Industry 4.0

The state of manufacturing is changing rapidly. Regardless of sector or location, manufacturing decision-makers across the world are signaling a desire for better supply chain resiliency, manufacturing flexibility, increased speed of innovation and stronger environmental sustainability. Singapore’s manufacturing sector, a significant contributor to its gross domestic product, is always evolving and today is shifting away from its traditional focus on producing highly customized products using flexible manufacturing processes, but at significantly lower efficiencies. Today, with Industry 4.0, we can design manufacturing systems that optimize both efficiency and flexibility. And this is possible because of the convergence of technologies such as artificial intelligence (AI), data analytics, robotics and the Industrial Internet of Things (IIoT). This blend of technologies helps reduce the cost of technological solution ownership – a derivative of Right’s Law – as a function of cumulative production. In HP Singapore, driving innovation in our product and processes is part of our DNA, and over time our products have grown in complexity and breadth. We have embraced Fourth Industrial Revolution (4IR) technologies in our advanced manufacturing lines. We started our Industry 4.0 journey in 2016 with Vision and Mission 2020 to modernize our production facilities to smart factories that strengthen our competitive edge. Our focus was on upskilling our employees with future skill sets, build new technological capabilities and partner with higher education institutes. To drive these transformations, we have formulated five pillars: Additive Manufacturing Data Analytics Cyber-Physical Integration Digitalization Workforce Transformation These five pillars have enabled us to move from labor-intensive and reactive processes to processes that are highly digitized, automated, and AI-driven, enabling us not only to increase quality and productivity but also to reskill our people in anticipation of jobs they will need in the future. Technicians have been upskilled and promoted to techno-operators which has, in turn, freed up technical specialists to explore other roles. Engineers have retrained as data scientists, or have moved to new product development, for instance. In 2017, HP’s Ink Supplies Operations (ISO) set up Smart Manufacturing Applications and Research Centre (SMARC) to adopt 4IR technologies and implement these innovations in production lines. Today, SMARC is the home ground for HP engineers to experience, trial and prototype solutions, bringing innovative and sometimes unexpected solutions to manufacturing. It is also a showcase for industry partners, government agencies and schools. Here is how each pillar of the SMARC contributed to transformation to augment the manufacturing workforce: Cyber-Physical Integration – Move Role of robotics/automation – By standardizing automation standards for robotics, we have deployed collaborative robots (Cobots) and autonomous intelligent vehicles (AIVs) to perform manual and routine tasks to drive productivity, while reducing errors from operator fatigue and protecting our operators’ physical well-being. Digitalization – Sense Role of IIoT – Devices are a treasure trove of data that can provide clarity on how the entire manufacturing line is performing in real time. Building a platform that connects devices and collects data while allowing factory floor managers to dynamically visualize on an Integrated Command Centre (ICC) and manage factory performance is central to HP’s digital transformation journey. And IIoT is not restricted to just devices that are already wired for data sharing. HP has also connected off-the-shelf analogue devices using a standardized data transportation protocol, allowing HP to collect essential data across all types of devices and eliminating manual data entry. Additive Manufacturing – Build By embracing additive manufacturing (use of HP MultiJet Fusion 3D printers), HP introduced more flexibility in operations through on-site rapid prototyping, light production, and replacement of parts needed on our manufacturing floors, shortening production timelines. We 3D printed pallets, which are cheaper and faster to produce, and replaced original pallets for transportation on conveyor belts, improving the efficiency and productivity of our operators. Director Jamie Neo with HP’s MultiJet Additive Manufacturing Printer. (Photo Credit: HP) The HP Multi Jet Fusion 3D printing technology has helped HP to replace traditional manufacturing methods and streamline processes in our supply chain. For example, HP is 3D printing the Drill Extraction Shoe, a tool that is essential to the removal of waste products from laser-drilling in HP’s printhead manufacturing line. Through 3D printing, HP has consolidated the production of the tool from nine parts to one 3D printed model, thereby optimizing the design of the tool and reducing its production time from three to five days to 24 hours. Data Analytics – Think By deploying advanced analytics and machine learning models, HP has enabled real-time detection, diagnostics, and prediction of product quality across our manufacturing lines. Predictive models are replacing traditional “destructive testing,” reducing waste and allowing HP to meet unique product specifications more accurately. Machine learning is diagnosing and recommending the right set up for tools and manufacturing lines, when necessary, to reduce downtime and increase precision. Workforce Transformation – Grow The pivot to becoming an advanced manufacturing leader not only requires HP to invest in 4IR technologies but also skill sets to operate 4IR technologies. We embarked on a Workforce Transformation program to help our employees stay competitive in a fast-changing world. Today 35% of HP technical workforce have had the opportunity to take on new roles even as needs evolve, thanks to internal and external training and reskilling. Beyond technology and training, the glue that binds these together and makes it successful is our culture at HP. We are ambition-led, which means that we do not see the world as it is, but what we can be. And we do so by collaboration. Plans for the Future After accomplishing our Mission 2020, in late 2020 we launched Mission 2025 to extend our end-to-end smart factory capabilities through advanced connectivity, intelligence and automation to optimize and drive sustainable manufacturing flexibility and efficiency. Pyramid of HP’s smart manufacturing focus Advanced technologies such as additive manufacturing, IIoT, automation and robotics, data analytics, machine learning and AI are central to the connectivity and the end-to-end intelligence of our smart factories, enhancing production efficiency and flexibility while improving the quality of our products. For example, the deployment of IIoT sensors in our wafer plant has helped to reduce downtime in replacing CO2 gas cylinders. What’s more, AI enables us to more accurately monitor the dispensing of structural adhesive to eliminate lost yield. We believe that by enhancing manufacturing efficiency and flexibility, we were able to shorten resolution time, reduce our carbon footprint, and improve the resiliency of our manufacturing and supply chain systems. HP smart factory model In April 2021, two lines in HP Singapore joined the World Economic Forum’s Global Lighthouse Network after being recognized for pivoting from a labor-intensive factory