Skip to main content
Is there a way that we can acquire innovativeness? How can we keep providing innovative solutions? Dr. Melissa A. Schilling, a professor of innovation management strategy at New York University had a talk session with Dr. Umeshwar Dayal, General Manager of Global Center for Social Innovation, Hitachi, Ltd., who is the expert in data management, analytics and AI. The interview focused on the meaning of open-innovation and how Hitachi is going to keep providing innovative solutions to the social problems.
Schilling: My name is Melissa Schilling,and I’m a professor of management at New York University Stern School of Business. My research area is technological innovation and I teach classes on social innovation strategy. I have been studying serial breakthrough innovators such as Elon Musk, Marie Curie or Steve Jobbs, people who have introduced disruptive breakthrough innovations repeatedly. When you look at serial breakthrough innovators, you can extract elements they commonly have, and studying about them helps us understand keys to create innovative ideas.
Professor of Management at New York University Lenard N. Stern School of Business. She received her PhD in strategic management from the University of Washington. Her research focuses on innovation and strategy in high technology industries such as smartphones, video games, pharmaceuticals, biotechnology, electric vehicles, and renewable energies. Her textbook” Strategic Management of Technological Innovation”, is the number one innovation strategy text in the world. In Japan, she is known as the author of “Quirky”, which analyzes the commonalities of innovators.
Dayal: I am Umeshwar Dayal and am leading Hitachi Global Center for Social Innovation (CSI), which is one of the three Hitachi Research & Development (R&D) initiative bases alongside with Center for Technology and Innovation (CTI) and Center for Exploratory Research (CER). I have over 35 years of experience in data management and analytics and am currently responsible for open innovation with customers and partners worldwide.
Please let me talk a little more about Hitachi R&D activities. As you can imagine, we originally followed the traditional approach to technology innovation that most industrial research labs follow: we hired a lot of smart people, let them invent a lot of interesting technology artifacts, and then we tried to find good uses of these artifacts. A few years ago, we realized that to create innovative solutions, we had to start with understanding what disruptions are occurring in the various industries in which we play, what our competitors are doing, and what challenges
customers are facing. We really needed to change the way we innovate.
So, while keeping the traditional research and futuristic technology research hubs as CTI and CER, we established CSI. At CSI, we use the NEXPERIENCE methodology to engage with customers and understand their challenges, then we co-create with customers to build innovative solutions that meet those challenges.
General Manager of Global Center for Social Innovation, Hitachi, Ltd., and Senior Vice President and Senior Fellow (Information Research) at Hitachi America, Ltd.
In this role, he is responsible for open co-innovation worldwide, leading to the creation of novel AI and analytics solutions that deliver social, environmental, and economic value in various industries, including manufacturing, energy, natural resources, healthcare, financial services and mobility. Prior to joining Hitachi America, Ltd., he was an HP Fellow and Director of the Information Analytics Lab at Hewlett-Packard Labs, Palo Alto, California, a senior technical consultant at DEC's Cambridge Research Lab, Chief Scientist at Xerox Advanced Information Technology and Computer Corporation of America, and on the faculty at the University of Texas-Austin. He received his PhD in Applied Mathematics from Harvard University, and received the Edgar F. Codd Award from the ACM Special Interest Group on Management of Data (SIGMOD) for fundamental contributions to data management.
In May 2019, our President & CEO Higashihara announced Hitachi’s 2021 Mid-term Management Plan, committing to improving customers’ three values; Social Value, Environmental Value, and Economic Value. Social Value helps solve problems that a society challenges, whether these are in mobility, healthcare or in manufacturing operations area and so on. But we also provide Economic Value to ourselves and our customers, and lastly, we help to improve customers’ Environmental Value, because obviously we think this is important.
Schilling: I think it is great that Hitachi really focuses on social problems and looking at what customers need opposed to what you can do better. We call that science-push vs demand-pull. Science push is valuable because as an inventor you see opportunities, but demand-pull focuses on realizing what solutions we really need to solve. There is no point in chasing down a bunch of incremental improvements down a path that is easy if you are not really addressing a fundamental human need. So, it is great that Hitachi really focuses on that.
