Early Bird Special $250 – eSMART Factory Conference 2018
We are extremely pleased to announce the official Kickoff of the eSmart Factory Conference initiative, in partnership with Lab One. It all starts at 6:00 p.m. on Thursday, May 23 at 3055 Orchard Drive, Suite 10, San Jose CA and concludes with a reception, compliments of Lab One. This is a great opportunity to stimulate new business relationships, rekindle old ones, and get you in the mood for the “hot key technologies shaping the electronics manufacturing industry”.
Lab One is located in the heart of the electronics industry, and is setup such that visitors can walk in and discuss and collaborate on projects, use the equipment, create samples and even small series production runs. They provide inspection services to stimulate thoughts around new solutions and innovations, solving visitor’s problems together and then also bringing the new technologies to their visitors. Lab ONE provides all the inspection facilities and services of the Comet Group under one roof with experts from their RF Power (Semiconductor), YXLON (X-ray) and Ebeam Treatment System divisions. All technologies offered by the Comet Group – from ebeam to RF power and X-ray – can be tested and demonstrated there. The trends for miniaturization, safety, security and sustainability bring us all together to share technologies and make them sustainable and resource efficient. These are the common denominators. There will be demonstrations throughout the event.
After registering Online, you will receive a PayPal invoice confirming your payment to eSmart Factory Conference 2018. Please bring the PayPal invoice and a Photo ID with you to the conference and present it at the time of registration. You will receive your name badge and other supporting information!
On arrival at Plug ‘n Play Center in Sunnyvale, CA, please follow the posted signs to eSmart Factory’s Registration Desk on the second floor.
Abstract: Among the electronics firms currently employing AI and IoT, a group of Reinventors is leveraging these technologies in a particularly transformative and visionary way. Understanding the commonalities between these firms is crucial in this era of fast-paced change for electronics, as industry boundaries blur, and rich ecosystems develop around newly available data sources. A fundamental step towards success in this age of Digital Reinvention is the development of an appropriate hybrid cloud strategy, to support advanced applications of AI and IoT.
Christophe Begue bio: Christophe Begue has twenty years of experience in business transformation, solution selling and the development of vertical solutions for the Electronics, High-Tech and Manufacturing Industries. He developed the IBM Supply Chain Management practice, winning some of the largest global supply chain solution implementations in Electronics, and is a member of IBM Industry Academy.
Laura Ong bio: Laura Ong is a recent Princeton graduate, working as a Business Transformation Consultant aligned to Distribution and Cognitive Business Decision Support.
Introduction: In manufacturing, we work systematically at ways to save time while maintaining the highest yields. During the last 50 years this was evident with the methodologies born which would help streamline operations and indoctrinate staff. These have helped form what is today the best practices which are followed in every modern manufacturing company globally. These processes and systems can be seen in the assembly line, TQM (Total Quality Management), Six Sigma and LEAN. As a growing global electronic manufacturing powerhouse, with an eye on the future, we considered more about where we needed to be in two years and consider ways to leap frog our nearest competitors of future ones. It meant should we do what others are doing and make the incremental improvements and follow a classic path. Or should we look further ahead and consider what manufacturing of the future would work and transact like. We looked inward at the business and painstakingly reviewed all our activities down to each task to discover what could be done. Looking at many ways to decrease waste while improving the overall functions of the departments during our planned growth. We focused on two major areas, Production and the Indirect Functions which support the entire business.
Challenge: Transcend and disrupt inside our business space. We knew that today and future products which our clients demanded, are every more design savvy with intricate and custom fabrication methods and materials. From diagnostics equipment through to the most advanced automotive systems we could see what was needed to out pace while providing no decrease in quality. Then we must consider the challenges our clients face forecasting, dealing with supply chain concerns and the distribution of key materials that our clients products require. We needed a manpower automation solution which could help in Indirect Labor and lend some hand in Production. We wanted to automate the Indirect Labor functions over a period with the end goal of maturing into a 50/50 machine to human task force. This would allow us to maintain a “mentor” role over the systems while have an unlimited workforce on call.