into a digitized, automated one with the help of AI. In doing so, we managed to improve manufacturing costs by 20% and productivity by 70%. Under Mission 2020, we saw the following successes: Improved manufacturing costs by 20% Improved productivity by 70% Brought most HP employees onboard to our smart manufacturing journey Equipped HP employees with skill sets in areas such as additive manufacturing, data analytics, AI, robotics and Internet of Things Established a Model Factory playbook With Mission 2025, we will: Continue to train employees in future skillsets by partnering with institutes of higher learning Scale our Model Factory playbook across more manufacturing lines to reduce costs and improve productivity Enhance our knowledge in additive manufacturing by building an ecosystem as a service platform to help manufacturing companies Enable a sustainable manufacturing system to reduce our carbon footprint and help enable a circular economy We believe in innovating with purpose by focusing on solving real-world problems and creating technology in the service of humanity. That is why we built the SMARC to create the solutions for our lines and showcase these solutions to encourage industry participation. We are driven by values and ambition, which means that it is not just what we do, but also how we execute it. We make sure our values inform everything we do – for instance, helping us make a greater impact to environmental sustainability, people, and our community. We believe this is a crucial step in coalescing industry support, which is necessary to move the needle on advanced manufacturing. Robert Ronald is Master Program Manager, Cost Structure, Model Smart Factory and Sustainability, at HP.

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.

Connecting product development to manufacturing to the field in the semiconductor equipment industry by becoming a Model-Based Enterprise Manufacturers across many industries, including the semiconductor equipment industry, are facing pressure to dramatically reduce product development cycle and production ramp times while also enhancing product quality and reliability. This challenge is complicated when multiple configurations are deployed and then maintained, enhanced and upgraded in the field. Buzzwords like digital thread, digital twin, smart manufacturing, Industry 4.0, and digital supply networks point toward a fusion of process, technology, data, and talent that promise the game-changing outcomes needed to address these challenges. Yet, there remains uncertainty about how these various elements come together into a cohesive approach across the product lifecycle. One such approach involves establishing a Model-Based Enterprise (MBE), with alignment across organizational silos, processes, and technologies. Too often, however, an effort to transform to an MBE is frustrated by “random acts of digital” that pursue implementation of certain digital technologies but fall far short of delivering value. What is needed instead is a deeper understanding of the characteristics of a true MBE and the tools and approaches to transform the organization, processes, and data to an MBE and thrive in the marketplace. Elements of a Model-Based Enterprise Modeling in engineering is not new, but what’s emerging are MBEs that comprise digital models connected upstream and downstream over the entire product lifecycle, from a product’s conception and development through its production and end market installation and use. At the heart of such MBEs are digital threads—integrated sets of processes executed within an interconnected technology ecosystem that drive the end-to-end product lifecycle and provide MBE data traceability front to back. A digital thread in the MBE environment includes all of the process, data, and system capabilities that enable digital representations of the product lifecycle stages, or digital twins, of which there are three primary types: The product digital twin is a virtual or simulated representation of the product and each of its components and configurations. While most manufacturers manage engineering models with CAD/CAE solutions, just 15% use product digital twins. Leaders in this space have seized a competitive advantage in product development and accelerated engineering. Yet, becoming a true MBE requires more than product digital twins. The process digital twin is a model of the manufacturing equipment, processes, and the workforce required to carry out related operations. The process digital twin represents the operation of the physical factory floor and its assets, complete with workflows and instructions that describe how the manufacturing processes are performed. A process digital twin relies on data from the product digital twin and allows the enterprise to build according to a product plan and predict what may happen on the factory floor. About 5% of enterprises use process digital twins. After production, a service digital twin represents the installation, use, maintenance, and repair of each product operating in the end market. The service digital twin is informed by product and process digital twins to facilitate adjustments and enhancements based on real-world data. Less than 5% of manufacturers use service digital twins, which is perhaps expected due to their reliance on the existence of the product and process digital twins. When underlying data (e.g., models, specifications, and configurations) are standardized and integrated across a digital thread, an enterprise has the capacity to monitor and refine a product over the span of the thread while also injecting insights and improvements back into the thread. The result is that engineering and manufacturing and end customer usage feedback is continuous and efficient, achieving in weeks what once took months. Visibility into materials, costs, suppliers, and more enable the enterprise to pivot and keep production moving, even when dealing with unforeseen challenges. Real-time monitoring that synthesizes live data also helps reveal performance insights and end market issues (e.g., installation issues, quality issues, etc.) that allow improvements to the offerings and operations of the enterprise. While the manufacturing and product development benefits of an MBE may be clear, the path to becoming an MBE and achieving these benefits can be challenging. Figure 1. Representative MBE end-to-end digital thread that connects product development, manufacturing, and the field (Source: Deloitte Development LLC, 2021) Accelerating Transformation New and emerging manufacturers have an opportunity to build toward their MBE vision without the historic data constraints legacy systems can impose. For more established manufacturers, such is more typical in the semiconductor equipment space, legacy technology and processes can present obstacles to an MBE transformation path. Stakeholders who have invested time and resources implementing certain enterprise platforms (e.g., CAD, PLM, ALM, MES, ERP, etc.) often look for how these can be used to enable a digital thread and digital twins over the enterprise. This limiting view frustrates a broader, more holistic opportunity to transform to an MBE and thrive in the marketplace. There is thereby a dual imperative to define a modernized and scalable end-to-end technology architecture for managing the MBE product data while also establishing a capabilities implementation roadmap that rapidly leads to MBE maturity. To be successful in