Artificial Intelligence

The adage “the only thing constant is change” has never been more universally applicable than this past year – across the globe, across industries, across buyers. All manner of ways in which we work and consume has changed and continues to change, driving innovation, disrupting industries, and transforming buyers’ behavior. To survive, companies must follow the old adage: to remain a constant, they must change. Overnight, we shifted to work-from-home, and, after a few days to adjust and align, we discovered surprising benefits. By working remotely, we gained time by losing our commute, and we increased exponentially the number of meetings we could hold – and the number of people we could meet with – in a typical business day. Executives, customers, and decision makers were suddenly more accessible, and we could share a ‘face-to-face’ call in far more intimate settings, allowing us to meet family and pets, which in turn deepened relationships. Beyond productivity and a healthier work-life balance, remote work obliterated any constraints of geography, enabling companies to consider employees across the country and globe, thereby expanding talent pools, creating retention opportunities, and bolstering diversity efforts. Now, despite the easing of restrictions, published studies and employee surveys (even our own annual Tell Dell survey) show that many employees want and expect to continue to work remotely at least part-time. No surprise there, but it is important to note: These changes in preference and expectation are not limited to how we work; They apply to every aspect of our lives. In 2020, with never-before-seen speed, we adopted distance learning, telehealth, online entertainment, 3D printing of PPE, online grocery/restaurant orders, and digitally-enabled deliveries and curbside pickup – and we aren’t going back. Just like employees now prefer the flexibility of work-from-home, buyers now prefer – and expect – the flexibility of shop-from-home. While these changes were in progress well before 2020, the pandemic accelerated and normalized adoption, and now buyers approach business decisions with the same preferences, expectations, and behaviors of consumers. In fact, according to Gartner, by 2025, 80% of B2B sales interactions between suppliers and buyers will occur in digital channels.[1] Buyers have already embraced online research and digital buying. They expect authentic, personal experiences and relationship-driven online interactions. Like the consumers they are at home, B2B buyers are researching online well before they engage with a person. To survive, companies must meet customers where they are and how they want to buy: online. For marketing and sales, the handoffs have changed. In marketing, our messaging and content is touching decision makers and potential customers far before they meet with a sales rep. Enabled by artificial intelligence (AI) and enhanced analytics, sales teams will need to follow the data, and be ready to respond to buyers’ needs at the exact time they realize the need. At Dell Technologies, we have not only embraced this digital transformation change, but we are also leveraging marketing automation technology to help our partners learn and activate digital marketing and selling. We are training our sales and marketing teams while also providing enablement, training and support to enable our partners to navigate the new buyer’s experience. Our teams are organized to move quickly and lead through change so, together with our partners, we can address the ever-changing needs of our customers. Are you and your team ready for this change? Do you have the digital skills needed to adapt? Are your organizations agile and open to new ways of working? Do you have the right leaders in place to lead through change? Your buyers are in the driver’s seat: They determine if, when, and how they interact with suppliers. Are you in the right place at the right time to meet your customer if, when and how they want? To remain a constant – to remain in business – you need to embrace the change in your buyer and embrace the technology available to meet your buyers where they are – online. Join me July 13 at my session Digital Leadership – Embracing the Buyer Evolution at the SEMI Innovation for a Transforming World virtual event to learn more. Senior Vice President at Dell Technologies, Cheryl Cook spearheads development and strategy for the Global Partner Marketing organization. Beyond her main global responsibilities for branding, partner program marketing, channel events, partner communications, and MDF/BDF program investments and execution, Cheryl drives long-term partner marketing strategy, together with Dell’s Global Alliances, OEM, and global and regional business teams. A vocal advocate for the partner community, Cheryl is a 20+ year partner veteran, known as an innovative, collaborative leader who creates compelling business solutions that accelerate partners’ success. [1] Gartner Press Release, Gartner Says 80% of B2B Sales Interactions Between Suppliers and Buyers Will Occur in Digital Channels by 2025, September 15 2020.

Dr. Mousumi Bhat, senior director of External Manufacturing at Micron Technology and first female member of the SEMI Southeast Asia Regional Advisory Board, discusses her passion for sustainable business practices and her work driving the transformation of Micron’s manufacturing plant in Singapore.

