CEO Interview: Doluca preps Maxim for IoT era
Q1) How would you characterize the differences between Maxim under your predecessor Jack Gifford and under yourself?
Tunc Doluca: I worked with Jack Gifford for many years and respected him greatly. An effective teacher, he had passion, intellect and brilliant ideas. He drove people hard to deliver results, but he also rewarded them. I became CEO in 2007 and, although we have made several changes (such as centralizing our supply chain to better serve our customers and reorganizing to focus on end markets), we continue to execute on Maxim’s vision of creating highly integrated products. That started under Jack; it is not new for Maxim Integrated. In fact, we modified our company name in 2012 to underscore that focus and expertise. We accelerated the changes to adapt the company to our strategy.
Q1a) Is it a move away from analog building block products and towards mixed-signal and digital solutions including software?
TD: Maxim has built a leadership position in analog and mixed-signal integration. We continue to grow operational excellence and revenue in areas such as multichip system-in-packages and fully integrated mixed-signal SoCs. While highly integrated products, which now comprise about half of our revenues, are our differentiator, we continue to create great building-block products. That will never change. Over time, we have built a rich portfolio, which grows to this day. These ICs showcase our innovation, are successful in the market, and help our customers differentiate their systems. Ultimately, many of our standalone products will emerge as candidates to be integrated with others onto a chip. However, some of these products will remain clever, best-in-class discrete components.
Q2) Is it more application and market oriented? In which case what are Maxim’s markets and what characterizes them in terms of volumes, average selling prices, etc.?
TD: In 2011, we reorganized to focus on end markets so, yes, I would say we are market-oriented. Driven by demand for smartphones, mid-range phones and tablets, our mobility business grew the fastest. It is currently the largest segment in terms of revenues. As smaller form factors and power efficiency become more critical, we target other applications that can benefit from our expertise in analog integration. These include smart meters, automotive infotainment and portable medical equipment. However, we value and target the broad market, too, primarily small and medium-sized customers in the industrial space. Our business model effectively balances high-growth mobility with the stability and long term annuity of the industrial and communications markets.
Q3) Where do you manufacture and what is the strategy?
TD: Several years ago, we centralized our supply chain operations. That enabled us to better serve our customers and realize efficiencies. To provide flexibility and additional capacity, our hybrid manufacturing model has served us well. Our internal wafer fabs are located in the United States: San Jose, Calif.; San Antonio, Texas, and Beaverton, Oregon. Our test and assembly sites are in the Philippines and Thailand. We engage foundry partners to augment our internal facilities, and about half of our products are made at partners’ fabs using Maxim’s unique process technologies. These are true partnerships, not simple outsourcing.
Q3a) I think under Jack Gifford a number of mature manufacturing facilities were acquired right up until 2007. Under yourself none! Are you becoming more fab-lite over time?
TD: Keeping some manufacturing in-house differentiates us from our fabless competitors because in analog, there are specialty technology requirements. We designed our hybrid manufacturing model – i.e. using a blend of internal fabs and foundry partnerships – to flex according to demand and market needs. It is working very well for us; I do not anticipate it changing.
Q3b) As the hybrid (fab-lite) manufacturing model is not changing, does that mean Maxim would consider acquiring or creating more internal manufacturing capacity to maintain the proportional balance with foundry-sourced wafers as Maxim sales expand?
TD: At this time, we are satisfied with our internal capacity and have no current plans to add new fabs. Our primary effort in current internal fabs is to enable specialized technologies such as high-voltage, sensors and next-generation BCD rather than expand capacity.
Q3c) You have a proprietary process technology at 0.18-micron but what about mixed-signal at 90nm, 65nm, 28nm. Do you rely on foundries for that?
While we utilize our foundry partners to manufacture products on some of the smaller-geometries, they are using Maxim’s proprietary process technologies. As a result, we continue to heavily invest in this area. Obviously, we do not publicly discuss “which specific process technologies are where.” However, I can tell you we have qualified and are ramping several leading-edge analog processes in our own fabs. Several product lines require deep sub-micron technologies and for those we utilize merchant foundry process at and below 65nm.