your MBE transformation, four core functional capabilities that enable a digital thread must be considered: digital engineering, industrial simulation, manufacturing execution, and real-time monitoring. Addressing each with optimized tools allows a manufacturer to rapidly move from strategy to reality. Recognizing the complexity of addressing these functional needs, Deloitte has developed preconfigured solutions to help expedite and enhance transformation across each of these core areas. Design with D-PLM Simulate with D-Sim Execute with D-MES Monitor with D-IoT Accelerates product and application lifecycle management transformations with a multi-phased approach, including a phased, multi-year PLM/ALM roadmap and business case. Facilitates the ability to test and refine processes in a virtual environment, rapidly revealing the most efficient and effective industrial processes more quickly than is possible in a real-world environment. Integrates pre-defined processes related to production planning, execution, tracking and tracing, quality management, data collection, and visualization, with integrations to PLM and ERP. Delivers fast implementation of IoT capabilities that connect, collect, and analyze a broad scope of production data to drive quicker returns for high-impact areas while cultivating digital adoption. While many enterprises have various initiatives in model-based systems and manufacturing, few have tied them together with an end-to-end digital thread and set of data standards over the entire product life cycle. The foregoing preconfigured solutions can help enterprises transform to a true MBE that can typically achieve: 15% – 20% better development efficiency 30% – 50% faster time to market 8% – 20% product cost reduction 10% – 30% cost of quality reduction With such improvement potential, this could be the right time to map out and accelerate your MBE transformation to support your evolving business models and products. Deloitte Consulting LLP Co-Authors Kevin Prendeville Principal, Product Strategy Lifecycle Management [email protected] Vijay Santhanam Managing Director, Product Strategy Lifecycle Management [email protected] Kenneth Norton Senior Manager, Product Strategy Lifecycle Management [email protected] Dan Hamling Specialist Master, Technology Semiconductor [email protected] As used in this document, “Deloitte” means Deloitte Consulting LLP, a subsidiary of Deloitte LLP. Please see www.deloitte.com/us/about for a detailed description of the legal structure of Deloitte USA LLP, Deloitte LLP and their respective subsidiaries. Certain services may not be available to attest clients under the rules and regulations of public accounting. This publication contains general information only and Deloitte is not, by means of this publication, rendering accounting, business, financial, investment, legal, tax, or other professional advice or services. This publication is not a substitute for such professional advice or services, nor should it be used as a basis for any decision or action that may affect your business. Before making any decision or taking any action that may affect your business, you should consult a qualified professional advisor. Deloitte shall not be responsible for any loss sustained by any person who relies on this publication. Copyright © 2021 Deloitte Development LLC. All rights reserved.

The SEMI Smart Manufacturing Americas Chapter, a key driver of the Global Smart Manufacturing Initiative, accelerates awareness of digital and data-driven strategies and implementations to help speed adoption of smart manufacturing. In 2021, the Chapter will focus on expanding its work across the industry to include academic and research initiatives. The semiconductor industry saw an unprecedented focus on improving digital monitoring of manufacturing activity in 2020, partially due to COVID-19. The Americas Chapter shared case studies on new tools and techniques for social distancing in fabs, aides for remote maintenance, and tips for remote workers. The Chapter also introduced its three pillars of Sensing, Connecting and Predicting and offered related programs. The Global Smart Manufacturing Conference (GSMC) highlighted the significance of universities and research institutions in the development of smart manufacturing with their focus on joint research for broad dissemination. To help drive smart manufacturing advances, at GSMC several offered non-proprietary tutorials on topic including the following: Integrating sensors for acquisition – CEA-Leti Applying new AI and ML tools and strategies to manufacturing – Binghamton University Digital tools for planning, qualifying and management and scheduling in fabs – MINES Saint-Étienne. Adding AI tools to robot work in a smart factory – KAIST Institutes By continuously highlighting the activities of these and other institutions through presentations, interviews, articles and blog posts, we will draw more attention to what is on the horizon for smart manufacturing in 2021. The SEMI Smart Manufacturing Americas Chapter also plans to elevate activities important to the Outsourced Semiconductor Assembly and Test (OSAT), Surface-Mount Technology (SMT) and Printed Circuit Board Assembly (PCBA) segments of the industry including programs on inspection, traceability and the SEMI SMT-ELS Standard for SMT automation. Thurston Taylor, marketing expert at Tokyo Electron and Vice Chair of the Americas Chapter, notes that “With increasingly more demanding requirements for bump, assembly and test, smart manufacturing and applied data science are necessary to achieve back-end goals now and in the future.” Also, many companies are implementing smart manufacturing applications and assessing various strategies to increase their smart manufacturing capabilities. Members of the Americas Chapter plan to review and develop self-assessment documents and maturity models that apply to front-end wafer fabs all the way through packaging and assembly facilities. “Moving forward it is imperative for all of us to up the intensity on specific ROI vectors such as quality, cost, productivity, sustainability and safety leveraging our smart manufacturing SEMI framework of Sensing, Connecting and Predicting,” said noted Bobby Mitra, worldwide director of Smart Manufacturing at Texas Instruments and Americas Chapter Chair. “By offering special flagship events, invited talks, ROI case-studies and ROI criteria in maturity models, we’ll bring high value to the smart manufacturing industry.” Chapter members also will begin mapping the skills needed to implement and support increasingly digital manufacturing capabilities, including any new skill sets, to help companies develop their hiring, training and management strategies. The mapping effort aims to support companies in building a strong pipeline of employees who can efficiently manage and operate smart manufacturing facilities. For its part, the Americas Chapter’s Go Green Subcommittee will focus on applying smart manufacturing technology to reducing the electronic industry’s carbon footprint by accurately tracking energy waste improving overall fab efficiency. Stay tuned for details on activities planned for our chapters in Europe, China, Japan, Korea, Southeast Asia and Taiwan. To learn more about each chapter and how to get involved, please visit the SEMI Smart Manufacturing Hub and sign up for our newsletter. Ayo Kajopaiye is senior project coordinator, Collaborative Technology Platforms, at SEMI.