Traditionally, defect classification is done manually by operators or using Automated Optical Inspection (AOI) machines, often leading to classification inconsistencies. Also, rules-based AOIs may at times be unable to fully satisfy project requirements due to the rigidity of inspection recipes. SixSense – Breaking the Status Quo with Artificial Intelligence Enter SixSense, an AI-powered defect classification software platform that has been making breakthroughs in defect detection and classification for semiconductors to make manufacturing smarter and more efficient. Founded in 2018, SixSense has already amassed a wealth of experience and chalked up a number of successes such as automating the manual image classification process, reducing manufacturing false rejects, and capturing escapees. Infineon Technologies and GlobalFoundries were amongst the early adopters of SixSense’s platform: classifAI. With Infineon, classifAI has allowed over-rejection rates to be precisely quantified. classifAI – Simple UI, Easy Usage, Powerful Models As a UI-based assistive software platform, classifAI, SixSense’s automated defect classification platform is built with the defect and yield engineer in mind. SixSense takes care of all the back-end complexities – such as coding, algorithm modelling and deployment – to enable end users to get started and use the platform with a simple GUI. The simplified end-to-end AI pipeline offered on the platform includes data labelling to make data AI-ready, model training, and model testing. Ultimately, models are deployed on the production floor for 24/7 inferencing of hundreds of millions of images every year, at scale, across processes, tools and sites. Machine learning models built by the SixSense team have seen strong results, with model accuracy of up to 98% in certain use cases. Track Record of delighting IDMs, Foundries and OSAT Customers SixSense has consistently solved visual inspection problems and enabled the success of IDMs, foundries and OSATs since its inception. The AI technology has helped a range of customers across 100mm-300mm wafer standards, both pure silicon and compound wafers, and caters to specific end-use market requirements such as RF and automotive. Partnerships between startups and established manufacturers are key to actualizing the value of AI in manufacturing. “Our collaboration with AI startup SixSense has enabled us to explore opportunities in yield gain, improving cycle time, and real-time monitoring of process shifts,” said Dato’ Tan Soo Hee, Executive Vice President, Global Backend Operations at Infineon Technologies Asia Pacific. “SixSense has been very attentive to the needs of our engineering team, addressing project requirements using a customer-first approach evident in the design of the intuitive software platform,” said Melvyn Peh, Principal Engineer, Automation-Scan-Pack, Infineon Technologies Asia Pacific. The intelligent annotation module is one of many offered by SixSense, which uses AI to train AI and accelerate the data annotation process by focusing on the semiconductor-specific requirements. Another valuable module in classifAI is advanced analytics that capture the heatmap for defect distribution on the images. Images are stacked on top of each other, with the location of defects aggregated to provide the defect heatmap. Through this, systematic failure patterns were identified that allowed defect engineers to zero in on key sources of failure and assist in root-cause analysis. Infrastructure – Scale Fast, Adapt Quickly, Accelerate Value Creation In the dynamic world of technology, machine learning and AI projects must meet changing infrastructure demands. A cloud-first approach is often favored for the plethora of benefits it offers. “We’re looking forward to a great partnership with SixSense, treading together hand in hand exploring fresh ideas and possibilities,” said Manju Jalali, Vice President of digital manufacturing at GlobalFoundries, who oversees the company-wide roll out of classifAI. For use cases where on-premise deployments are preferred, SixSense offers such options for infrastructure integration, satisfying all possible infrastructure requirements in the market. Contributing to a vibrant innovation ecosystem SixSense was mentioned by Singapore’s Deputy Prime Minister Heng Swee Keat during an event that marked Infineon’s 50th anniversary in Singapore: “I am heartened that Infineon will be investing more than $27 million over three years on an AI initiative in Singapore. Under this initiative, Infineon Singapore will be partnering academia, industry, and local startup SixSense AI to develop new AI solutions and courses.” Explosive Growth of AI in Chip Manufacturing According to a McKinsey Company report, AI contribution to semiconductor company earnings is projected to rise to between $85 billion and $95 billion per year in the coming years. SixSense has been taking great strides in creating value for their semiconductor customers. “SixSense offers tremendous value in a high-growth vertical in the semiconductor industry, marrying the latest deep learning algorithm with the compute power of the cloud,” said Rajan Rajgopal, CEO of DenseLight Semiconductor. “This leads to faster root-cause analysis that helps reduce the cost of non-conformance and improve quality.” Dominic Teo is Enterprise Business Development Representative at SixSense. He can be reached at [email protected].

Adnan Hamid, CEO, founder and visionary of Breker Verification Systems, an ESD Alliance member based in San Jose, Calif., once described his job in chip design verification at AMD as “breaking things.” When it came to naming his startup, Breaker was a natural choice. After some consideration, the “a” was dropped and the company became Breker. Now Hamid is breaking the most complex semiconductor designs and Breker, moving from a startup to a scale-up company, is a noted part of the functional verification space. Smith: Why does verification continue to take the most amount of time in a project cycle? Hamid: The project cycle for semiconductor design has changed. Design abstraction has been raised to a much higher level than the days when developers were connecting logic gates. Today’s developers are typing functions that don’t include lower-level implementation details. Designs incorporate more blocks of reusable IP. Both reduce design time. Meanwhile, designs are getting bigger with more blocks of IP stitched together, all in need of testing. As design complexity grows, the amount of testing and verification increases as a square of design effort. One block requires one functional verification effort. Four blocks of IP mean up to 16 functional interactions require verification. While design is moving up the abstraction level, that’s not the case for verification, where plenty of detail must be reimplemented. Verification has certainly evolved, but engineers still think at the level of independent stimulus, response and coverage, driving the need to allocate so much time for verification. Smith: Are chips targeting artificial intelligence and machine learning applications more difficult to verify? If so, why? Hamid: Yes, absolutely and it’s an interesting challenge, especially given that machine learning is based on massively connected processing element arrays. Attempting to verify the individual processing elements and the critical interconnects is complex. AI device arrays and, interestingly, verification test content operation may both be thought of as a mathematical graph of processing elements and interconnect. Their operation involves walking through the graph form to generate a result. Finding the optimum path through these arrays is key. To understand how these systems may be effectively verified, it is worth investigating planning algorithms. Originally proposed by IBM, these hold the key to this type of verification process. The AI- style algorithm starts backward at the end of the processing element array and tracks down the most optimal and likely paths through it. At Breker, we have used these planning algorithms extensively to drive our graph-based test content synthesis process. Smith: Does system integration require verification? Hamid: Yes, it does. In the past, most functional verification has been performed at the block level. However, with the increase in more specialized SoCs, functionality is spread across multiple blocks, as well as the software running on the processors, driving full system-on-chip (SoC) functional verification. In addition, new requirements such as security and safety must be validated. A system-level infrastructure such as cache coherency and power domain execution has become more complex and these must also be tested. The new frontier in verification is ensuring a fully operational SoC. Of course, given the size of these SoCs, hardware-assisted verification such as emulation is essential, and porting tests from block simulations to SoC emulations has become a requirement. This porting process is problematic and this in turn has driven portable tests, giving rise to the idea behind Accellera’s Portable Stimulus Standard (PSS), of which Breker was a major participant. Indeed, some companies are taking this to the next level by composing their system-level testbench at the same time as they commence SoC architectural design, and then developing the hardware design, software design and test content all in parallel, in the so-called “shift-left” manner. Smith: Is “shift-left” a growing trend that are you seeing in verification? Hamid: Yes. Shift-left is taking hold in hardware and software design, giving way to an increase in early test content composition. Then as individual blocks are finished and connected, their verification is driven from this same test content, saving a significant amount of time and effort. This is a huge verification and test generation change that was inevitable given the increased time-to-market constraints and SoC complexity. Figure 1: Shift-left is ushering in the next generation of SoC verification. Source: Breker Smith: As an entrepreneur, what advice would you give someone founding a startup or thinking about starting one? Hamid: Do not take the attitude “Build it and they will come.” My best advice for an entrepreneur or fledgling entrepreneur is to solve a specific customer problem, however narrow it might seem. Including services as part of a product offering and developing partnerships with other vendors helps with this and turns your company into a solution provider not a product developer. This is essential for getting the right products to market on time and within budget, and then ultimately scaling them across the market. The ESD Alliance and Accellera are hosting a two-part webcast series on the work-from-home experience titled Remote Work, Remote Chip Design: Building Chips During a Pandemic. The first panel, Wednesday, June 9, at 9:00am PDT, will feature a discussion led by Tom Fitzpatrick, strategic verification architect from Siemens EDA verification engineers through their experiences converting their home offices into verification test labs. The second panel in July will explore how executives managed a remote workforce and explain how they plan to bring employees back to physical offices. About Bob Smith Robert (Bob) Smith is executive director of the ESD Alliance, a SEMI Technology Community. He is responsible for the management and operations of the ESD Alliance, an international association of companies providing goods and services throughout the semiconductor design ecosystem.