Process: Discovery then Plan, Define, Design, Deploy and Improve. During this planning period, we meet with several technical consultancies which could help with incremental change and perhaps provide some opportunities. All had a common approach and opportunity but may not make the impact necessary to transcend and provide us with the resource improvements for our planned expansion. We met with a startup in Silicon Valley, Neural Corp, which was founded based on early use of machine learning through technical university’s in Europe and Asia. Neural was solving problems which reduce “Networks” where Human-to-human transactions caused delays or time constraint leading to costly processes. With their depth in business consulting, machine learning and artificial intelligence it became the most opportune relationship. AsteelFlash will become the first manufacturing firm globally to have a hybrid-staff with a growing ratio of machine to human counterparts. We targeted the entire portion of the business where a cluster of sensitive networks caused the most challenge for manufacturing and ones which would also provide more direct value to our customers programs. These networks are common to all manufacturing and notable when you understand the processes which they oversee. They also have the most severe causality to programs when not in focus. We would classify these as major nervous systems which must be working in concert with the highest communication possible to truly support the programs which they support. We undertook an extensive planning and testing to pilot phase where we had to find the best fit into the live systems, splicing into workflows without disruption to any program whilst making an elegant presentation that Artificial Intelligence does transform. This utilized over 20 years consulting and dealing with change from factory floor to life sciences firms. And in a place where change is welcome but difficult to execute. Results (Status) Today we are in the early stages of brining online portions of the intelligence and weaving them into the daily routines of current team members. Using Neurals proprietary methodologies we targeted chunks of these identified networks and follow a system deployment processes to maintain quality adherence. To help with transformation we utilize an advanced onboarding concept where we operate as a startup within our own company consuming the processes internally via Smith (SM) the named Artificial Intelligence so to maintain control during the Neural Corps, Apprentice to Mentor model. Using these methods and strategies it will allow us to have the lowest change causality while having the most immediate impact from each step. Roles which have undergone the initial changes are from Document Control to Sourcing, Buying and Planning. They lead us into other supportive intelligence which would then directly help Program Management, Forecasting and our clients own Programs and Product teams.
Summary: We set out to increase our workforce, save time on human processes, decrease human to human networks which would allow us endless capacity and increased communication speed which were previously staffed by human based roles. Smith can provide value in directly realized time equivalent of hundreds of man hours. As it is trained it evolves and will time take on additional duties and solve more problems that imagined.
The whole manufacturing world is going digital, though with many parallel courses in manufacturing, product engineering and supply-chain. The Digital “whatever” series of messaging from different companies provides inspiration, but also confusion as to what actual digital entities are needed, who provides them, and how they interact. As digital data standards emerge, we need to put some perspective on what true digital manufacturing of the future, including supply-chain and quality as well as manufacturing itself, will need. How vendors in the industry cope with and gain benefit from working alongside each other.
Taking the IPC standards of 2581 (Digital Product Model), 1782 level 4 (Digital Traceability), and the latest Connected Factory Exchange (CFX) shop-floor Industrial IoT standard as examples, we take a fresh look at the roles and responsibilities of machine and robot vendors, human operators, MES, ERP and PLM solution providers to see how and where these technologies plug and play together to achieve manufacturing business goals. Connecting business and technology takes IIoT from an academic discussion into a real-world revolution. This is how the complete digital manufacturing factory is likely and practicably to be seen in just a short time to come.
Coffee and tea will be made available within the main conference area. You are invited to use the high top tables to network with your peers and take time to visit the table top displays around the room.
“Danger Will Robinson…Danger”, Classic line from the 1960s TV series “Lost in Space” depicts how far back artificial intelligence and robots have been part of our lives. Artificial intelligence (AI) makes it possible for “machines” to learn from experience, adjust to new inputs and perform human-like tasks. While Hollywood movies and science fiction novels depict artificial intelligence as human-like robots, AI can encompass anything from Google’s search algorithms to IBM’s Watson to autonomous weapons.
The current evolution of AI technologies believe AI has evolved to provide many specific benefits in every industry, from health care to media and entertainment, legal, retail and even manufacturing. According to Wikipedia, the field of AI research was founded at a workshop held on the campus of Dartmouth College during the summer of 1956. The Dartmouth Conference of 1956 was organized by Marvin Minsky, John McCarthy and two senior scientists: Claude Shannon and Nathan Rochester of IBM. The proposal for the conference included this assertion: “every aspect of learning or any other feature of intelligence can be so precisely described that a machine can be made to simulate it”.
At the conference McCarthy persuaded the attendees to accept “Artificial Intelligence” as the name of the field. In 1950 Alan Turing published a landmark paper in which he speculated about the possibility of creating machines that think. He theorized that if a machine could carry on a conversation that was indistinguishable from a conversation with a human being, then it was reasonable to say that the machine was “thinking”. The Turing Test was the first serious proposal in the philosophy of artificial intelligence. Fast forward to present day, and there are notable personalities that have weighed in on the risks of artificial intelligence. Stephen Hawking, Elon Musk, Steve Wozniak, Bill Gates, and many other big names in science and technology have recently expressed concern in the media and via open letters about the risks posed by Artificial Intelligence, joined by many leading AI researchers. Artificial intelligence is a combination of technologies that have their roots in learning algorithms.