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.

Electric mobility, renewable energy and other technology innovations like IoT, 5G, smart manufacturing and robotics all require reliability, efficiency, and compact power systems, fueling the adoption of Silicon Carbide (SiC) and Gallium Nitride (GaN) to support lower voltages in significantly smaller devices. But chip designers must overcome the technological and economical challenges of integrating the two semiconductor materials into power systems.SEMI spoke with Elisabeth Brandl, Business Development Manager at EV Group about trends and new developments within the power electronics industry and the devices' application in smart mobility. Brandl shared her views ahead of her presentation at the SEMI SMART Mobility Forum, 18 February, as part of the SEMI Technology Unites Global Summit, 15-19 February 2021, online event. Join us to meet experts from EV Group and other key industry influencers. Registration is open. SEMI: What is driving new developments in power electronics?Brandl: Globally there are significant changes in infrastructure requirements for communication, automotive and power conversion. We need to look no further than the rising adoption of 5G, electric and hybrid vehicles, and renewable energy as examples of drivers of these changes. The device level, particularly in the field of power electronics, figures prominently in these shifts.The power electronics industry faces a growing number of scenarios where conventional silicon power devices are no longer suitable and are easily outperformed by new architectures mainly based on wide bandgap semiconductor materials like Silicon Carbide (SiC) and Gallium Nitride (GaN).SEMI: What industry challenges is power electronics innovation aiming to solve? Brandl: Power conversion efficiency is very important and needs further improvement as the related losses significantly contribute to the overall power consumption. For green power and a better environmental footprint, renewable energy is crucial, but so is overall power-consumption efficiency, yet the role of power devices is often underestimated. High-frequency and high-power applications, such as data center applications and inverters for renewable energy, where silicon power electronics are reaching their limits, are also important areas in power electronics.SEMI: How will the transition from silicon to compound semiconductor materials help?Brandl: The superior material properties of several compound semiconductors can tackle the need for lower losses in power conversion or better high-frequency behavior. Today, we mainly talk about GaN and SiC power devices as they are materials well-suited to address these needs. However, other materials like diamond and gallium oxide are in development for these applications. Material properties of SiC that enable thinner materials with lower power losses and better thermal behavior address power conversion efficiency as well as form factor challenges. GaN, especially in a high electron mobility transistor (HEMT), can be used for high-frequency applications.SEMI: What enables a better and more cost-effective manufacturability of SiC and GaN power devices?Brandl: For the end customer, a typical figure of merit regarding the cost effectiveness is $ per Ampere or Watt. While this seems simple, the reality is of course more complex. It is important to understand the main cost contributors within the manufacturing area. For SiC, this is clearly the substrate cost. In my presentation, I will show a way to reduce this cost via wafer bonding. For GaN, epitaxy – a method for growing or depositing mono crystalline films on a substrate – is the critical parameter. And of course, yield has a very big impact on cost effectiveness too, which means that good process control including metrology is very important.SEMI: Many semiconductor companies are already transitioning to silicon carbide and gallium nitride. Can you give us an example of a success story?Brandl: All the big power device manufacturers have either acquired or developed their SiC and/or GaN power device technology, so they also see a bright future for these wide bandgap semiconductors in the power device market. The most prominent success story is STMicroelectronics with its SiC MOSFET power devices, which have been implemented by Tesla in its Model 3 vehicles since 2018.SEMI: What is coming next?Brandl: New materials for power devices are being explored, such as diamond and gallium oxide. For SiC, the trend is moving toward 8-inch substrates, which is the focus of the funded EU project REACTION under the coordination of STMicroelectronics. Cost reduction and substrate availability also play a big role. All major power device manufacturers have contracts to secure the supply chain for SiC substrates because material availability is the main uncertainty at this time. Finally, collaborations along the supply chain are crucial and generally beneficial for all parties, as development requirements are better communicated and prioritized.Elisabeth Brandl is Business Development Manager at EV Group. She received her master in technical physics from the Johannes Kepler University Linz, Austria in Semiconductor and Solid State Physics. Since 2014, she has been responsible for Product Marketing Management for temporary bonding and compound semiconductors at EVG. The SMART Mobility Forum is the digital platform of SEMI Europe’s Global Automotive Advisory Council (GAAC) for industry stakeholders along the automotive and electronics value chains, from Design, Semiconductor Equipment and Materials Suppliers to Automotive OEMs.Smart Mobility is one of four SEMI initiatives focused on building communities, content, and activities around critical and emerging electronics markets. Read more about our Regional Chapters.Serena Brischetto is senior manager of Marketing and Communications at SEMI Europe.