On April 21, 2021, the European Commission put forward the long-awaited Proposal for a Regulation on a European approach for Artificial Intelligence, introducing for the first time harmonized rules for the development, placement and use of secure and ethical artificial intelligence (AI) in Europe. The proposed regulation’s wide scope subjects providers, importers, distributors, and users of AI systems to regulatory scrutiny. It also includes providers and users outside of the European Union (EU) that deploy AI systems or use AI system outputs in the EU. With extraterritorial scope, the proposed regulation aims to further strengthen the EU’s leadership in shaping global standards and norms for new technologies. A European Path to Safe AI Following a risk-based approach, the proposed regulation classifies AI systems according to the level of danger they pose in the following categories: Unacceptable risk: AI poses unacceptable risk to systems or applications with the potential to manipulate human behavior and exploit vulnerabilities of groups of people to cause psychological or physical harm. Examples of prohibited AI systems include social-scoring systems and biometric identification that can be used by public authorities. High risk: The proposal identifies two main categories: AI systems used as safety components of products (or are a product themselves), and other stand-alone AI systems that have fundamental rights implications. Considering their intended purpose, the proposal identifies specific conformity assessment measures for both groups. AI systems intended to be used as security components will require a conformity assessment by an independent third party. Such systems will also be subject to the same ex-ante and ex-post compliance and enforcement mechanisms as products of which they are part. In contrast, stand-alone AI systems assessed through internal checks would require ex-ante compliance with all requirements of the regulation as well as with robust standards for quality and risk management and post-market monitoring. The proposed regulation identifies eight areas of high risk including AI systems in safety components of products (e.g., machinery, radio equipment, AI applications in robot-assisted surgery), critical infrastructures (e.g., transport), educational and vocational training (e.g., exam scoring) and employment (e.g., monitoring or evaluation of persons in work-related contractual relationships). Prior to their introduction to market, high-risk AI systems will be a subject to strict requirements including the use of high-quality datasets; adequate risk assessment and mitigation systems; high levels of robustness, accuracy, and security; clear and adequate user information; detailed system documentation and logging of activities to ensure traceability of results. Human oversight and control must be ensured. Low and Minimal risk: For limited-risk AI systems (e.g., chatbots), the regulation proposes only minimum transparency obligations, while minimal risk AI systems posing little to no risk (e.g., AI in spam filters) will not be regulated. Boosting AI Excellence from Lab to Market Continuous innovation of AI requires a secure environment that can support responsible validation of AI technologies. To that end, the proposal encourages the set-up of testing and experimentation facilities or so called AI regulatory sandboxes. Established by one or more Member States, the sandboxes will provide a controlled environment to test innovative technologies under strict oversight before their market introduction. These facilities could play an instrumental role in connecting the Europe’s R D ecosystem, creating new partnerships among numerous stakeholders. In addition to regulatory sandboxes, the European Commission intents to set up: A Public-Private Partnership (PPP) on AI, data and robotics designed to implement and invest in strategic research innovation and a deployment agenda for Europe Additional Networks of AI Excellence Centers to foster exchange of knowledge and advance collaboration with the industry Testing and experimentation facilities to test state-of-the-art technology Digital Innovation Hubs, one-stop shops to provide access to technical expertise and experimentation An AI-on-demand platform as a central European toolbox of AI resources (e.g., expertise, algorithms, software frameworks and development tools). Next Steps The proposed regulation is the at the start of a lengthy legislative process and will be debated by the European Parliament and European Council in the coming months. Given the importance of AI, and number of stakeholders involved, it is likely the proposed regulation will face various changes before being applied across the EU. For its part, SEMI Europe will maintain discourse with key public and private stakeholders on the proposed regulation, closely monitoring related policy developments as they unfold. Marek Kysela is senior coordinator of Advocacy at SEMI Europe.

