to pair computational fluid dynamics simulations and machine-learning techniques, the Argonne and 3M collaboration sought to reduce energy consumption by 20% without compromising material quality.
Because the process of making a new nozzle is very expensive, the information gained from the machine-learning model can equip material manufacturers with a way to narrow down to a set of optimal designs. ”Machine-learning-enhanced simulation is the best way of cheaply getting at the right combination of parameters like temperatures, material composition, and pressures for creating these materials at high quality with less energy,” Blaiszik said.
Machine learning facilitates “turbulence tracking” in fusion reactors
A multidisciplinary team of researchers is now bringing tools and insights from machine learning to aid this effort. Scientists from MIT and elsewhere have used computer-vision models to identify and track turbulent structures that appear under the conditions needed to facilitate fusion reactions.
Monitoring the formation and movements of these structures, called filaments or “blobs,” is important for understanding the heat and particle flows exiting from the reacting fuel, which ultimately determines the engineering requirements for the reactor walls to meet those flows. However, scientists typically study blobs using averaging techniques, which trade details of individual structures in favor of aggregate statistics. Individual blob information must be tracked by marking them manually in video data.
The researchers built a synthetic video dataset of plasma turbulence to make this process more effective and efficient. They used it to train four computer vision models, each of which identifies and tracks blobs. They trained the models to pinpoint blobs in the same ways that humans would.
When the researchers tested the trained models using real video clips, the models could identify blobs with high accuracy — more than 80 percent in some cases. The models were also able to effectively estimate the size of blobs and the speeds at which they moved.
Machine learning-aided engineering of hydrolases for PET depolymerization
Plastic waste poses an ecological challenge1,2,3 and enzymatic degradation offers one, potentially green and scalable, route for polyesters waste recycling4. Poly(ethylene terephthalate) (PET) accounts for 12% of global solid waste5, and a circular carbon economy for PET is theoretically attainable through rapid enzymatic depolymerization followed by repolymerization or conversion/valorization into other products6,7,8,9,10. Application of PET hydrolases, however, has been hampered by their lack of robustness to pH and temperature ranges, slow reaction rates and inability to directly use untreated postconsumer plastics11. Here, we use a structure-based, machine learning algorithm to engineer a robust and active PET hydrolase. Our mutant and scaffold combination (FAST-PETase: functional, active, stable and tolerant PETase) contains five mutations compared to wild-type PETase (N233K/R224Q/S121E from prediction and D186H/R280A from scaffold) and shows superior PET-hydrolytic activity relative to both wild-type and engineered alternatives12 between 30 and 50 °C and a range of pH levels. We demonstrate that untreated, postconsumer-PET from 51 different thermoformed products can all be almost completely degraded by FAST-PETase in 1 week. FAST-PETase can also depolymerize untreated, amorphous portions of a commercial water bottle and an entire thermally pretreated water bottle at 50 ºC. Finally, we demonstrate a closed-loop PET recycling process by using FAST-PETase and resynthesizing PET from the recovered monomers. Collectively, our results demonstrate a viable route for enzymatic plastic recycling at the industrial scale.
CircularNet: Reducing waste with Machine Learning
The facilities where our waste and recyclables are processed are called “Material Recovery Facilities” (MRFs). Each MRF processes tens of thousands of pounds of our societal “waste” every day, separating valuable recyclable materials like metals and plastics from non-recyclable materials. A key inefficiency within the current waste capture and sorting process is the inability to identify and segregate waste into high quality material streams. The accuracy of the sorting directly determines the quality of the recycled material; for high-quality, commercially viable recycling, the contamination levels need to be low. Even though the MRFs use various technologies alongside manual labor to separate materials into distinct and clean streams, the exceptionally cluttered and contaminated nature of the waste stream makes automated waste detection challenging to achieve, and the recycling rates and the profit margins stay at undesirably low levels.
Enter what we call “CircularNet”, a set of models that lowers barriers to AI/ML tech for waste identification and all the benefits this new level of transparency can offer. Our goal with CircularNet is to develop a robust and data-efficient model for waste/recyclables detection, which can support the way we identify, sort, manage, and recycle materials across the waste management ecosystem.
Lufthansa increases on-time flights by wind forecasting with Google Cloud ML
The magnitude and direction of wind significantly impacts airport operations, and Lufthansa Group Airlines are no exception. A particularly troublesome kind is called BISE: it is a cold, dry wind that blows from the northeast to southwest in Switzerland, through the Swiss Plateau. Its effects on flight schedules can be severe, such as forcing planes to change runways, which can create a chain reaction of flight delays and possible cancellations. In Zurich Airport, in particular, BISE can potentially reduce capacity by up to 30%, leading to further flight delays and cancellations, and to millions in lost revenue for Lufthansa (as well as dissatisfaction among their passengers).
