INTRODUCTION / SITUATIONAL ANALYSIS
This is the fifth of the eight-part episode tutorial on the topic MEMS and Sensors Marketing, Oxymoron or Opportunity. The topic of this episode is putting the “s” back into MEMS and the challenges of doing so. Welcome back to those who have been following this bi-weekly tutorial series and a cordial welcome to those who are joining us for the first time. Previous episodes included:
- Episode 1: Listening to the Voice of the Customer
- Episode 2: The Role of Marketing in the Funding Process
- Episode 3: Is the “Shine” off of MEMS
- Episode 4: Insider Tips for MEMS/Sensor Marketers
It is interesting to consider that this tutorial series focuses on the theme of ‘oxymoron’ from a marketing perspective. Additionally, and for this episode, the topic of micro-electro-mechanical systems (MEMS) itself is truly another oxymoron or misnomer from several perspectives:
- MEMS as we have gotten to know them, have essentially been unique devices providing a singular functionality and certainly are not a system
- MEMS, while they can, and frequently do, include a mechanical function, can also be realized in other cross-domains including electro-chemical, electro-photonic, micro fluidic as well as actuators and structural elements which are not sensors at all.
Thus, we have a situation that could be considered as an oxymoron squared.
When I joined the MEMS community approximately 38 years ago as a marketing manager at Foxboro ICT in San Jose, California, we referred to these as integrated circuit sensors. Several years later as a consultant and member of the founding team of Nova Sensor, we referred to these devices as micromachined sensors.
At Nova Sensor, we successfully positioned the organization as the leader of this technology. To this day, there exists an annual international conference called Micromachine Summit, on whose organizing committee I am a member. The MEMS acronym, I believe, was created by Kaigman (Ken) Gabriel at an IEEE technical conference in Cape Cod, Massachusetts over 25 years ago, and the name has stuck to this day.
With the many responses obtained of the various topics from the approximately 20 interviewees, the one issue that received the most attention and corroboration was that there exists a lack of a systems/solutions approach to marketing of MEMS/sensors. It was quite clear, that to be optimally successful, a sensor/MEMS supplier cannot only offer a device/component, but rather needs to provide a systems/solutions approach that best fits customer needs and adds value to the offering.
Prof. Kensall Wise, one of the pioneers of MEMS, created the Wireless Integrated Micro Systems (WIMS) Center at the University of Michigan in 1990s. Ken and his researchers’ goal was to develop and integrate various elements of a sensor-based solution for many different applications including gas chromatography optical instruments. The development of low-power CMOS and energy harvesting functions was part of a 10-year National Science Foundation (NSF) Engineering Research Center (ERC) grant.
The center exists to this day with an updated name: Wireless Integrated Micrsensors and Systems (WIMS2). It currently has 40 faculty, 124 researchers and has spun off 15 companies, created 1796 publications and has been granted 86 patents. I was an early adopter of this concept after I was asked to provide a keynote presentation at a conference in Manaus Brazil and had the opportunity to meet and discuss this concept with Prof. Thomas Gessner of Fraunhofer ENAS in Chemnitz Germany. His Smart Systems Integration approach made great sense to me and from that point forward I became a disciple and began aggressively proselyting this concept in the early 2000’s. I serve on the organizing committee and regularly present papers at a European Conference which bears the same name and which meets annually in the Spring at different locations in Europe. Additionally, I authored the first of several articles  on this topic as early as 2008. I named this concept MEMS-Based System Solutions (MBSS) and it forces the design engineer to think outside the chip and to listen to the voice of the customer as we discussed in Episode 1.
I believe it is safe to say that Prof. Gessner and Prof. Wise were truly visionaries and certainly evangelists of this systems integration concept. Figure 1 illustrates this concept.
My definition of MBSS includes the following functional elements:
- Sensor (or actuator, structure) front end
- Signal conditioning
- Power and power management
- Communications…either wired or wireless
- Most importantly, packaging/integration and interconnects.
This last item is critical to the creation of a successful solution since it typically represents up to 65% of the total cost . These are all supported by strong systems engineering principals.