Over the past 50 years, the field of engineering simulation has developed numerical methods that enable engineers to solve 3D physics problems faster, easier, with greater accuracy and more robust results. Finite element analysis (FEA), finite volume methods (FVM) and finite different time domain (FDTD) have increased solver efficiency while dynamic visualization techniques improve what is often called user-friendliness. Despite these improvements, certain challenges still remain. Specifically, simulation requires the simultaneous trade-off of: Accuracy of results Speed of results Ease of use of the workflow Robustness of the workflow Take, for example, mesh generation, the building block of multiphysics solutions. It is well known that using coarser meshes increases simulation speed but will result in loss of accuracy. Similarly, easy-to-use workflows with simpler meshes also reduce accuracy, and can introduce other issues: The simulation may not converge and the robustness fails. Ansys is exploring the use of AI/ML to solve all of these problems. Simultaneous Improvements Commercialization of AI began in the 1970s, but the field actually got its start a decade earlier with the development of rules-based expert systems. The simplest form of AI, these systems rely on curated human expertise to solve problems that would normally require human intelligence. We’d expect that AI/ML applications would be actively used in science and medicine, from streamlining drug discovery and advancing robot-assisted surgery to automating medical records that can be instantaneously accessed by providers anywhere in the world. But AI/ML is rapidly being successfully adopted by an increasingly broad range of industries and users. It’s helping consumer brands mine their social media to find out customers’ feel about their products (sentiment analysis), giving investors a leg up on stock trade opportunities (financial algorithmic trading) and enabling e-commerce owners to personalize offerings to online shoppers (recommendation engines). At Ansys, we can use AI/ML methods to automatically find the parameters of simulation to simultaneously improve speed and accuracy. We believe applying AI/ML will enable us to: Further improve customer productivity Augment simulation, including accelerating chip thermal solutions and developing a fluids solver that combines high-fidelity solutions in local regions with ML methods in coarse regions Optimize design space exploration Drive business-intelligence decisions such as resource-prediction needs for our solvers Combine data analytics-based and simulation-based digital twins to create accurate and fast digital twin hybrids In other words, we believe that AI/ML will help us narrow the gap between the ideal world, where time, effort, efficiency and results are perfectly balanced, and what happens in real life – and make productivity, ease of use, and accuracy a little less of a trade-off. To learn more about applying AI/ML to autonomy, click here. Prith Banerjee is Chief Technical Officer at Ansys, Inc.

Introduction Automated production in electronics manufacturing can produce high-quality products, but it might lead to a particular failure without human interventions. With the rapid technology development, such as the Industrial Internet of Things (IIoT), big data analysis, cloud computing, artificial intelligence (AI), many manufacturing processes can be more intelligent, and Industry 4.0 can then be realized in the near future[1]. Smart manufacturing adopts real-time decision-making based on operational and inspectional data and integrates the entire manufacturing process as a unified framework. Then, the future manufacturing process transforms cyber-physical systems digitally and responds to any uncertain situations proactively while ensuring higher efficiency. In Surface Mount Assembly (SMA) lines, equipment status and quality data can be collected via IIoT technology. Data-driven solutions, such as AI and machine learning algorithms, can be applied to diagnose abnormal defects and adjust optimal machine parameters in response to unexpected changes/situations during production. Collaborating with various SMT industry partners, the research team at the State University of New York at Binghamton (aka Binghamton University) developed a novel framework based on AI-based closed-loop feedback control and parameter optimization to implement a smart manufacturing solution in the PCB assembly for yield and throughput improvement. This AI-based framework could provide a potential road map for data-driven process control in SMA. Machine Intelligence in SMA Each SMA process has a critical effect on the final PCB product quality and throughput. Notably, the solder printing process is a critical operation because over 60% of the PCB assembly soldering defects can be traced back to this stage. An inadequate volume of solder paste transferred to any PCB pad is a printing fault, which leads to board failure and substantial reworking and repair costs. The pick and place (P P) process is the highest cost procedure, including expensive machine investment and extended production time. In the soldering reflow process (SRP), the reflow oven temperature and other related settings determine the solder joints' quality and reliability. Hence, multiple inspection machines in the SMA processes have been introduced, including solder paste inspection (SPI) and automated optical inspection (AOI) machines. Particularly, two independent AOIs could be employed to detect the components' defects before and after SRP. Because many electronics components become small-scale (e.g., ), more assembly-related failures are often observed in recent SMA processes. The Smart Electronics Manufacturing Laboratory (SEML) at Binghamton University is fully equipped with two solder paste printers, two chip-mounters, and a reflow oven along with SPI and AOI machines. The research team tested more than 8,000 PCB at SEML. The results show that numerical methods based only on physical properties might have practical limitations in explaining small-scale components' behavioral patterns. It might be caused by unknown environmental factors (e.g., temperature and humidity), machine calibration, measurement accuracies, vibrations, etc., which could have influenced the quality of the SMA outcomes. However, recent research shows that AI-based methods can increase product quality up to 35%, reduce scrap rates, and optimize fab operations in semiconductor manufacturing, compared to traditional approaches[2]. It implies that a data-driven intelligent SMA process control has the potential to advance SMA processes. The goal of the smart SMA is to maintain optimized settings in both offline and online scenarios. The AI and data analytics solution can optimize all SMA process parameters before production (i.e., offline control) and during production (i.e., online control). The overall schematic of the AI-based closed-loop feedback control framework is illustrated in Figure 2. Intelligent SMA Modules In the solder printing process, four machine intelligence modules are considered: Printing advising module (PAM) Printing optimization module (POM) Printing diagnosis module (PDM) Dynamic stencil cleaning process control (CPC) Figure 2. A schematic diagram of the AI-based closed-loop feedback platform Figure 3. PAM effectiveness over customer’s best-known printing parameter setting PAM aims to recommend the ideal initial setting of the printer critical parameters, such as printing speed, printing pressure, and separation speed, using hybrid machine learning and heuristics optimization techniques[3]. As a case study, the research team validated the PAM's performance with an automotive PCB testbed and compared the printing results to the best-known printing parameters. The experimental results show that PAM can achieve over 50% higher Cpk (i.e., process capability index, as shown in Figure 3). POM optimizes printing parameters in real-time by monitoring printing quality and fine-tuning offset and process parameters for adapting to dynamic conditions[4]. The experimental results