Nexperia became a standalone company about four years ago after our divestiture from NXP Semiconductors. Last year we started our journey towards smart manufacturing at our back-end factories in Asia by developing a roadmap to help steer us in the right direction.Our first step to creating a convincing and workable smart manufacturing roadmap was to define the very meaning of smart manufacturing to Nexperia. Since the definition of smart manufacturing varies widely, we started by looking at two different and distinct technology adaptations: Physical automation Data-driven manufacturing, or using analytics at the core to develop and adopt machine learning and artificial intelligence (AI) models It is important to find the right balance of investments between physical automation and data-driven manufacturing to steer clear of deployment inefficiencies since only connected solutions deliver full value. Our approach involved the following high-level steps. Meeting with internal management teams for their inputs and examining factory needs and maturity Meeting with other semiconductor factory operators, subcontractors and partners to review their smart manufacturing approaches and challenges Evaluating our needs and status against the Singapore Smart Industry Readiness Index model Physical AutomationEvaluating the maturity of available solutions and adaptions by the industry and our own shop floor helped simplify the thought process quite well. Logistic automation is not new. Very mature solutions, even for custom layouts and preferences, are readily available. Shop floor automation is far more difficult than logistic automation since variability is simply too high. Traditional shop floor investments were always driven from quality or OEE perspectives and not necessarily very well connected. Our approach is outside-in – deploy logistic automation first and then move to the shop floor.Data-Driven ManufacturingHow smart manufacturing becomes depends on the extent to which a factory is data-driven. Enabling data-driven manufacturing requires foundational investments to improve traceability, connectivity and real-time operations. We believe real-time awareness can drive machine-level and closed-loop process control critical for predictive, cognitive control of the shop floor.Real-Time Awareness and Traceability is at the CoreDeveloping real-time awareness requires wide-ranging manufacturing protocols. The following focus areas have helped us simplify the challenge: Connectivity Core systems for areas including MES, quality and SAP Analytics and AI Digital shop floor featuring one operator interface with real-time control systems Readiness of engineers, technicians and managers Each of these pillars has different level of complexity due to legacy equipment and systems, legacy processes and inexperience of engineers with automation. This makes deployment of data-driven operations a complex challenge. We looked at different project approaches for each of the focus areas: Core Systems – Build additional technology enablers and roll them out with prioritization planning. Analytics – Focus mainly on OEE and yield with automated root cause analysis and predictive approaches. Real-Time Control – Merge the initiative with factory-level programs to improve productivity and quality. With a strong smart manufacturing roadmap, the next challenge is to secure long-term buy-in on the plan and required investments from executive management. Visiting and otherwise connecting with peer sites that have already deployed smart manufacturing infrastructure is vital in this effort. Thanks to SEMI members, we were allowed to visit their factories with our management team for go-and-see tours since seeing is believing in the smart manufacturing journey. Our executives also met with subcontractors and vendors to better understand the value of this transformational undertaking.A long-term outlook is necessary to successfully develop a smart manufacturing roadmap, and executive commitment goes a long way to ensuring its success. We are excited about our smart manufacturing journey and believe it is a game changer for our factories.Adarsha MARPALLI is director of Factory Automation at Nexperia B.V.

Semiconductors play an essential role in modern society by enabling ground-breaking technological advances. The manufacture of high-volume and advanced semiconductors requires the use of fluorinated chemicals known as PFAS. Representing the voice of SEMI members, I explained the important role of these substances and their “essential use” in the semiconductor manufacturing supply chain at a Chemical Watch conference for industry and European Union decision-makers on 3rd of December 2020.In order to achieve the European Green Deal’s zero pollution ambition for a toxic-free environment, the European Commission announced in its recently published Chemicals Strategy for Sustainability its intention to restrict the use of the most harmful chemicals, except in cases where they are deemed essential for society. Per- and polyfluoroalkyl substances – known as PFAS – are the first group of chemicals facing regulatory scrutiny on this basis. This begs the question: What chemicals should be characterized as essential for society and what uses will they encompass? The key and enabling role of semiconductors in modern lifeSemiconductors are essential and ubiquitous in our lives. They are integral to enabling modern society to function – driving advancements in mobile communication technologies for the smartphones and computers that help us work more efficiently and connect us with our loved ones. These benefits have never been more evident than in 2020 with billions of people finding themselves working and studying remotely and safely from home.At the same time, technologies relying on semiconductors have been vital in the effort to combat COVID-19 – in ventilators, medical imaging devices and digital healthcare solutions. In addition, semiconductors will also enable the next leap in society to Industry 4.0 and as essential building blocks in connected and electric vehicles, artificial intelligence (AI) and quantum computing.The Commissioner for Internal Market, Thierry Breton, has highlighted the strategic importance of semiconductors in achieving European digital sovereignty (for instance, in his speech at Hannover Messe Digital Days), and the EU’s New Industrial Strategy[1] also points to the importance of semiconductors and microelectronic systems. What must also be appreciated are the cost and complexity of producing these valuable technologies. Setting up a cutting-edge fabrication plant with the hundreds of pieces of semiconductor manufacturing equipment typically required can cost around €15 billion.