Rationale for the adoption of this systems approach include:
- Provides the supplier with a highly differentiated product…especially required in a mature/commodity market which has significant price pressures and thus provides a competitive advantage in the market
- Overcomes the problem that many customers have with limited resources in internal system engineering resources and require a complete solution from the supplier which minimizes interface issues…think of it as a general contractor for remodeling your home versus hiring individual workmen…e.g. plumbers, carpenters, electricians
- Provides a higher value added to the supplier’s offering, and, hopefully, a higher profit margin
However, if a provider of solutions is to be successful, it must have the broad range of resources to help a system engineer to create a solution with a very strong and capable applications engineering team in the front end of the sales process to properly engage with the customer. Regrettably, these requirements eliminate many MEMS/sensors companies from engaging in the game since they do not have the necessary resources to permit them to do so successfully.
In my opinion, most of the producers of MEMS have only recently excised themselves from the non-systems approach starting with the addition of an application specific integrated circuit (ASIC) that provided signal conditioning and compensation. Amongst the functionalities included in the ASIC were A/D conversion, filtering, amplification, embedded processing including proprietary algorithms.
I think that it is safe to say that most sensors sold today include an ASIC and primarily as a heterogeneous circuit element. Several companies including MEMSIC (accelerometers) and early Analog Devices accelerometer designs used a monolithic approach. ADI has moved away from this strategy many years ago and now uses a discrete ASIC in their accelerometers.
At the cost of being redundant, I will provide Rob O’Reilly’s (Analog Devices) input to the last episode here since, I believe he was spot on in addressing the issue. He said, “As sensor component suppliers begin to move up the chain, the approach to marketing both system and IoT solutions will need to adapt to the changing playing field. Winning the socket is no longer in style and initial conversation with potential customers have become more complex. Discussions beyond the simple sensor have expanded to include; power management, edge computing, gateways, micro controllers, radios, cloud and connectivity. Working knowledge of the system is more of a requirement during initial system/IoT discussions for both existing and potential customers. With the expansion of tools, cloud partners, sensor clusters and applications, marketing is becoming a team event. As the market expands the approach to marketing must also expand to keep up with the diversity, depth, complexity and constant challenges in applying sensors to the digital world. Yes, it will take a village.”
The following are some of the more poignant of the responses on this topic from the 20 interviewees:
Juan Figueroa (NSF) said, “I saw a major shift in the focus of proposals submitted to the NSF (National Science Foundation) SBIR (Small Business Innovative Research Program) during the 2002-2007 time frame from MEMS components to systems and applications solutions. MEMS appeared to have morphed into sensors and MEMS have virtually disappeared from the vocabulary of the submitted proposals. I believe that this was due to an incorrect view of MEMS and that the use of the term MEMS only complicated the message.”
From the perspective of Janusz Bryzek (eXo Systems), “to be successful, you need to sell the information obtained from the sensor-based application and thus need to include functions including data analytics and thus move up the food chain, the sensor needs to be smart.
Sandeep AkKaraju (eXo Imaging) said, “MEMS suppliers are increasingly offering their customers systems and application solutions approaches versus components which only sell features. Successful suppliers provide unique solutions and do not compete on price. The price will be determined by the degree to which it provides the necessary solutions, that is, provides actionable information.”
Steve Whalley (Whalley Consulting) said, “most sensors/MEMS are now viewed as commodities and marketers have not been successful in maximizing the value of their offerings. They need to move up the food chain and add sensor hubs/sensor fusion, data analytics and deep learning software to their solutions to be able to successfully compete. As sensor use grows dramatically in IOT products and many others, so does the data generated. We can utilize that data to improve the overall application solution by fine tuning the sensor algorithms and the sensors themselves. The rate of growth though is surpassing, or in some cases has already surpassed, how quickly engineers can turn this data into useful insights and keep improving product offerings. Thankfully Machine Learning, Deep Learning and AI can do a much better job of identifying subtle patterns in massive data sets, without the need for traditional programming methodologies. This new tool is opening up new applications that were once thought unsolvable. If you are not using these techniques in your sensor based solutions, you probably should.”
Jim Knutti (Acuity) says, “It is important to remember that the real strength of MEMS technology is to provide a link between the physical world and electronic signal processing. The recent focus has been on MEMS Sensors. However, actuators, packaging, energy harvesting, and such other structures as mechanical amplifiers, microfluidics, and cell sorting critically enable the interface of low-cost signal processing with the physical world. A typical complete autonomous system relies on collecting information, powering signal processing to interpret the data, packaging the system, and responding. MEMS technology can provide the key interface at each of these steps and many opportunities for new systems.”