show that POM achieves more than 30% production quality improvement in terms of the Cpk by adjusting printing parameters compared to the offline control. PDM provides anomaly detection and diagnosis of the potential printing failure cases to improve process quality and reduce downtime[5]. The experimental results show that the PDM can achieve more than 87% accuracy in predicting different types of defects, improper printer hardware issues in the board support, squeegee, paste conditions, etc. The CPC uses the SPI information to estimate residue buildup level on the stencil undersurface and assess the stencil cleaning profile and cycle control, as illustrated in Figure 4[6]. Upon the implementation of CPC, it is expected that the robustness of the printing quality and the Cpk can be improved by 34% and 10%, respectively, compared to the best-known cleaning parameters using in the production line. Figure 4. Smart residue buildup prediction for stencil cleaning operation control During the P P procedure, the mounter optimization module (MOM) and the mounter diagnosis module (MDM) can be applied as a machine's automation process while optimizing the P P machine's parameters. By utilizing self-alignment effects appropriately, MOM identifies the optimal placement position by predicting the component's post-reflow positions based on the data collected by SPI, Pre-AOI, and Post-AOI machines, as shown in Figure 2. MOM also offers the placement positions adaptively during active production. In the MOM framework, multiple dynamic placement options are first generated based on the solder paste offset information. The components' final offsets in both x and y directions are predicted by a hybrid AI model that stacks on the k-nearest neighbor regression and the gradient boosting regression models. The optimal placement, which has the minimum predicted post-reflow misalignment, can be identified by MOM. The experimental results show that MOM can decrease 18% of the final misalignments compared to a conventional P P placement method (i.e., placing a component on the pad center). MDM is a prescriptive and predictive maintenance method that uses P P machine operational and AOI inspection data to trace back the root causes of P P defects and prevent future failure. MDM can achieve an accuracy of 84.50% in identifying the known root causes of certain defects, such as improper nozzle size, parts' contamination, and feeder problem. It shows that different mounting defects can be detected and classified automatically when the abnormality is detected through AI-based diagnosis algorithms. Figure 5. The illustration of the optimal placement position in the MOM One of the reflow oven issues to be addressed is to find the optimal reflow oven temperature settings, which would affect the final quality of the PCB products. Solder paste manufacturers usually provide a target profile based on the solder paste composition's physical properties, and solder joint temperature is required to meet the given profile. Hence, reflow engineers should fine-tune reflow oven temperature manually to ensure a thermal profile outcome from the reflow oven to correctly meet a target profile, requiring substantial cost and effort. The research team proposes an automated reflow recipe optimization model based on the PCB thermal profile and its recipe. Figure 6. Optimized thermal recipe and thermal profile First, the initial recipe collects the data for the prediction model and identifies the relationship between the thermal profile and the corresponding recipe. Then, an AI-based model is developed to predict the thermal profile based on the input recipe. Compared to traditional methods, the AI-based method generates an optimal reflow oven recipe to minimize the gap between the predicted temperature and the given profile. As a result, the AI-based prediction model allows us to achieve promising results, such as 97% of fitness in the given profile temperature curve within one hour of processing time. The proposed model has other significant advantages, such as saving time, labor, and materials. It enhances the degree of automation of the PCB reflow process. In the future, data from multiple inspection machines will be integrated so that the reflow optimization process is fully automated and generates more reliable results. Summary and Conclusion The small-scale electronics products make the SMA processes much more complicated to maintain high-quality PCB products, and theoretical interpretations of the SMA processes can be challenging due to many uncertain factors. With the help of AI and big data collected from various inspectional operations, SMA processes can be intelligent and flexible in response to dynamic environmental situations. While retaining the optimal control parameters throughout the SMA processes, the final PCB product quality can be enhanced while maintaining the designed throughput. Automated and smart systems bring about the opportunity to next level of electronics manufacturing, which utilizes the data and information from the end-users through edge/cloud computing and fastens the customized product manufacturing with increasing efficiency for high-mix/low-volume manufacturing. Also, it can increase verities of design and fasten the delivery time. About the Author Prof. Sang Won Yoon is a recipient of the SUNY Chancellor’s Award for Excellence in Scholarship and Creative Activities in 2019 and a highly successful researcher who leads many productive long-term industry collaborations. Prof. Yoon received his doctoral degree in School of Industrial Engineering at Purdue University, and he joined the faculty of the Watson School in the Department of Systems Science and Industrial Engineering at State University of New York at Binghamton in 2010. Prof. Yoon has been studying how to extract useful insights from expanding data sets to support intelligent decision-making processes. His research not only resides in better understanding large-scale data set by using statistical learning methodologies, but also leverages optimization, soft computing, simulation, and complex theories with conventional machine learning algorithms. As a result, Prof. Yoon has published in over 130 internationally renowned journals and conference proceedings. He was also a member of the Data Science Transdisciplinary Area of Excellence (TAE) initiative and is an active member of the Health Sciences TAE at his institution. The author recognizes the following for their assistance with this article: Daehan Won, [email protected], Assistant professor Jingxi He, [email protected], Ph.D. candidate Shrouq M. Alelaumi, [email protected], Ph.D. candidate Yuanyuan Li, [email protected], Ph.D. candidate Yuqiao Cen, [email protected], Ph.D. candidate References ​​​​​​​[1] Qi, Q., and Tao, F., 2018. Digital twin and big data towards smart manufacturing and industry 4.0: 360 degree comparison. IEEE Access, 6, pp.3585-3593. [2] 10 Ways machine learning is revolutionizing manufacturing in 2019. https://www.forbes.com/sites/louiscolumbus/2019/08/11/10-ways-machine-learning-is-revolutionizing-manufacturing-in-2019/?sh=7cd2e9e22b40. [3] Khader, N. and Yoon, S.W., 2018. Stencil printing process optimization to control solder paste volume transfer efficiency. IEEE Transactions on Components, Packaging and Manufacturing Technology, 8(9), pp.1686-1694. [4] Lu, H., Wang, H., Yoon, S.W. and Won, D., 2019. Real-Time stencil printing optimization using a hybrid multi-layer online sequential extreme learning and evolutionary search approach. IEEE Transactions on Components, Packaging and Manufacturing Technology, 9(12), pp.2490-2498. [5] Alelaumi, S., Wang, H., Lu, H. and Yoon, S.W., 2020. A Predictive Abnormality Detection Model Using Ensemble Learning in Stencil Printing Process. IEEE Transactions on Components, Packaging and Manufacturing Technology, 10(9), pp.1560-1568. [6] Alelaumi, S., Khader, N., He, J., Lam, S. and Yoon, S.W., 2021. Residue buildup predictive modeling for stencil cleaning profile decision-making using recurrent neural network. Robotics and Computer-Integrated Manufacturing, 68, p.102041.