Machine learning (ML) can help airports and airlines to better anticipate and manage these types of disruptive weather events. In this blog post, we’ll explore an experiment Lufthansa did together with Google Cloud and its Vertex AI Forecast service, accurately predicting BISE hours in advance, with more than 40% relative improvement in accuracy over internal heuristics, all within days instead of the months it often takes to do ML projects of this magnitude and performance.
Improving Yield With Machine Learning
Machine learning is becoming increasingly valuable in semiconductor manufacturing, where it is being used to improve yield and throughput.
Synopsys engineers recently found that a decision tree deep learning method can classify 98% of defects and features at 60X faster retraining time than traditional CNNs. The decision tree utilizes 8 CNNs and ResNet to automatically classify 12 defect types with images from SEM and optical tools.
Macronix engineers showed how machine learning can expedite new etch process development in 3D NAND devices. Two parameters are particularly important in optimizing the deep trench slit etch — bottom CD and depth of polysilicon etch recess, also known as the etch stop.
KLA engineers, led by Cheng Hung Wu, optimized the use of a high landing energy e-beam inspection tool to capture defects buried as deep as 6µm in a 96-layer ONON stacked structure following deep trench etch. The e-beam tool can detect defects that optical inspectors cannot, but only if operated with high landing energy to penetrate deep structures. With this process, KLA was looking to develop an automated detection and classification system for deep trench defects.
“We see AI as a disruptive technology that will in the long run eliminate, and in the near term reduce the need for verification,” says Anupam Bakshi, CEO and founder of Agnisys. “We have had some early successes in using machine learning to read user specifications in natural language and directly convert them into SystemVerilog Assertions (SVA), UVM testbench code, and C/C++ embedded code for test and verification.”
There is nothing worse than spending time and resources to not get the desired result, or for it to take longer than necessary. “In formal, we have multiple engines, different algorithms that are working on solving any given property at any given time,” says Pete Hardee, director for product management at Cadence. “In effect, there is an engine race going on. We track that race and see for each property which engine is working. We use reinforcement learning to set the engine parameters in terms of which engines I’m going to use and how long to run those to get better convergence on the properties that didn’t converge the first time I ran it.”
Batch Optimization using Quartic.ai
Using machine learning techniques in wine quality testing
The Profiling capability from Thermo Scientific™ SampleManager™ LIMS software provides an innovative way for laboratories to predict test results using historical data and novel machine learning (ML)-based techniques. For example, a food and beverage company might apply the Profiling capability to enable supervised learning in the food production process. In this case, SampleManager LIMS would use historical data to gain an understanding of the critical variables that determine whether a product is safe for consumers. This holistic approach considers not only the values of the individual critical variables themselves, but also the relationships between them. If a sample were to be flagged as failing, the system would alert stakeholders in advance to issue adjustments or investigations to avoid any risk to finished products.
In a wine production facility, the result of the “Quality Test” is of utmost importance. The laboratory has great flexibility and control over the testing process, so they could use the Profiling capability to redefine the order of the standard tests conducted to a wine sample.
Ericsson’s next-gen AI-driven network dimensioning solution
Resource requirement estimation, often referred to as dimensioning, is a crucial activity in the telecommunications industry. Network dimensioning is an integral part of the Ericsson Sales Process when engaging with a prospective customer – find out more about our approach to network dimensioning and the critical importance of accuracy.
The telco dimensioning problem can be conceived as a regression problem from an AI/ML perspective. The proposed solution is Bayesian Regression which proved to be more robust to multi-collinearity of features. Additionally, our approach allows the incorporation of domain knowledge into the modeling (for example, in the form of priors, bounds and constraints), to avoid dropping network features that are critical for the domain and interpretability requirements, from a model’s trustworthiness perspective.
Decentralized learning and intelligent automation: the key to zero-touch networks?
Decentralized learning and the multi-armed bandit agent… It may sound like the sci-fi version of an old western. But could this dynamic duo hold the key to efficient distributed machine learning – a crucial factor in the realization of zero-touch automated mobile networks? Let’s find out.
