Breaking the Bottleneck | Issue 20
[8/7/2023] Nvidia Metropolis, Retail's Manufacturing Shift & a Tesla Solar Roof Review
Breaking the Bottleneck is a weekly manufacturing technology newsletter with perspectives, interviews, news, funding announcements, and a startup database. For a high-level market map on discrete and continuous manufacturing click the link here! If you know anyone looking to chat about manufacturing tech, I’d love to talk!
Content I Enjoyed This Week 🏭🗞️🔬
News:
Digital Twins as the Key to Product Development [McKinsey]
In advanced industries, around 75% of companies have already adopted digital-twin technologies, with players in automotive and aerospace being more advanced in their use. Digital twins provide a risk-free product development environment, improve testing and validation, offer deeper insights into product behavior, and allow real-world data to inform product improvements. Companies that have implemented digital twins report significant improvements in development times, cost reduction, fewer quality issues, and higher customer satisfaction. The global market for digital-twin technologies is projected to grow rapidly, reaching $73.5 billion by 2027.
Nvidia Metropolis Using Cutting-Edge GPU Technology [Engineering]
The article highlights the NVIDIA Metropolis for Factories platform, which enables manufacturers to reduce product defects as well as improve preventive maintenance, remote asset monitoring, safety, and compliance. The platform utilizes advanced vision AI capabilities and GPU-accelerated SDKs to build, deploy, and scale AI-enabled video analytics applications from factory edge deployments to the cloud. Metropolis provides various AI workflows for developing automated optical inspection, allowing manufacturers to create accurate AI solutions faster. It offers pre-trained models and transfer learning options to speed up AI model development without the need for massive training datasets. The platform also includes NVIDIA's TAO Toolkit for quick training, adapting, and optimizing models with custom organization data. The article mentions that NVIDIA hardware, particularly NVIDIA GPUs, accelerates the Metropolis run-time stack to maximize object detection, tracking, classification, and anomaly detection throughput.
AI Build Simplifying Large Format 3D Printing with AI [3D Printing Industry]
AiSync uses AI models, including OpenAI's GPT-4, for natural language processing to create advanced 3D printing toolpaths using simple text prompts. The software also provides machine-learning-driven quality control, improving first-time print success rates and automating anomaly detection during the printing process. In the future, Ai Build envisions AI as a strong companion to engineers and designers, making communication with machines seamless and enabling a higher level of automation in 3D printing. The company's ultimate goal is to simplify the additive manufacturing process, allowing for high-quality, fast, and accurate production without the need for extensive technical competence and constant iteration.
Colgate Palmolive Scaling Manufacturing Operations [Industry Week]
Warren Pruitt discussed Colgate-Palmolive's successful approach to scaling manufacturing technologies. The company established a community of practice comprising engineers and managers from plants around the world to meet regularly to identify new manufacturing technologies, pilot them in select plants, and then scale the successful ones across the organization. Furthermore, Pruitt emphasizes the importance of forming external partnerships with technology providers who are willing to collaborate rather than just seeking customers. Examples of successful technology deployments include the Augury predictive maintenance system and a digital twin system developed in partnership with Twin Thread.
Mercedes-Benz Investing $2.2B+ to Reskill Workforce [Manufacturing Dive]
Mercedes-Benz will invest more than $2.2 billion by 2030 to train employees as data and artificial intelligence specialists, the company said last week. The Turn2Learn initiative’s two pilot programs are training over 600 employees in production, production-related and administrative roles in digitalization and AI. The automaker will award certificates to the initiative’s first graduates this week.
Inside Walmart’s Warehouse of the Future [WSJ]
Walmart is implementing automation at its Brooksville warehouse in Florida, transitioning from traditional manual labor to roles managed by robotic arms, conveyor belts, and screens to improve load accuracy and speed up the supply chain. Not all workers are eager about the transition. However, the company provides workers with around six weeks of training to learn how to work with the automated systems, troubleshoot issues, and track inventory as it moves between robots. Some have expressed skepticism and fear about dealing with new technology but workers who have transitioned to automated roles have reported positive aspects, such as experiencing less physical strain and more mental stimulation during their shifts.
Research:
Jabil 3D Printing Survey Results [Jabil]
A majority of the participants (97%) currently are using 3D printing to produce functional or end-use parts. Nearly three-quarters of participants produced at least 10,000 3D-printed parts over the past year, and more than a third printed up to 100,000 parts. Overall, participants anticipate an uptick in the use of 3D printing for production parts or goods in the next 3-5 years, despite being less bullish about overall 3D printing industry growth than in previous surveys.
Other key findings included:
The top three use cases for 3D printing are prototyping (97%), research and development (75%), and production parts (59%).
3D printing use for bridge production (moving from prototyping to initial production) grew from 23% in 2017 to 59% in 2023, while 3D printing jigs, fixtures, and tooling nearly doubled from 2017 (30%) to 2023 (58%).
Prototyping is recognized for delivering the most significant impact to product lifecycles by 95% of the participants, followed by product designs (52%) and small-scale production (27%).
IDC Manufacturing Tech Survey Results
Podcasts/Video:
Intel’s Transition to Net Zero Emissions [SupplyChainBrain]
Chinese EV Maker Selling More Than Tesla [Odd Lots]
Chart of the Week:
About 61% of apparel retail CEOs have stopped using China as their primary supplier, up from 30% before the pandemic, according to a new survey by the US Fashion Industry Association (USFIA Survey).
Manufacturing Deals
Transcend - A company creating generative design software for sustainable infrastructure
$20 million [Series B] - From Autodesk, HG Ventures, PureTerra, Arosa Capital, and Riverstone Holdings
Chemify - A company operating its proprietary molecular design, discovery, and chemical manufacturing technology
$43 million [Series A] - Led by Triatomic Capital and joined by Horizon Ventures, Rocketship Ventures, Possible Ventures, & Alix Ventures
Weekly Planned Downtime
Tesla Solar Roof Review