Gary Winzeler (DunAn Sensing ) said, “It is so easy to rely on the way it has been done in the past when it comes to the customer applications, testing and manufacturing. To compete today you really need to step back and forget the past and start with a new slate. We recently had a customer ask for a pressure transducer to measure refrigerant in a HVAC system. In the past, all one had to do was meet the detailed pressure needs and you’re done. But we decided to think outside the box and try a different approach, a system approach. We asked ourselves… why do you need to measure pressure for this application? The answer came back as it was required to measure superheat of the refrigerant and the calculation also requires knowing the temperature, and made ideally at the same location of measuring the pressure. What if we include a temperature sensor with the pressure transducer? That would save having to buy a temperature sensor plus the added expensed of wiring, a.k.a., cost of implementation and will be more accurate because of location. Okay, great design, but now we must manufacture the device and be competitive in pricing. Again, let’s step back and forget how it has been done in the past. We also needed to streamline the design to allow rapid testing and easy integration into any customer applications. As a result, we designed Dun An’s DURAsense Core which reduced mass making the test fixture easier to assemble (no threads) and require less testing time. It can now be fully tested and calibrated then fit into any housing/connector configuration required by the customer and at a significantly low cost of ownership to make the superheat measurement.”
Prof. Matteo Rinaldi (Northeastern University) said, “A key challenge posed by the Internet of Things revolution is how to power and maintain the rapidly increasing connected devices worldwide, expected to grow from 15 billion today to 50 to 200 billion by 2020. Supporting such a large network requires sensors and actuators with dimensions and power consumption significantly smaller than current versions. State-of-the-art sensors consume power continuously to monitor the environment, even when there is no relevant information present. Constant power consumption limits the sensor’s lifetime and causes high deployment and maintenance costs. Our Northeastern team recently broke this fundamental paradigm of wasting energy in standby with the invention and demonstration of a completely passive digitizing sensor microsystems that can detect and discriminate events of interest by exploiting only the energy contained in their specific physical signatures. This new device concept paves the way for the development of persistent, event-driven sensing capabilities, where physical, electromagnetic and other sensors can remain dormant, with zero power consumption, until awakened by specific physical signatures associated with events of interest. The capability of these zero-power sensors of consuming power only when useful information is present results in a nearly unlimited duration of operation, with a groundbreaking impact on the proliferation of the Internet of Things. The deployment of a myriad of such miniaturized zero-power sensors embedded in the environment, the city infrastructure, the buildings and the houses would enable real-time acquisition of information related to environmental stressors, security threats, mobility, traffic, indoor/outdoor air quality and pollution patterns resulting into the development of smart and connected cities, buildings, houses and transportation that make everyday life safer, simpler and more efficient enhancing the overall quality of life.”
Mark Laich (Laich Advisory Group) said, “The value of providing sensor solutions rather than components should be clear to every MEMS company. However, the challenge to our industry is in attaining compensation for the intellectual property which is a significant part of the investment which we make in our reference designs and other types of sensor solution deliverables. As noted by the panel of investors at the Fireside Chat during the 2017 MEMS & Sensor Executive Congress, MEMS companies too often make the mistake of giving away their system level intellectual property just to sell a MEMS component. Our customers are happy to obtain such cheap design knowledge for the price of a component. So, while providing MEMS system solutions help us gain competitive advantage, we need to develop more sophisticated revenue generating models for these solutions to amortize our investments in solution IP and provide better results to our shareholders. This takes a disciplined approach by our sales and marketing organizations to break from the traditions that our industry has set in the past.”
Rick Russell (Merit Medical) said, “As the market looks to IoT for the next big technology push, established industries, such as automotive, medical, and aerospace focus on system integration. Integration, however, seems to differ immensely from one industry to the next. For example, the medical industry, driven primarily by governing councils, tends to steer towards longevity, creating solutions that are not so cost sensitive, but application focused. The automotive market, to the contrary, tries to reduce cost by consolidating sensors and electronics. The airbag control module is a good example. It includes not only the complex airbag circuitry, but also the MEMS sensors for crash detection, rollover, and vehicle stability. This is a good example of sensor fusion. As this trend continues to grow, we see more and more companies looking to expand their capabilities in electronics. Sensors will become increasingly valued for how they integrate into a system not only for sensing capabilities but also for collecting and reporting data.”