Early work in Neural Network Algorithms from the 1950’s through the 1970’s stirred excitement as data predictions with some level of accuracy was starting to be realized. As data sources grew, and data point grew even faster, we ushered in Machine Learning algorithms from the 1980’s to 2010s. Finally, as data availability became more a part of a conscience strategy and readily available by the petabytes, computing power was more accessible, and data storage limitations dramatically removed, we arrive at the Deep Learning stage of AI’s evolution to present day. With it comes amazing opportunities for human advancements in health, technology and the human experience.
For example, your interactions with Alexa, Google Search and Google Photos are all based on deep learning and artificial intelligence. In the medical field, AI techniques from deep learning, image classification and object recognition can now be used to find cancer on MRIs with the same accuracy as highly trained radiologists, but faster and without fatigue. Artificial Intelligence is also big business. PwC estimates that artificial intelligence could add $15.7 trillion to global GDP by 2030. According to CB Insights, China accounts for a mere 9% share of deals going to AI startups globally. But China’s AI startups took 48% of all dollars going to AI startups globally in 2017, surpassing the United States for the first time for share of dollars. China’s Bytedance in the news and media industry, leads funding with $3.1 billion. Of the top 100 startups tracked by CB Insights’ Mosaic algorithm, 76 are U.S. based. As with any technology advancement acceleration, funding has always been key, and AI is no exception.
Do we temper progress with fear or do we drive ahead and forge new technological solutions for challenges in various fields, including manufacturing? This presentation will go through an in-depth review of Artificial Intelligence, a brief history, its related ethical dilemma, how it affects our daily lives, the business and monetary impact of AI and finally the current, near-term, and potential future impact to us and on the manufacturing space.
This paper will present the key elements that must be considered when creating a roadmap for the Smart Factory. It will explain the need for a clear assessment of the current situation for each factory, including the legacy systems in place, and the specific challenges and opportunities for improvement.
The paper will also present the complete portfolio of existing manufacturing software systems and the key functions associated with each system. In addition the paper will present emerging technologies, such as digitial twins, artificial intelligence, deep learning, big data, and how these systems will be positioned relative to traditional PLM, MES and ERP solutions.
Finally the paper will present practical considerations for creating a practical step-by-step roadmap for the Smart Factory.
To survive and prosper in today’s economy, leading manufacturers must assemble high-quality products at the lowest possible cost. The total cost of production must take into account the complete product lifecycle including warranty, recalls and repairs. Monitor, track, trace, and control systems are an essential element of success in achieving these strategic objectives. Data collection combined with analytics provide real-time process outputs and speed up decision making. Generating useful insights from the data allow process operators to act on those ideas. This real-time information helps them to develop the process window, optimize process settings and reduce downtime. The intelligence helps the process operators to learn, reduce variability and to form a repeatable and reproducible product.
The purpose of this research paper is to present monitoring, tracking and data analytics targeted at the reflow and cleaning process. Sensors strategically placed within the reflow oven and cleaning process capture the conditions that each assembly sees during the cleaning process. Analytics programmed within the system software provides process operators with real-time data outputs for monitoring the operation.
The inputs equate to knowledge. This knowledge sends corrective factors to clean production assemblies within the process window. The integrated system enables the process engineer with the capacity to learn and improve. Real-time diagnostics creates an intelligent system that works interactively.
Lunch will be served at the self-service cafeteria adjacent to the main conference room. Feel free to use the seating in the dining area and take time to visit the table tops around the perimeter of the main conference room.
The digital transformation with concepts like Industry 4.0, Smart Factory and the Internet of Things will change production, work processes and business models in electronics manufacturing radically and forever. It is all about connectivity, gathering and exchange of big data. Smart manufacturing should be a proactive and independent decision-making environment where connected systems or machines will “talk” to each other. That “discussion” is, however, based on algorithms.
By definition an algorithm can give us only one single answer. But does this provide us the right answer or one from three or four or five alternatives? Will it provide the same response every time or different ones for the same issue? Many MIS systems are not ready to manage big data, due to the lack of smart analytic tools. Without human intervention to make those critical decisions, based on knowledge and experience, we still cannot talk about completely effective and efficient smart production. Therefore the reliable solution combines human brains and algorithms together, allowing Process Engineers to make informed decisions after seeing all the data from various collection points. It is obvious that the quality of the raw data is one of the core components of the Smart Factory. In the connected Smart Factory, all machines and systems become essentially smart sensors, collecting all possible data from the production line and the boards themselves.
Therefore, the accuracy and sensitivity of those sensors is very critical, especially in quality control systems like AOI and AXI. Without diminishing the strong characteristics of the in-line inspection equipment, these still need assistive technologies for the verification of faults. The connected at-line x-ray would be that additional sensor making the final Pass/Fail decision. This paper will investigate how the connected at-line x-ray system can improve the production yield and lower production costs of the manufacturing process when connected to the production line via in-line AXI (or 3DAOI). This paper will demonstrate the logic and benefits of SmartLink technology where at-line x-ray can compensate the shortcomings of in-line inspection solutions.