Next-generation autonomous mobile networks will be complex ecosystems made up of a massive number of decentralized and intelligent network devices and nodes – network elements that may be both producing and consuming data simultaneously. If we are to realize our goal of fully automated zero-touch networks, new models of training artificial intelligence (AI) models need to be developed to accommodate these complex and diverse ecosystems.
How Drishti empowers deep learning in manufacturing
During his talk at the MLDS Conference, ‘New developments in Deep Learning for unlikely industries’, Shankar outlined Drishti’s industrial applications of AI in manufacturing. The company leverages deep learning and computer vision to automate the analysis of factory floor videos. Essentially, the company has installed cameras on assembly lines that capture videos on which the company runs object detection, anomaly detection and action recognition. Then, the data is sent to industrial engineers to improve the line.
Fingerprinting liquids for composites
Collo uses electromagnetic sensors and edge analytics to optimize resin degassing, mixing, infusion, polymerization and cure as well as monitoring drift from benchmarked process parameters and enabling in-situ process control.
“So, the solution we are offering is real-time, inline measurement directly from the process,” says Järveläinen. “Our system then converts that data into physical quantities that are understandable and actionable, like rheological viscosity, and it helps to ensure high-quality liquid processes and products. It also allows optimizing the processes. For example, you can shorten mixing time because you can clearly see when mixing is complete. So, you can improve productivity, save energy and reduce scrap versus less optimized processing.”
Why AI software companies are betting on small data to spot manufacturing defects
The deep-learning algorithms that have come to dominate many of the technologies consumers and businesspeople interact with today are trained and improved by ingesting huge quantities of data. But because product defects show up so rarely, most manufacturers don’t have millions, thousands or even hundreds of examples of a particular type of flaw they need to watch out for. In some cases, they might only have 20 or 30 photos of a windshield chip or small pipe fracture, for example.
Because labeling inconsistencies can trip up deep-learning models, Landing AI aims to alleviate the confusion. The company’s software has features that help isolate inconsistencies and assist teams of inspectors in coming to agreement on taxonomy. “The inconsistencies in labels are pervasive,” said Ng. “A lot of these problems are fundamentally ambiguous.”
How pioneering deep learning is reducing Amazon’s packaging waste
Fortunately, machine learning approaches — particularly deep learning — thrive on big data and massive scale, and a pioneering combination of natural language processing and computer vision is enabling Amazon to hone in on using the right amount of packaging. These tools have helped Amazon drive change over the past six years, reducing per-shipment packaging weight by 36% and eliminating more than a million tons of packaging, equivalent to more than 2 billion shipping boxes.
“When the model is certain of the best package type for a given product, we allow it to auto-certify it for that pack type,” says Bales. “When the model is less certain, it flags a product and its packaging for testing by a human.” The technology is currently being applied to product lines across North America and Europe, automatically reducing waste at a growing scale.
Transfer learning with artificial neural networks between injection molding processes and different polymer materials
Finding appropriate machine setting parameters in injection molding remains a difficult task due to the highly nonlinear process behavior. Artificial neural networks are a well-suited machine learning method for modelling injection molding processes, however, it is costly and therefore industrially unattractive to generate a sufficient amount of process samples for model training. Therefore, transfer learning is proposed as an approach to reuse already collected data from different processes to supplement a small training data set. Process simulations for the same part and 60 different materials of 6 different polymer classes are generated by design of experiments. After feature selection and hyperparameter optimization, finetuning as transfer learning technique is proposed to adapt from one or more polymer classes to an unknown one. The results illustrate a higher model quality for small datasets and selective higher asymptotes for the transfer learning approach in comparison with the base approach.
Using Machine Learning in Bosch IoT Insights
Artificial intelligence optimally controls your plant
Until now, heating systems have mainly been controlled individually or via a building management system. Building management systems follow a preset temperature profile, meaning they always try to adhere to predefined target temperatures. The temperature in a conference room changes in response to environmental influences like sunlight or the number of people present. Simple (PI or PID) controllers are used to make constant adjustments so that the measured room temperature is as close to the target temperature values as possible.
We believe that the best alternative is learning a control strategy by means of reinforcement learning (RL). Reinforcement learning is a machine learning method that has no explicit (learning) objective. Instead, an “agent” with as complete a knowledge of the system state as possible learns the manipulated variable changes that maximize a “reward” function defined by humans. Using algorithms from reinforcement learning, the agent, meaning the control strategy, can be trained from both current and recorded system data. This requires measurements for the manipulated variable changes that have been carried out, for the (resulting) changes to the system state over time, and for the variables necessary for calculating the reward.