Q3d) What are the analog processes that Maxim is ramping in its own fabs? Anything at 130nm, 90nm? Special processes for optical or other?
TD: We are currently running 180nm BCD as the advanced node that is being ramped.
Q4) What is the role of MEMS and other sensors at Maxim? Do you have all the building blocks for wireless sensor networks and the Internet of Things?
TD: Sensors are a critical adjacent function to analog semiconductors, and the sensor market is expected to grow rapidly as new applications emerge. Through both internal development and acquisitions, sensors and MEMS have an impactful role at Maxim.
While our gesture sensor in Samsung’s Galaxy S4 received considerable attention, we are actually investing, developing and manufacturing sensors for numerous applications and markets. In fact, very few semiconductor companies have touch, motion/MEMS and optical sensor technologies in their portfolios. We have all three. It is possible to fuse various types of sensors with our analog technology, enabling all sorts of applications, including the Internet of Things and wearables.
Does Maxim make inertial MEMS internally and if so where?
TD: Yes, we have successfully qualified MEMS processes in our internal fabs. However, for proprietary reasons, we choose not disclose at which facilities we are making these devices.
Q5) You have made a number of acquisitions in recent years including Phyworks, SensorDynamics and Genasic in Europe. Will the acquisition pace decrease, increase and is the strategy for Maxim to be a "consolidator" to achieve critical mass? Is there a goal in terms of annual revenue or in terms of the global ranking of chip vendors (top 20, top 10?)
TD: Maxim has been an acquisitive company since 2007. A majority of our acquisitions have been technology tuck-ins, providing us with highly valued talent and IP. We have made some sizeable acquisitions as well, including Teridian (energy ICs), SensorDynamics (MEMS) and Volterra (integrated power management). In each case, these companies have shown a potential for growth and have technology that fits well with our own. What is our goal for annual revenue growth? We do not forecast revenue beyond the current quarter. However, my general goal for our company is to continue growing faster than our analog peers while maintaining solid profitability.
Q5a) Such consolidation implies that consolidators will increasingly address multiple markets and look alike. How is Maxim differentiated from other consolidators?
TD: How is Maxim differentiated? We create clever, highly innovative building-block products, are an industry leader in analog integration, and are rapidly ramping our sensor capabilities. We have a vast product portfolio, a centralized supply chain, deep customer relationships and valued distributor partners helping us serve customers in the broad market. I am proud of our employees and our success. Although I am pleased about Maxim’s position in the market, this is a very competitive industry. We cannot sit still so must keep innovating and adapting.
Q6) How will the end of Moore’s Law disrupt the semiconductor industry, and affect Maxim? What technical revolutions are possible?
TD: Maxim is, in essence, an analog company. For now, Moore’s Law actually enables us to build better products by developing better digital circuits to support the analog. Process geometries continue to shrink for analog and mixed-signal products. The end of Moore’s Law will enable a vast amount of applications and markets about which we cannot yet grasp. However, I will gladly place my bets on a company with a broad analog portfolio, sensor capabilities and, of course, analog integration expertise.
Q7) How do you view the rise of southeast Asia? Must development tend to follow manufacturing to the eastern hemisphere? Are the European Commission’s plans to increase chip manufacturing in Europe a dream?
TD: For Maxim, development, whether product or process technology, will occur wherever there is superior engineering talent – the U.S., Europe or Asia. For example, we have talented engineers in design centers all over the world. In some cases, they are near our manufacturing operations. In some instances, they are close to our major customers. In others, they are in towns that have extraordinary engineering talent and/or are great places to live. It is difficult for me to predict the results of the European Commission’s plans as they affect chip companies. We currently do not manufacture in Europe and do not have current plans to start.
Related links and articles:
News articles:
Maxim forms venture capital group
Maxim acquires Edinburgh mixed-signal design house
If you enjoyed this article, you will like the following ones: don't miss them by subscribing to :