As we pass the work-from-home one-year mark, most of us still work remotely and will do so for the foreseeable future. As live trade shows and technical conferences were cancelled one after the other, virtual events became the norm. And, teleconferencing became a way of life. While possibly overstating our role, we have the semiconductor industry – from system design through manufacturing and system integration – to thank for a long history of achievement that made the transition to working remotely relatively seamless and straightforward. The shift, in some cases, took some time to sort out as we set up a workable home office, moved to video conferencing with intermittent connections and settled into a routine. Nonetheless, many of us became more productive and, in some cases, even too productive. Each spoke in the global electronic products hub contributed through creativity and innovation with a pinch of ingenuity and grit. Of course, we could have worked remotely 10 years ago, but not nearly as efficiently. Over the last 10 years, the economy moved to the cloud, producing new opportunities across the global market. Many of these opportunities were made possible by the electronic system supply chain and combination of semiconductor technology, electronic product innovation and people who figured how to leverage it with software platforms to tie it together. Zoom, one of our teleconferencing lifelines, is a good example, as are Netflix, our ongoing source of entertainment, and Roblox, a platform to build games. Facebook, Twitter, LinkedIn and the like sourced the news for us and kept us in touch. Amazon delivered our online purchases and GrubHub brought us our takeout dinners. All rely on cloud computing with thanks to the semiconductor industry. Another great example are data centers powered by semiconductors and the amount of data they processed last year. According to International Data Corporation (IDC), 64.2 zettabyte (ZB) of data was created or replicated due to the dramatic increase in the number of people working, learning and entertaining themselves from home. (Its revised model for global data creation and replication predicts the CAGR will grow to 23% over the 2020-2025 forecast period, a sure bet that the semiconductor industry will address ways to manage the growth, possibly through new AI chips.) Our connectivity is driven by smartphones optimized for low power and the performance of more complex chips. Over the last 10 years, design tools have been enhanced and new methodologies have been introduced to respond to the needs of the increasing complex chips for applications that demand high bandwidth, low latency and reduced power consumption and area. Manufacturing is retooling for higher automation under smart manufacturing initiatives and packaging is even more sophisticated with increasing integration and the 2.5D and 3D packaging rollouts. Let’s take stock of our success. The semiconductor industry has a storied tradition of breakthrough technology since its inception. The consumer electronic product craze started when the first PCs were rolled out in 1971, notes the Computer History Museum. Primitive laptops that followed in 1986 gave way to notebooks in 2007 and the ubiquitous smartphone in 2002 – and the rocket fuel for much of this was the buildout of computer networks, hyperscale datacenters and the cloud. Nothing’s been the same since. The next time we turn on our laptop, click on the link for the latest teleconference from our remote home office in comfortable sweats sitting in our ergonomic chair, let’s take a minute to acknowledge our industry’s grand achievement. And, thank one and all for their contribution and consider what’s coming next. About the Author Robert (Bob) Smith is Executive Director of the ESD Alliance, a SEMI Strategic Association Partner. He is responsible for the management and operations of the ESD Alliance, an international association of companies providing goods and services throughout the semiconductor design ecosystem.