Quality prediction of ultrasonically welded joints using a hybrid machine learning model
Ultrasonic metal welding has advantages over other joining technologies due to its low energy consumption, rapid cycle time and the ease of process automation. The ultrasonic welding (USW) process is very sensitive to process parameters, and thus can be difficult to consistently produce strong joints. There is significant interest from the manufacturing community to understand these variable interactions. Machine learning is one such method which can be exploited to better understand the complex interactions of USW input parameters. In this paper, the lap shear strength (LSS) of USW Al 5754 joints is investigated using an off-the-shelf Branson Ultraweld L20. Firstly, a 33 full factorial parametric study using ANOVA is carried out to examine the effects of three USW input parameters (weld energy, vibration amplitude & clamping pressure) on LSS. Following this, a high-fidelity predictive hybrid GA-ANN model is then trained using the input parameters and the addition of process data recorded during welding (peak power).
Machine learning predictions of superalloy microstructure
Gaussian process regression machine learning with a physically-informed kernel is used to model the phase compositions of nickel-base superalloys. The model delivers good predictions for laboratory and commercial superalloys. Additionally, the model predicts the phase composition with uncertainties unlike the traditional CALPHAD method.
Graph-based semi-supervised random forest for rotating machinery gearbox fault diagnosis
Random forest (RF) is an effective method for diagnosing faults of rotating machinery. However, the diagnosis accuracy enhancement under insufficient labeled samples is still one of the main challenges. Motivated by this problem, an improved RF algorithm based on graph-based semi-supervised learning (GSSL) and decision tree is proposed in this paper to improve the classification accuracy in the absence of labeled samples. The unlabeled samples are annotated by the GSSL and verified by the decision tree. The trained improved RF model is applied to the fault diagnosis for the rotating machinery gearbox. The effectiveness of the proposed algorithm is verified via hardware experiments using a wind turbine drivetrain diagnostics simulator (WTDDS). The results show that the proposed algorithm achieves better accuracy of classification than conventional methods in gearbox fault diagnosis. This study leads to further progress in the improvement of machine learning methods with insufficient and unlabeled samples.
Hybrid machine learning-enabled adaptive welding speed control
This research presents a preliminary study on developing appropriate Machine Learning (ML) techniques for real-time welding quality prediction and adaptive welding speed adjustment for GTAW welding at a constant current. In order to collect the data needed to train the hybrid ML models, two cameras are applied to monitor the welding process, with one camera (available in practical robotic welding) recording the top-side weld pool dynamics and a second camera (unavailable in practical robotic welding, but applicable for training purpose) recording the back-side bead formation. Given these two data sets, correlations can be discovered through a convolutional neural network (CNN) that is good at image characterization. With the CNN, top-side weld pool images can be analyzed to predict the back-side bead width during active welding control.
Fabs Drive Deeper Into Machine Learning
For the past couple decades, semiconductor manufacturers have relied on computer vision, which is one of the earliest applications of machine learning in semiconductor manufacturing. Referred to as Automated Optical Inspection (AOI), these systems use signal processing algorithms to identify macro and micro physical deformations.
Defect detection provides a feedback loop for fab processing steps. Wafer test results produce bin maps (good or bad die), which also can be analyzed as images. Their data granularity is significantly larger than the pixelated data from an optical inspection tool. Yet test results from wafer maps can match the splatters generated during lithography and scratches produced from handling that AOI systems can miss. Thus, wafer test maps give useful feedback to the fab.
Adoption of machine learning technology for failure prediction in industrial maintenance: A systematic review
Failure prediction is the task of forecasting whether a material system of interest will fail at a specific point of time in the future. This task attains significance for strategies of industrial maintenance, such as predictive maintenance. For solving the prediction task, machine learning (ML) technology is increasingly being used, and the literature provides evidence for the effectiveness of ML-based prediction models. However, the state of recent research and the lessons learned are not well documented. Therefore, the objective of this review is to assess the adoption of ML technology for failure prediction in industrial maintenance and synthesize the reported results. We conducted a systematic search for experimental studies in peer-reviewed outlets published from 2012 to 2020. We screened a total of 1,024 articles, of which 34 met the inclusion criteria.
Accelerating the Design of Automotive Catalyst Products Using Machine Learning
The design of catalyst products to reduce harmful emissions is currently an intensive process of expert-driven discovery, taking several years to develop a product. Machine learning can accelerate this timescale, leveraging historic experimental data from related products to guide which new formulations and experiments will enable a project to most directly reach its targets. We used machine learning to accurately model 16 key performance targets for catalyst products, enabling detailed understanding of the factors governing catalyst performance and realistic suggestions of future experiments to rapidly develop more effective products. The proposed formulations are currently undergoing experimental validation.