In this episode, with a little help from my friends, I have attempted to demystify the oxymoron issues associated with MEMS/sensors marketing and the issue of their system aspect, or lack thereof. I believe that in both situations, the MEMS/sensors community is moving forward slowly to overcome these issues that have plagued it for many years.
It is interesting to know that Prof. Ken Wise at the Wireless Integrated Microsystems Center at the University of Michigan initiated the activity of systems integration in 2000 and he and his researchers have had exceptional success in their activities. Same is true with Prof. Thomas Gessner at Fraunhofer in Chemnitz. However, in my opinion, the MEMS/sensors industry has not vigorously embraced much of the successful outcomes of these two visionaries.
It is quite well known in the industry and frequently mentioned in presentations at industry events that organizations cannot survive by making pure components. It is necessary to move up the value chain and add functionality to the solution. Many industry experts are advocating that suppliers not only need to provide sensor-based hardware systems but additional services that are based on the information collected from these systems and sold to their customers.
A good example of this is in agriculture where providers install sensors and sell information and recommendations on irrigation and fertilization to the farmers to maximize yield and minimize cost and pollution. This, in my opinion, is the ultimate system solution. The distilled message here is that it is not about the sensor, but about the solution.
In Episode 6, we will address the topic of MEMS/sensors market research and its past and future ability, or lack thereof, to support the MEMS/sensors community with valuable information as well as the caveats and pitfalls for its effective use.
The author would like to acknowledge the contributions of the following individuals:
- Sandeep Akkaraju, President /eXo Imaging (Formerly: CMO Jyve)
- Robert Andosca, Ph.D. CEO / Inviza (Formerly: CEO microGen)
- Matt Apanius, President and Managing Director/ Smart Microsystems
- Janusz Bryzek, Ph.D., CEO/ eXo Imaging (Formerly: CEO Jyve)
- Juan Figueroa, Ph.D. CEO/Abenaki Connect (Formerly: SBIR Program Manager/National Science Foundation)
- Alyson Hartzell, Managing Engineer/ Veryst
- Brian Kinkade, Founder/Positive Impact (Formerly: Marketing and Sales VP, Spec Sensors)
- Jim Knutti, Ph.D., CEO/ Acuity Inc.
- Mark Laich, CEO/Laich Advisory Group, (Formerly: VP Sales and Marketing/ MEMSIC and V.P. Business Development /Qualtre)
- Keith Myers, V.P. Marketing/ TE Connectivity
- Tom Nguyen, CEO/ Dun An Sensing
- Rob O’Rielly, Consulting Engineer/ Analog Devices
- Rick Russell, President/Merit Medical
- Steve Ohr, Semiconductor Industry Analyst and Reporter (Formerly: Semiconductor and Sensors Analyst/Gartner)
- Kurt Petersen, Ph.D., Member/Silicon Valley Band of Angels (Formerly: CEO/SiTime)
- Paul Pickering, V.P. Business Development /Micralyne
- Swaminathan (Swami) Rajaraman, Ph.D., Assistant Professor/University of Central Florida
- Paul Werbaneth, Director of Marketing / Intervac
- Steve Whalley, Whalley Consulting (Formerly: MEMS and Sensors Industry Group/Chief Strategy Officer)
- Gary Winzeler, V.P. Sales and Marketing/ Dun An Sensing
 R. Grace, Thinking Outside the Chip: MEMS-based systems solutions, Small Times, Nov.-Dec. 2008, pp.25-29, www.rgrace.com
 R.Grace, M. Maher; Think Outside the Chip at the Package Level; Electronic Products; Nov. 2010 ; www.rgrace.com
About the author
Roger H. Grace is president of Roger Grace Associates (RGA), a Naples Florida-based marketing consulting firm specializing in high technology, which he founded in 1982. His background includes over 40 years in analog circuit design engineering, manufacturing engineering, application engineering, project management, product marketing, and technology consulting. He can be reached at 239-596-8738, [email protected].