[2] A single semiconductor manufacturing tool typically consists of millions of articles, and a typical fab may house several hundred pieces of equipment. Furthermore, according to SEMI estimates, the fabrication of semiconductor wafers requires approximately 500 highly specialized process chemicals. In many cases, these processes, equipment and facilities rely on the unique properties offered by PFAS.“SEMI has worked diligently to highlight the strategic importance of semiconductors in achieving European digital sovereignty, and we are pleased that the critical role of microelectronics has been fully recognized by the EU and Member States. Fluorinated chemicals are essential for semiconductor manufacturing. "These specific chemicals are necessary due to their unique properties, and no alternatives are currently available that can adequately provide the functional properties required in semiconductor manufacturing. The essential use concept, therefore, must enable technological innovation, must apply across the entire supply chain, and must enable EU’s critical infrastructure and strategic objectives.” What are PFAS, and why and where are they used in semiconductor manufacturing?PFAS are a broad and highly diverse group of substances with unique properties and characteristics. The Organisation for Economic Co-operation and Development (OECD) has compiled a list of approximately 4,700 substances,[3] a handful of which are used in the semiconductor manufacturing industry. These very specific chemicals are necessary due to their unique and unparalleled properties that enable them to be used in the demanding conditions of semiconductor manufacturing.Semiconductor chemicalsAt the very core of semiconductor manufacturing is the photolithography process, where microscopic geometric patterns are transferred onto a film or substrate. Photolithography specialty formulations containing fluorinated compounds are used in various steps of this process to ensure quality and reduce the probability of defects. PFAS must be used due to their low surface tension and compatibility with other chemicals. PFAS are typically no longer present in the finished product. However, there are applications where PFAS are present in the final semiconductor device, particularly in imaging semiconductors used in cameras, displays and some medical devices, amongst others. Semiconductor manufacturing equipmentPFAS are also essential to semiconductor manufacturing equipment and factory infrastructure. The exceptional combination of their heat and chemical resistance and their chemical inertness allows fluoropolymers to be used both in equipment components (tubing, gaskets, containers, filters, etc.) and lubrication (such as various oils and greases). These same properties are also needed to ensure the functioning of the surrounding infrastructure. Finally, some fluorinated gases, which are already regulated by specific legislation,[4] are used as refrigerants and to clean the facilities.These are a handful of examples of how PFAS are used in semiconductor manufacturing. Today, there is no other way to undertake these processes or to build semiconductor manufacturing equipment without PFAS. No alternatives are currently available that can adequately provide the functional properties required. Even if alternative chemicals and technologies were discovered today, due to the extremely complex qualification process throughout the value chain, it would take another 15 years to deploy them in high-volume manufacturing. Therefore, continued access to PFAS is a prerequisite for high-volume and advanced semiconductors. Lack of continued access to PFAS could lead to an inability to produce and supply the EU with semiconductor manufacturing technology.How should we think about essential uses?Regulators have started to think about what uses of PFAS are essential and in which cases their use should be allowed. In developing this concept, there are a few aspects to keep in mind.Essential use must enable, not hinder, technological innovationFirst and foremost, the essential uses concept should enable continued technological innovation instead of acting as a hindrance. Semiconductors and manufacturing technology are constantly evolving and becoming more diverse to help meet increasing societal demands. What we see as innovative today may be commonplace in the future, while future innovations may be unimaginable today. We must therefore be careful not to accidentally limit our future potential for innovation.Essential use must apply across the entire supply chainSecondly, classifying a use as essential should apply throughout the entire supply chain. We must, for example, avoid defining semiconductors as essential while classifying the semiconductor manufacturing equipment and chemicals used to produce semiconductors as not essential. In the semiconductor manufacturing supply chain, where one manufacturer can have up to 16,000 suppliers, this risk is evident.[5]Essential use must enable critical infrastructures and the EU’s strategic objectivesFinally, we should keep Europe’s societal priorities in mind. The EU needs to be able to maintain and protect its critical infrastructures. Similarly, we should not lose sight of the EU’s strategic objectives of a green and digital Europe.Semiconductors, in conjunction with their corresponding manufacturing equipment and chemicals, are essential technologies in everyday life and the backbone of the EU’s strategic value chains. Manufacturing semiconductors is a very expensive and complex process that would not be possible without the unique properties of PFAS, making them essential to achieving the EU’s strategic objectives today – whether the European Green Deal or digital autonomy – and in the future. Therefore, we must ensure that essential uses will enable the continued use of PFAS in semiconductor manufacturing.The SEMI presentation delivered at the Chemical Watch event can be accessed here.Emir Demircan is director of Public Policy and Advocacy at SEMI Europe.[1] “The EU will also support the development of key enabling technologies that are strategically important for Europe’s industrial future. These include robotics, microelectronics, high-performance computing and data cloud infrastructure, blockchain, quantum technologies, photonics [etc.]”[2] Emerging technologies in electronic components and systems (ECS) Opportunities Ahead – A study by DECISION, 2018 for the European Commission[3] Available here[4] Regulation (EU) No 517/2014, “F-Gas Regulation”[5] SIA Nathan Associates, 2016, https://www.semiconductors.org/wp-content/uploads/2018/06/SIA-Beyond-Borders-Report-FINAL-June-7.pdf