Getting Industrial About The Hybrid Computing And AI Revolution
Beyond Limits is applying such techniques as deep reinforcement learning (DRL), using a framework to train a reinforcement learning agent to make optimal sequential recommendations for placing wells. It also uses reservoir simulations and novel deep convolutional neural networks to work. The agent takes in the data and learns from the various iterations of the simulator, allowing it to reduce the number of possible combinations of moves after each decision is made. By remembering what it learned from the previous iterations, the system can more quickly whittle the choices down to the one best answer.
Real-World ML with Coral: Manufacturing
For over 3 years, Coral has been focused on enabling privacy-preserving Edge ML with low-power, high performance products. We’ve released many examples and projects designed to help you quickly accelerate ML for your specific needs. One of the most common requests we get after exploring the Coral models and projects is: How do we move to production?
- Worker Safety - Performs generic person detection (powered by COCO-trained SSDLite MobileDet) and then runs a simple algorithm to detect bounding box collisions to see if a person is in an unsafe region.
- Visual Inspection - Performs apple detection (using the same COCO-trained SSDLite MobileDet from Worker Safety) and then crops the frame to the detected apple and runs a retrained MobileNetV2 that classifies fresh vs rotten apples.
The Journey of Additive Manufacturing and Artificial Intelligence
The Machine Economy is Here: Powering a Connected World
In combination with the real-time data produced by IoT, blockchain, and ML applications are disrupting B2B companies across various industries from healthcare to manufacturing. Together, these three fundamental technologies create an intelligent system where connected devices can “talk” to one another. However, machines are still unable to conduct transactions with each other.
This is where distributed ledger technology (DLT) and blockchain come into play. Cryptocurrencies and smart contracts (self-executing contracts between buyers and sellers on a decentralized network) make it possible for autonomous machines to transact with one another on a blockchain.
Devices participating in M2M transactions can be programmed to make purchases based on individual or business needs. Human error was a cause for concern in the past; machine learning algorithms provide reliable and trusted data that continue to learn and improve — becoming smarter each day.
How to integrate AI into engineering
Most of the focus on AI is all about the AI model, which drives engineers to quickly dive into the modelling aspect of AI. After a few starter projects, engineers learn that AI is not just modelling, but rather a complete set of steps that includes data preparation, modelling, simulation and test, and deployment
Visual Inspection AI: a purpose-built solution for faster, more accurate quality control
The Google Cloud Visual Inspection AI solution automates visual inspection tasks using a set of AI and computer vision technologies that enable manufacturers to transform quality control processes by automatically detecting product defects.
We built Visual Inspection AI to meet the needs of quality, test, manufacturing, and process engineers who are experts in their domain, but not in AI. By combining ease of use with a focus on priority uses cases, customers are realizing significant benefits compared to general purpose machine learning (ML) approaches.
Machine Learning Keeps Rolling Bearings on the Move
Rolling bearings are essential components in automated machinery with rotating elements. They come in many shapes and sizes, but are essentially designed to carry a load while minimizing friction. In general, the design consists of two rings separated by rolling elements (balls or rollers). The rings can rotate can rotate relative to each other with very little friction.
The ability to accurately predict the remaining useful life of the bearings under defect progression could reduce unnecessary maintenance procedures and prematurely discarded parts without risking breakdown, reported scientists from the Institute of Scientific and Industrial Research and NTN Next Generation Research Alliance Laboratories at Osaka University.
The scientists have developed a machine learning method that combines convolutional neural networks and Bayesian hierarchical modeling to predict the remaining useful life of rolling bearings. Their approach is based on the measured vibration spectrum.
Tree Model Quantization for Embedded Machine Learning Applications
Compressed tree-based models are useful models to consider for embedded machine learning applications, in particular with the compression technique: quantization. Quantization can compress models by significant amounts with a trade-off of slight loss in model fidelity, allowing more room on the device for other programs.
The realities of developing embedded neural networks
With any embedded software destined for deployment in volume production, an enormous amount of effort goes into the code once the implementation of its core functionality has been completed and verified. This optimization phase is all about minimizing memory, CPU and other resources needed so that as much as possible of the software functionality is preserved, while the resources needed to execute it are reduced to the absolute minimum possible.