Ride the Wave of Smarter Manufacturing The year 2020 sparked a tremendous acceleration in the digital transformation worldwide, driving a sharp rise in demand for semiconductors and escalating pressure on chip factories to reduce manual functions on the shop floor. The mindset of the semiconductor industry saw a remarkable shift as it recognized with heightened urgency the need to deploy data-driven visualization, analysis, scheduling and dispatching solutions to increase automation to improve production speed and efficiency. Amidst the new excitement around Industry 4.0, chip manufacturers are rapidly deploying new technologies including IIoT, big data, machine learning and Autonomous Intelligent Vehicles (AIVs). Yet for many chip manufacturers, the path to building a smart factory is far from clear because they lack an overall digital transformation strategy. Smart manufacturing is a broad concept covering an array of technologies and solutions, making a holistic, mid- to long-term digitalization strategy rooted in the overall business strategy crucial. There are no shortcuts that can move a manufacturer instantly to Industry 4.0. Instead, this transformation is a step-by-step undertaking with a natural evolution. Some Factory Tasks Must Remain Manual – For Now The semiconductor industry has reached a point where manual processes are no longer efficient enough to support mass chip customization and remote operations. The many technological and standardization advances behind automation can help streamline some of a factory’s most labor-intensive tasks including the loading or unloading of machines or lot tracking and data collection while reducing operational costs. Still, some tasks remain very difficult to automate. For example, handling errors and exceptions presents the greatest challenge since some errors are hard to anticipate. What’s more, the cost of automating error handling can be prohibitive. Eliminating Gaps in Connectivity Often, critical data sources aren’t available due to lack of equipment integration, incomplete product quality monitoring or gaps in material tracking. Closing these gaps in connectivity enables the collection of data and provides rich, reliable information for analysis and reporting that can drive continuous operational improvements, optimizations and efficiencies throughout a factory. But keep in mind that data integration alone can be a challenging task. The selection and proper enrichment of relevant data is, in many cases, not just a technical problem but requires a detailed and in-depth knowledge of the manufacturing steps to be analyzed and optimized. Even when data is available, it might be still difficult to make decisions or implement improvements if it is in siloed systems that require manual processes to integrate and translate into useful information. Problem solving at this level is possible but extremely time-consuming. Manual integration is not only ineffective but costly, draining time, human resources and money from the factory. The right contextual information for the data is vital to unleash its potential and make improvements possible. Dispersed solutions cannot control processes because they span functional areas and people, physical and business entities. Backbone software for shop-floor operations that controls all other applications is central to smart manufacturing. Data-Driven Manufacturing The semiconductor industry is expert in data collection and leads many other industries in this area. The problem is often that chip companies use only a fraction of the information they collect for the analysis and insights needed to improve operational efficiency. By comprehensively integrating all distributed data into a single version of truth – in one location where it is always available – companies can make data analysis and problem solving almost frictionless. Keep in mind that data platforms and edge solutions, within the context of manufacturing, will not be adopted as part of a greenfield initiative. Building a solid automation architecture is only feasible and beneficial by deploying new technologies such as machine learning and artificial intelligence (AI). Analysis of historical data provides important context and reveals deviations such as unexpected process time, uncommon material accumulations or issues with material transport. By integrating swift control actions for new data point collected, manufacturing operations can shift from reactive problem-solving to proactive analysis and operational improvements. The tremendous increase in interest and investment in AI for manufacturing automation only became possible with the availability of low-cost sensors that generate huge volumes of data and solutions for storing and processing that at low cost. AI and other leading-edge technologies transform the tedious but critical process of extracting insights from data, making it instantaneous, streamlined and achievable for every manufacturer. The maturity of smart manufacturing hinges on the extent to which a factory is data-driven. This requires foundational investments to improve traceability, connectivity and real-time operations – and finally making sure that data helps us what to do and when to do it. Ricco WALTER is managing director of SYSTEMA Automation in Singapore.

SEMI spoke with Andreas C. Zimmer, Executive Search and Selection Consultant at ZIAN Co industrial consulting and recruitment, about strategies for attracting and retaining talent and promoting careers in semiconductor industry. Zimmer shared his views ahead of his presentation at the SEMI Fab Management Forum, 17 February, as part of the SEMI Technology Unites Global Summit, 15-19 February 2021, an online event. Join us to meet experts from ZIAN Co. and other key industry influencers. Registration is open. SEMI: What makes the semiconductor industry such a great career destination? Zimmer: The semiconductor industry is an interesting world for anyone involved in or just fascinated by high-end technology. But if we think about our mobile phones, personal computers or cars, we should all ask ourselves what technology is behind these devices we use in our daily life. The classical Newtonian physics does not reveal the source of the pixels in our mobile phones or why a navigation system knows where I currently am and how I’m supposed to drive to avoid the traffic jam ahead. The semiconductor industry truly is the technological pacesetter. The technologies and applications developed by SEMI and its members are the multipliers directly impacting our daily life. Moore's law not only affects the development of chips themselves, but also how we use the applications and devices they enable. Think about the size-performance ratio of modern smartphones compared to the first- and second-generation devices in the 1970s and 1980s, or compare today's BMW with one from the 1960s. The problem is that the industry is too hermetic. We perceive a lack of willingness to go out and tell in a generally understandable way what this industry is all about! Everyone knows Apple, Samsung, Nokia, but who, besides the specialists, knows NXP, Infineon, TSMC or LFoundry? Many companies are largely unknown to the general public! So why should a graduate from a technical university choose a company such as Applied Materials, TEL or ASML? During their studies students will inevitably have come in touch with IC or MEMS companies, but do they also know what is behind them? Do they really know the value chain that leads to the end product? SEMI: What can the chip industry do to better attract talent? Zimmer: Our industry is extremely attractive for anyone who is interested in technology and would like to push things ahead, but unfortunately access to this industry is almost reserved to the initiated who, in whatever way, came in touch with the industry at some point. Let me get this straight: This is not a conscious, willful attitude. It is just the result of our industry’s hermetic attitude. In my opinion, there is no overarching, uniform strategy in marketing, communications or advertising to promote the potential of the semiconductor industry to a wider audience. That’s why SEMI and the cooperation of its members in attracting talent is essential. SEMI: What concrete actions do you suggest for attracting and retaining talent? Zimmer: In German there is the saying “Do good and talk about it!” – and this is exactly what should be implemented. It is not enough to place an ad when necessary, to promote something here and there, perhaps to sponsor a chair or to provide a device free of charge. These are