While Artificial Intelligence (AI) emerged in the 1950s, only in recent years have AI applications proliferated with the explosion of data and continuing improvements in Moore’s law that have driven rising processing speeds. Voice assistants, image analysis software, search engines, and speech and facial recognition systems were among the first applications to use AI. Today, adoption has spread to sectors such as agriculture, cybersecurity, healthcare, software development, e-government and the intelligent enterprise to generate jobs and help spur economic growth. The Edge AI Opportunity and the Microelectronics IndustryAI can be embedded in hardware devices such as advanced robots, autonomous cars, drones or Internet of Things (IoT) applications. Today, according to the EU’s digital strategy, data centres and other centralized computing facilities account for the vast majority – 80% – of AI data processing and analysis, with smart connected objects such as automobiles, home appliances and manufacturing robots that bring the compute function closer to the user representing 20%. The latter, known as Edge AI applications, are powered by edge-based machine learning chipsets, not the AI chipsets designed to run cloud-based machine learning algorithms.The EU’s white paper on AI published in February 2020 anticipates that the way data are stored and processed for AI applications will change significantly over the coming five years as edge computing applications proliferate. Most AI applications need to connect with devices that collect data and manage data flows. When the applications connect with cloud infrastructures to train large volumes of data for a machine learning model, the interface devices often require hardware support. Edge AI can minimize data transport by processing data directly from local devices to accelerate data analysis and decision-making and make data transport or accelerator hardware unnecessary, critical in reducing power consumption and enhancing data security for applications such as autonomous driving. Over the past 40 years, the ICT sector has been continuously increasing greenhouse gas (GHG) emissions despite efforts to shift to renewable energy. Cloud-based AI applications require an ICT infrastructure for high-performance computing and high-speed connectivity. According to MIT Technology Review, data centres’ AI workloads could account for a tenth of the world’s electricity usage by 2025. a mass update of cloud-based AI applications may significantly increase energy consumption, unlike with Edge AI. This is why the strategy for developing Edge AI is well-aligned with the EU’s Green Deal objectives. Europe aspires to play a leadership role in Edge AI to strengthen the sector’s competitiveness and protect the European digital sovereignty. Europe’s strong industrial competencies in embedded systems and microcontrollers will help the region promote development of European domestic AI solutions for emerging high-value IoT applications in industrial processes such as Industry 4.0, Connected and Automated driving (CSA), smart cities, climate action, healthcare, and national defence and security. With this strong strategic position in technology, Europe is well-positioned to invest to become the leader in the Edge AI global market.Preparing the Workforce for the Microelectronics IndustryTo design and manufacture leading Edge AI chipsets, European education providers and industry will need to work closely together to train the current and future workforces. Within the framework of the METIS project, a four-year project co-funded by the European Commission through the Erasmus+ programme, SEMI and imec deployed experts in the field to survey and interview focus groups. The survey identified the following key focus areas for workforce development: 1. True Capability of AI and Data Science With AI’s heavy dependence on data, the workforce of the future must be trained in areas of data science including data integrity to ensure quality, unbiased sourcing, collection and accurate analysis necessary to interpret huge volumes of data. Europe also needs to train the next generation of AI chip designers in data security and privacy – key challenges to the widespread deployment of Edge AI chips. 2. Climate Change, Sustainable Development Goals (SDGs) and Social Inclusion TrainingSince the industry must be able to develop Edge AI solutions to enable the digital transformation while limiting GHG emissions, microelectronics engineers need to be schooled in climate change and understand how their work contributes to meeting the United Nation’s Sustainable Development Goals (SDGs). Workplace diversity and social inclusion are also important target areas for education since Edge AI applications should serve various groups of people with different needs.3. EthicsChip industry workers must also be educated in ethical issues of AI related to the technology’s potential societal impact in the near future[1]. With AI applications capable of monitoring Internet searches based on users’ personal preferences and biases to deliver tailored advertising, news and other information, developers must recognize how the technology can influence thinking and behaviour of individuals and groups. This awareness can help developers strike a balance between supporting commercial interests and societal good so the microelectronics industry can ensure ethical implementation of AI. 4. Cross-disciplinary Skills Required for AIAI development requires a comprehensive, cross-disciplinary skill-set to be able to integrate the work of specialists from diverse educational, cultural and professional backgrounds critical to developing non-biased AI solutions. For example, in addition to technical expertise, microelectronics AI developers must be able to communicate clearly and work in close-knit teams with non-technical experts from business, law, medicine and the social sciences.What’s Next?The microelectronics industry has a tremendous opportunity to develop new chip-based solutions for AI architectures, and apply AI techniques to improve operational efficiencies of design and manufacturing. To seize this opportunity, the industry must work closely with education providers to groom the next generation of skilled workers. This tight collaboration is critical to designing and delivering specialised courses to college and university students as well as engineers now working in the chip sector. The stakes are high. By preparing workers to develop Edge AI chipsets, the microelectronics industry can help the world confront some of the greatest challenges it faces today.For more information, see SEMI Responds to European Commission White Paper on Artificial Intelligence.METIS is a Sector Skills Alliance project co-funded by the European Commission’s Erasmus+ Program and coordinated by SEMI. The four year project, launched in November 2019, will develop a Microelectronics Skills Strategy. Based on the strategy, the METIS project will design 43 training modules for 1,100 hours learning in four key areas of the microelectronics sector.We thank Patrick Blouet (STMicroelectronics) and Jeroen Geusens (imec) for their valuable contributions to this article.[1] Ethics of Artificial Intelligence and Robotics, Stanford Encyclopedia of PhilosophyDr. Yanying Li is senior manager of Collaborative Projects at SEMI Europe.Dr. Pushkar P. Apte is the strategic technology advisor for the Smart Data AI Initiative at SEMI