This process of creating embedded software from lab-based algorithms enables production engineers to cost-engineer software functionality into a mass-production ready form, requiring far cheaper, less capable chips and hardware than the massive compute datacenter used to develop it. However, it usually requires the functionality to be frozen from the beginning, with code modifications only done to improve the way the algorithms themselves are executed. For most software, that is fine: indeed, it enables a rigorous verification methodology to be used to ensure the embedding process retains all the functionality needed.
However, when embedding NN-based AI algorithms, that can be a major problem. Why? Because by freezing the functionality from the beginning, you are removing one of the main ways in which the execution can be optimized.
Google Cloud and Seagate: Transforming hard-disk drive maintenance with predictive ML
At Google Cloud, we know first-hand how critical it is to manage HDDs in operations and preemptively identify potential failures. We are responsible for running some of the largest data centers in the world—any misses in identifying these failures at the right time can potentially cause serious outages across our many products and services. In the past, when a disk was flagged for a problem, the main option was to repair the problem on site using software. But this procedure was expensive and time-consuming. It required draining the data from the drive, isolating the drive, running diagnostics, and then re-introducing it to traffic.
That’s why we teamed up with Seagate, our HDD original equipment manufacturer (OEM) partner for Google’s data centers, to find a way to predict frequent HDD problems. Together, we developed a machine learning (ML) system, built on top of Google Cloud, to forecast the probability of a recurring failing disk—a disk that fails or has experienced three or more problems in 30 days.
Ford's Ever-Smarter Robots Are Speeding Up the Assembly Line
At a Ford Transmission Plant in Livonia, Michigan, the station where robots help assemble torque converters now includes a system that uses AI to learn from previous attempts how to wiggle the pieces into place most efficiently. Inside a large safety cage, robot arms wheel around grasping circular pieces of metal, each about the diameter of a dinner plate, from a conveyor and slot them together.
The technology allows this part of the assembly line to run 15 percent faster, a significant improvement in automotive manufacturing where thin profit margins depend heavily on manufacturing efficiencies.
Start-ups Powering New Era of Industrial Robotics
Much of the bottleneck to achieving automation in manufacturing relates to limitations in the current programming model of industrial robotics. Programming is done in languages proprietary to each robotic hardware OEM – languages “straight from the 80s” as one industry executive put it.
There are a limited number of specialists who are proficient in these languages. Given the rarity of the expertise involved, as well as the time it takes to program a robot, robotics application development typically costs three times as much as the hardware for a given installation.
Multi-Task Robotic Reinforcement Learning at Scale
For general-purpose robots to be most useful, they would need to be able to perform a range of tasks, such as cleaning, maintenance and delivery. But training even a single task (e.g., grasping) using offline reinforcement learning (RL), a trial and error learning method where the agent uses training previously collected data, can take thousands of robot-hours, in addition to the significant engineering needed to enable autonomous operation of a large-scale robotic system. Thus, the computational costs of building general-purpose everyday robots using current robot learning methods becomes prohibitive as the number of tasks grows.
Intelligent edge management: why AI and ML are key players
What will the future of network edge management look like? We explain how artificial intelligence and machine learning technologies are crucial for intelligent edge computing and the management of future-proof networks. What’s required, and what are the building blocks needed to make it happen?
Using Machine Learning to identify operational modes in rotating equipment
Vibration monitoring is key to performing condition monitoring-based maintenance in rotating equipment such as engines, compressors, turbines, pumps, generators, blowers, and gearboxes. However, periodic route-based vibration monitoring programs are not enough to prevent breakdowns, as they normally offer a narrower view of the machines’ conditions.
Adding Machine Learning algorithms to this process makes it scalable, as it allows the analysis of historic data from equipment. One of the benefits is being able to identify operational modes and help maintenance teams to understand if the machine is operating in normal or abnormal conditions.
Amazon’s robot arms break ground in safety, technology
Robin, one of the most complex stationary robot arm systems Amazon has ever built, brings many core technologies to new levels and acts as a glimpse into the possibilities of combining vision, package manipulation and machine learning, said Will Harris, principal product manager of the Robin program.
Those technologies can be seen when Robin goes to work. As soft mailers and boxes move down the conveyor line, Robin must break the jumble down into individual items. This is called image segmentation. People do it automatically, but for a long time, robots only saw a solid blob of pixels.
AI In Inspection, Metrology, And Test
“The human eye can see things that no amount of machine learning can,” said Subodh Kulkarni, CEO of CyberOptics. “That’s where some of the sophistication is starting to happen now. Our current systems use a primitive kind of AI technology. Once you look at the image, you can see a problem. And our AI machine doesn’t see that. But then you go to the deep learning kind of algorithms, where you have very serious Ph.D.-level people programming one algorithm for a week, and they can detect all those things. But it takes them a week to program those things, which today is not practical.”