certainly all reasonable actions, but rather random and not long-term or strategic. Furthermore, these actions will reach only a relatively small group of people. The industry should organize structured recruitment activities under a long-term plan, over 10 years or even extending to the next generation. This shouldn't be a rigid corset, but rather a guideline closely informed by the chip industry’s technology roadmap and companies across the supply chain. If it is the task of an organization’s board and the management to define the strategic direction and to set specific goals, it should be the task of technical management to ensure that these goals can and will be achieved. However, this will only succeed if human resources is involved from the very beginning and can plan appropriate personnel resources accordingly. Employees retire, quit and change employers. New materials, technologies, applications and processes are being developed and require new, specific knowledge. Market requirements change. All of these components need to be recognized and considered in early planning. SEMI: What is your experience as a consultant? Zimmer: As consultants, we experience how organizations literally fall out of the clouds when the situation within the organization itself drastically changes, because a strategically important colleague is retiring or suddenly leaving the team for whatever reason. Then, quite surprisingly, the question “Where and how quickly can we find the suitable replacement?” arises. Instead, that departure should be considered as a possible development up front in overall talent planning – a plan B to keep in the drawer. Developing and implementing a long-term HR development roadmap, aligned with the technology roadmap, enables a company to anticipate when specific resources are needed, identify the right people and get them onboard without gaps. It is also important to keep your team informed and involved in all decisions and process changes, and to make sure they get the respect and appreciation they deserve. Employer-employee cooperation over the long term only works when the relationship is a win-win for both parties. If an organization sees the relationship as one-sided to its exclusive benefit, sooner or later the worker will be terminated or quit at the expense of the organization. Truly live the statement “Our people are our best and most valuable resources!” SEMI: When should organizations start attracting young talent? Zimmer: The sooner, the better! Communications aimed at attracting future employees should be designed to reach people of all ages and levels of education. For many years, the tobacco industry targeted young people by demographic, considering their age, education and cultural mindset to ensure they perceived cigarettes as cool. The result? Many people became addicted, mostly for life, just because some clever communications expert touched the right spot! Our industry will not attract teenagers like tobacco corporations did, but the strategy is basically the same: arouse the curiosity of your target group and speak their language. A possible scenario: A company starts and establishes a relationship with neighboring technical, middle and high schools by providing equipment, documentation, and employees who serve as teachers or coaches, and organizing guided tours, seminars and workshops in coordination with the school management. The cooperation continues with the university, where the respective chairs are supported and financed. With a little creativity there are endless possibilities! In our day-to-day business, we observe that large, well-known companies such as Bosch and Daimler are practically sitting on the lap of students in key universities and institutes, yet are unable to identify talent very early and bind them to their company. SEMI: How can organizations capitalize on shifting retirement patterns to help narrow their talent gap? Zimmer: The answer to this arises from considerations related to personnel planning in connection with a company’s technology roadmap. If the roadmap is linked to HR plans, you automatically have an overview of the time-critical moments when personnel gaps might arise. Then you can easily close these gaps, for example by arranging the onboarding of a successor for a specific position long before the job holder leaves. Considering notice periods and approval processes, a period of at least two years should be planned in order to be prepared for personnel changes. Of course, much of this varies depending on the importance of the position to the organization and the size of the talent pool. For example, it will probably be easier and faster to hire and train a sales engineer than the successor for a development manager, when you know there are maybe only 10 people worldwide who are, professionally speaking, at his level. And this is equally true for internal promotions: Always keep an eye on your own people and try to discover their greatest talent! Senior people tend to look outside the organization rather than just around the corner. Maybe the right talent is sitting next to you. Stay tuned and talk to your people to implement a strategic knowledge transfer as part of your organizational culture. Another aspect that is often overlooked is the deputy function: We often find functions in organizations that literally have a unique selling proposition. But there is no deputy, no one who can step in case of an emergency, because no other colleague possesses the knowledge and information to take over if necessary. Usually this is not a problem during a vacation or illness, but what do you do if a key job holder suddenly cannot work from one day to the other? SEMI: What is the role played by artificial intelligence? Zimmer: AI is both a risk and an opportunity. A new technology can always mean danger if it is used incorrectly, and I am not talking about job losses! This has always proven to be a mistake in the past. On the contrary, new technologies create new jobs! New technology accelerates communication, creates new platforms for interaction, shortens decision-making processes, and turns the world into a small village. In your interview with David Meyer CEO of Lynceus, he hits the nail on the head: The great advantage of AI in our industry is likely to be the management, handling, analysis and drawing of conclusions from an incredible amount of information at an unbelievable speed. Without AI, information cannot be controlled to this extent, not to mention accurately evaluated in real time. The mastery of these processes and the learning curve that results from them – for example for the determination of quality levels – should set completely new manufacturing standards. SEMI: How can technology unite us? What do you expect from your participation at SEMI Technology Unites Global Summit? SONAR GmbH has been in this industry as a personnel and business consultant firm for 25 years now. We have experienced many pig cycles since 1995 and accompanied our customers through all the ups and downs, only to have learned one thing in the end: The semiconductor industry is unfortunately still too fixated on technology and overlooks the fact that this technology is made by people for people. The EU's latest Pact for Skills, which was presented at end of November 2020 by Commissioners Schmidt and Breton, foresees 2 billion € investment to generate 250,000 new jobs in the electronics industry throughout Europe! In 2013, we aimed to sensitize semi industry executives, managers and CEOs to the importance of human resources to the well-being and success of organizations. It’s vitally important to invest in day-to-day relationships with your employees to foster their careers and address their needs. The SEMI Fab Management Forum will feature leading game changers of semiconductor operations to highlight best practices for achieving sustainable operations beyond 2020 and exploring the latest solutions for smarter tools and smarter processes. Andreas C. Zimmer is executive search and selection consultant at ZIAN Co industrial consulting and recruitment, specializing in recruiting talent for high-end technologies in areas such as LED, PV, semiconductors, electronics, and test and measurement. A personnel and industrial consultant with more than 20 years of experience, Andreas is active throughout Europe, the United States and Asia. For more insights about workforce and skills strategies, please see SEMI Workforce Development activities and the European METIS project. Serena Brischetto is senior manager of Marketing and Communications at SEMI Europe.