D-SIMLAB Technologies, a Singapore-based provider of simulation-based business analytics and optimisation software solutions, recently joined SEMI. I spoke with Peter Lendermann, the company’s co-founder and Chief Business Development Officer, about the company’s role in the smart manufacturing movement, how customers are benefiting from D-SIMLAB solutions, and what the future holds for smart manufacturing. Ng: What is D-SIMLAB’s mission?Lendermann: Our mission is to develop, market, and deliver high-performance simulation-based decision support solutions that enable corporations to enhance their performance in a sustainable manner leading to significant cost savings. In particular, we focus on semiconductor manufacturing material flow planning and optimisation but also do business in aviation where we help customers optimise their spare parts support operations. What these two domains have in common are three important attributes: They are capital intensive, their underlying operations are complex, and operations are also heavily affected by random, i.e. unpredictable events, which makes both planning and execution of manufacturing operations very challenging. D-SIMLAB is a spin-off from the Singapore Institute of Manufacturing Technology (SIMTech) under the Agency for Science Technology and Research (A*STAR). Our head office is in Silicon Island Singapore. We also have representations in Germany and the U.S. Most of our staff are industrial and computer engineers with up to 20 years of operations experience in their respective industry domain, as well as vast data analytics and software development capability.Ng: What solutions does D-SIMLAB offer to optimise semiconductor manufacturing?Lendermann: In the three-pillar smart manufacturing framework of Connect, Sense and Predict advocated by SEMI, our focus is on Predict though we emphasise the equal importance of the subsequent Act: Our solutions can Predict, for example, WIP waves or usage-based preventive maintenance due dates. But much more value-add can be realised once some decisions with regard to how to Act can be derived from such a prediction. The ability to pro-actively adjust action plans in a timely manner is essential to overcoming challenges arising from changing customer due dates, mix profile changes, untimely production line issues, and production capacity to be shared with R D lots effectively, so that ultimately our customers can enhance capacity, reduce cycle times and improve the due-date performance of their factories.To that end, our D-SIMCON solution suite spans the full spectrum of decision-support tools required to forecast, manage and optimise material flow – from operational scheduling and dispatching, WIP forecasting and dynamic and static capacity planning all the way to specific applications for fab load mix optimisation or for the enhancement of the product/layer dedication and resist allocation in the lithography area. Our solutions are implemented in numerous 6-, 8- and 12-inch wafer fabs operated by both IDMs and foundries worldwide with capacity ranging from 40,000 to 200,000 wafers per month.Ng: What are the key enablers of D-SIMLAB’s success?Lendermann: Our success lies in deploying production-ready solutions for our customers, allowing them to extract immediate value. Our solutions enable the portrayal of many domain-specific characteristics such as queue time constraints or specific equipment behaviour, which is absolutely essential to generating operationally feasible plans or schedules in order to be able to Act in the best possible manner according to what has been Predicted. Moreover, we have modules for automatic generation, calibration and maintenance of the underlying capacity model, including resolution of data inconsistencies as well as verification and validation of the model, to allow near real-time responses to continuously changing operations. And the associated optimisation approaches focus on creating maximum possible value with as few iterations as possible and within minimum time through smart heuristics and parallel computing infrastructure – a paradigm that is as powerful as it is cost-effective.Ng: What are a few of your more notable customer successes?Lendermann: As a result of the first implementation of our novel, multi-objective based Scheduler cum Dispatcher, a tool capacity gain of 8%, a transportation capacity gain of 10%, and an operator workload reduction of 25% were concurrently realised at one of the critical equipment groups in our customer’s fab. At another set of equipment groups in the same fab, a 7% increase of lots within the critical queue time limiting area was achieved.Another use case we successfully realised is fine-tuning of Preventive Maintenance plans: Based on a seven-day lot arrival forecast at each equipment generated with our WIP Forecaster, a recommendation is made when PM would be best possible without causing too much disruption in the WIP flow. The effect of this synchronisation of the PM plan with material flow enabled a dramatic reduction of the average queue lengths at critical equipment groups and the associated cycle times without incurring any capacity loss. Reduction of average queue length as a result of synchronising preventive maintenance with material flow. Ng: What challenges has D-SIMLAB been facing in the COVID-19 world?Lendermann: Obviously, software delivery projects have become more challenging for the time being since our engineers cannot be on-site frequently. But it also turned out that more and more services can be delivered remotely, which has the nice side effect of making the services more cost-effective for customers. Overall, we are confident that our solid customer base will enable us to sail steadily through these challenging times.Ng: Where does D-SIMLAB see the technological development heading?Lendermann: In the future, enriching decision support and manufacturing execution solutions with machine learning and other AI techniques will be critical in reducing dependency on human experience. This path is essential to making manufacturing operations fully Industry 4.0-compliant. D-SIMLAB will certainly be at the forefront of this development. Bee Bee Ng is president of SEMI Southeast Asia.