V-ONE, a smart factoring tool at your fingertips, visualize, respond and proactively predict the inspection results in various stages of the production lines making your process indicators live both pre and post inspections and giving the capability to control the production process in a smarter way with less time and resources.
Electronics Manufacturing around the world suffers from defects on the products that slips through inspection stations. This project involves in a software solution that serves in the concept of Industry 4.0. The purpose of this software solution strive to monitor the inspection stations, giving real time information and process defects in order for the engineers to predict the product defects in high accuracy when it happens and soon even before it happens.
This project draws upon mostly primary sources from the Machine Vision aspect as well as in the technology of deep learning artificial intelligence and big data analytics. The prior focus and objective of this project is to connect all the inspection systems and other equipment in the SMT production line. Visualize real time information and process defects from them and ProActively inform potential root causes of the defects that happens, where it happen and how it happens so that prevention steps can be implemented. Therefore it enables factories to be managed in a smarter way and optimizes factory resources across geographical locations. The enormous data collection can be stored in a centralized server or cloud platform. This project has highlighted the importance of the revolution in the manufacturing industry and beyond. The benefits; it has allowed reduction of time to market, reduction of humans managing machines, higher yield and utilization of investment on capital equipment.
Amazon currently employs over 45,000 robots across 20 fulfillment centers around the world. That’s a 50% increase in a single year. As the lights out concept rapidly grows with the giant retailer, automation adoption still finds resistance in the SMT manufacturing world. Despite the great progress in the development of the integrated production line with Industry 4.0, a critical part of the manufacturing chain still offline. Inventory rooms around the world must still operate with a far reduced level of sophistication. It is not uncommon to find thousands of reels densely packed in hundreds of linear feet of shelving space.
The simple exercise of counting back reels as they come back from the production floor is not standard practice. And even when they are counted, the count is often made by hand. The lack of focus on the automation of our inventory rooms leads to an increasing loss of productivity due to line stoppages. When associated with the recent increase in lead times for electronic components, line stoppage has become a crippling problem that makes inventory accuracy a necessity. In this presentation, we will discuss the state of the art in material management automation and the barriers of entry to a fully automated inventory room.
We will also explore the infrastructure needed to be in place to realize an accurate, fast, and efficient methodology for a fully automated inventory control. This investment is critical to realize the benefits in efficiency unleashed by Industry 4.0.
Coffee and tea will be made available within the main conference area. You are invited to use the high top tables to network with your peers and take time to visit the table top displays around the room.
Agile manufacturing requires quick decision making. This means product and process performance needs to be monitored and provide enough data for trend, root cause analysis, and prediction. Many times this data resides in disparate locations (i.e. databases, machines, etc.) and behind proprietary interfaces, making data integration a daunting task. This requires users to go to individual machines to pull data and extra effort to establish a meaningful data integration and actionable information from it. Moreover, if the data requires assimilation with other production data, engineers often have to run queries that need to be associated and merged before it can be analyzed. This is a task that makes data analysis costly and slow, while making effective real-time process control nearly impossible and expensive.
In our vision of smart manufacturing framework, we plan to leverage the best technology blends to achieve Digital Transformation on the plant floor. We envision this transformation as a leap step into an ecosystem of tools and information. In this work, we present initial steps in the direction of automated data acquisition, harmonization, storage and processing that will use the best of the Edge and Cloud computing. Our endeavor is to combine these initial two-level computing settings to create actionable analytics and visualization around our electronic assembly processes. This solution will allow the machine operators and production engineers to identify potential outliers in the production processes and to unveil product quality tendencies that can be resolved faster and with anticipation.
Looking into the future; machine learning and other artificial intelligence tools will help us identify out of controls, trends and new correlations. This will reduce downtime and defects that will drive operations productivity.
Structural Electronics is emerging as a new way of adding intelligence into products such as automobiles, turbine engines, medical devices, and building infrastructure. These functions can be used to sense, report on, and take or recommend corrective actions when products are deviating from their specifications, all part of the (Industrial) Internet of Things.
However, these front-end monitoring solutions have been underserved as the focus on the IoT has been on data analytics, security, and storage. Optomec has developed a scalable additive manufacturing solution that facilitates the incorporation of sensors and antennas directly on to high value products used in industrial settings. Optomec’s Aerosol Jet printers deliver the unique ability to print electronic and other materials onto any type of 2D and 3D structure, in dimensions ranging from 10 microns up to centimeters in scale.
Aerosol Jet 3D printed electronics solutions have been deployed to add intelligence to products enabling continuous monitoring and feedback. This added functionality facilitates just-in-time corrective action helping to insure steady state operation of the functionalized products. Real world examples will be presented including printed creep sensors on turbine blades to monitor metal fatigue.