That’s beginning to change. “We’re seeing faster deep-learning algorithms that can be more easily programmed,” Kulkarni said. “But the defects also are getting harder to catch by a machine, so there is still a gap. The biggest bang for the buck is not going to come from improving cameras or projectors or any of the equipment that we use to generate optical images. It’s going to be interpreting optical images.”
How To Measure ML Model Accuracy
Machine learning (ML) is about making predictions about new data based on old data. The quality of any machine-learning algorithm is ultimately determined by the quality of those predictions.
However, there is no one universal way to measure that quality across all ML applications, and that has broad implications for the value and usefulness of machine learning.
Go beyond machine learning to optimize manufacturing operations
Machine learning depends on vast amounts of data to make inferences. However, sometimes the amount of data needed by machine-learning algorithms is simply not available. SRI International has developed a system called Deep Adaptive Semantic Logic (DASL) that uses adaptive semantic reasoning to fill in the data gaps. DASL integrates bottom-up data-driven modeling with top-down theoretical reasoning in a symbiotic union of innovative machine learning and knowledge guided inference. The system brings experts and data together to make better, more informed decisions.
Adversarial training reduces safety of neural networks in robots
A more fundamental problem, also confirmed by Lechner and his coauthors, is the lack of causality in machine learning systems. As long as neural networks focus on learning superficial statistical patterns in data, they will remain vulnerable to different forms of adversarial attacks. Learning causal representations might be the key to protecting neural networks against adversarial attacks. But learning causal representations itself is a major challenge and scientists are still trying to figure out how to solve it.
What Walmart learned from its machine learning deployment
As more businesses turn to automation to realize business value, retail’s wide variety of ML use cases can provide insights into how to overcome challenges associated with the technology. The goal should be trying to solve a problem by using ML as a tool to get there, Kamdar said.
For example, Walmart uses a ML model to optimize the timing and pricing of markdowns, and to examine real estate data to find places to cut costs, according to executives on an earnings call in February.
AI project to 'pandemic-proof' NHS supply chain
With the ability to analyse NHS and global procurement data from previous supply contracts, the platform will aim to allow NHS buyers to evaluate credibility and capability of suppliers to fulfil their order. Each supplier would have a real-time ‘risk rating’ with information on the goods and services they supply.
Researchers at Sheffield University’s Information School are said to be developing Natural Language Processing (NLP) methods for the automated reading and extraction of data from large amounts of contract tender data held by the NHS and other European healthcare providers
How Machine Learning Techniques Can Help Engineers Design Better Products
By leveraging field predictive ML models engineers can explore more options without the use of a solver when designing different components and parts, saving time and resources. This ultimately produces higher quality results that can then be used to make more informed decisions throughout the design process.
Introducing Amazon SageMaker Reinforcement Learning Components for open-source Kubeflow pipelines
Woodside Energy uses AWS RoboMaker with Amazon SageMaker Kubeflow operators to train, tune, and deploy reinforcement learning agents to their robots to perform manipulation tasks that are repetitive or dangerous.
Leveraging AI and Statistical Methods to Improve Flame Spray Pyrolysis
Flame spray pyrolysis has long been used to make small particles that can be used as paint pigments. Now, researchers at Argonne National Laboratory are refining the process to make smaller, nano-sized particles of various materials that can make nano-powders for low-cobalt battery cathodes, solid state electrolytes and platinum/titanium dioxide catalysts for turning biomass into fuel.
Way beyond AlphaZero: Berkeley and Google work shows robotics may be the deepest machine learning of all
With no well-specified rewards and state transitions that take place in a myriad of ways, training a robot via reinforcement learning represents perhaps the most complex arena for machine learning.
AWS Announces General Availability of Amazon Lookout for Vision
AWS announced the general availability of Amazon Lookout for Vision, a new service that analyzes images using computer vision and sophisticated machine learning capabilities to spot product or process defects and anomalies in manufactured products. By employing a machine learning technique called “few-shot learning,” Amazon Lookout for Vision is able to train a model for a customer using as few as 30 baseline images. Customers can get started quickly using Amazon Lookout for Vision to detect manufacturing and production defects (e.g. cracks, dents, incorrect color, irregular shape, etc.) in their products and prevent those costly errors from progressing down the operational line and from ever reaching customers. Together with Amazon Lookout for Equipment, Amazon Monitron, and AWS Panorama, Amazon Lookout for Vision provides industrial and manufacturing customers with the most comprehensive suite of cloud-to-edge industrial machine learning services available. With Amazon Lookout for Vision, there is no up-front commitment or minimum fee, and customers pay by the hour for their actual usage to train the model and detect anomalies or defects using the service.
Rearranging the Visual World
Transporter Nets use a novel approach to 3D spatial understanding that avoids reliance on object-centric representations, making them general for vision-based manipulation but far more sample efficient than benchmarked end-to-end alternatives. As a consequence, they are fast and practical to train on real robots. We are also releasing an accompanying open-source implementation of Transporter Nets together with Ravens, our new simulated benchmark suite of ten vision-based manipulation tasks.
Artificial Intelligence: Driving Digital Innovation and Industry 4.0
Intelligent AI solutions can analyze high volumes of data generated by a factory to identify trends and patterns which can then be used to make manufacturing processes more efficient and reduce their energy consumption. Employing Digital Twin-enabled representations of a product and the associated process, AI is able to recognize whether the workpiece being manufactured meets quality requirements. This is how plants are constantly adapting to new circumstances and undergoing optimization with no need for operator input. New technologies are emerging in this application area, such as Reinforcement Learning – a topic that has not been deployed on a broad scale up to now. It can be used to automatically ascertain correlations between production parameters, product quality and process performance by learning through ‘trial-and-error’ – and thereby dynamically tuning the parameter values to optimize the overall process.
Edge-Inference Architectures Proliferate
What makes one AI system better than another depends on a lot of different factors, including some that aren’t entirely clear.
The new offerings exhibit a wide range of structure, technology, and optimization goals. All must be gentle on power, but some target wired devices while others target battery-powered devices, giving different power/performance targets. While no single architecture is expected to solve every problem, the industry is in a phase of proliferation, not consolidation. It will be a while before the dust settles on the preferred architectures.
Pushing The Frontiers Of Manufacturing AI At Seagate
Big data, analytics and AI are widely used in industries like financial services and e-commerce, but are less likely to be found in manufacturing companies. With some exceptions like predictive maintenance, few manufacturing firms have marshaled the amounts of data and analytical talent to aggressively apply analytics and AI to key processes.
Seagate Technology, an over $10B manufacturer of data storage and management solutions, is a prominent counter-example to this trend. It has massive amounts of sensor data in its factories and has been using it extensively over the last five years to ensure and improve the quality and efficiency of its manufacturing processes.
Building effective IoT applications with tinyML and automated machine learning
The convergence of IoT devices and ML algorithms enables a wide range of smart applications and enhanced user experiences, which are made possible by low-power, low-latency, and lightweight machine learning inference, i.e., tinyML.
Advanced Technologies Adoption and Use by U.S. Firms: Evidence from the Annual Business Survey
While robots are usually singled out as a key technology in studies of automation, the overall diffusion of robotics use and testing is very low across firms in the U.S. The use rate is only 1.3% and the testing rate is 0.3%. These levels correspond relatively closely with patterns found in the robotics expenditure question in the 2018 ASM. Robots are primarily concentrated in large, manufacturing firms. The distribution of robots among firms is highly skewed, and the skewness in favor of larger firms can have a disproportionate effect on the economy that is otherwise not obvious from the relatively low overall diffusion rate of robots. The least-used technologies are RFID (1.1%), Augmented Reality (0.8%), and Automated Vehicles (0.8%). Looking at the pairwise adoption of these technologies in Table 14, we find that use of Machine Learning and Machine Vision are most coincident. We find that use of Automated Guided Vehicles is closely associated with use of Augmented Reality, RFID, and Machine Vision.
How Instacart fixed its A.I. and keeps up with the coronavirus pandemic
Like many companies, online grocery delivery service Instacart has spent the past few months overhauling its machine-learning models because the coronavirus pandemic has drastically changed how customers behave.
Starting in mid-March, Instacart’s all-important technology for predicting whether certain products would be available at specific stores became increasingly inaccurate. The accuracy of a metric used to evaluate how many items are found at a store dropped to 61% from 93%, tipping off the Instacart engineers that they needed to re-train their machine learning model that predicts an item’s availability at a store. After all, customers could get annoyed being told one thing—the item that they wanted was available—when in fact it wasn’t, resulting in products never being delivered. ‘A shock to the system’ is how Instacart’s machine learning director Sharath Rao described the problem to Fortune.