If I quickly talk about the word root “innovation,” it actually comes from the Latin word “innovāre,” which is basically to create new things from new ideas so it's a very broad word.
Talk Session was held remotely between Los Angeles and Tokyo.
And I would like to clarify that for me technological innovation isn't just the shiny things with microprocessors inside phones or computers but in my field, innovation is any process by which an input becomes an output. So really cooking is a sort of innovation. What I'm studying is innovation in the way we do things to create, to take new ideas and to apply them into something new for useful purpose.
Dayal: Yes, innovation has a lot of meanings. When we think of innovation at our labs, we have developed a methodology called “NEXPERIENCE,” which visualizes insights and concerns of customers and partners to smoothly collaborate with them. We start with this sharing of a vision that leads to creating a concept of how we might solve or address some of these challenges, the kinds of solutions we might be able to create.
Schilling: I hear you are the expert in data science, but could you please talk a little more about how you joined Hitachi?
Dayal: Starting my career in Massachusetts doing industrial research, I did research on everything having to do with data, and I was always very interested in the impact of the work I was doing. Looking at the industry trend in 70s, 80s or 90s, a lot of excitement was around data management and we were always questioning ourselves how we could help enterprises collect data from their own business operations and make that available to decision makers. Later, the emphasis shifted to business intelligence, which is the use of analytics to derive useful insights from data, enabling the improvement of business operations. When I ran data and information analytics lab at HP, Hitachi reached out to me to create a Big Data Laboratory in the Silicon Valley. Given Hitachi’s strength as an industrial company, this gave me the charter to focus on using data analytics and AI to create innovative industrial solutions.
Schilling: Oh, that is probably the same time when I just started research in technology and found fascinating to know the fact that although cool technology seems to take a long time to come but suddenly appears without any sign.
For instance, back in 2000, I had all these students in my class writing cases about home automation technology which is the precursor to IoT. And as you imagine, we saw autonomy as not coming for a while until 2030. But today we see many car manufacturers that have got a lot of autonomous features. Technology just really sneaks up on us.
Dayal: Yes, that is a very good point, for instance, people have thought about flying cars.
It takes a lot of time, not necessarily the technology that's inhibiting us, but a lot of the other issues surrounding it, such as regulations. We have seen this with drones for instance. Many companies are ready to do package delivery with drones, but legislative regulations and policies just don't exist. So sometimes it's that technology might be there, sneaks up on us, but the social acceptance or regulations are not ready yet and we have not really figured out the best use of those technologies.
Schilling: The example I thought was interesting is home automation. As the technology was basically there, there was nobody who was doing it. The reason I think of was the ease of use and if installation is too complex, nobody would not do it.
Dayal：That is an important point. Please let me pick up one example from my experience. We were trying to build what we call a physicians’ workstation, bringing all the data a physician would need to look at, making them available on a single dashboard. Data such as patient vital signs, past history, and medical records, were available. However, as the system was made by software developers and technical engineers and we crammed all the data onto the single dashboard, making it so complicated that the physicians rejected it. So, we brought in anthropologists and had them interview physicians, trying to understand when they needed to see what kind of information and then improved the user interface, and now physicians just loved it. It is a small story, but I think that is true in a lot of other areas. I think there is always a barrier to acceptance due to how the technology is presented and this leads to our earlier discussion of science-push vs demand-pull. Another point is that often, the pieces of great technology exist, but how to bring in all these together to create innovative solutions is always a challenge.
Schilling: While great technologies have drastically changed our lifestyle more conveniently, some people fear that technology will eventually replace human workers as it progresses. However, humans have a capacity for insight and intuition that we're a long way from building that into a computer. But you can harness AI what humans lack is really great memory and really fast processing capability. Computers have both of those things, so the big next step is harnessing computers to improve peoples’ workstyle. What do you think about that?
Dayal: Fully agree. As you know, we have an aging society and many of our customers and we ourselves face with the issue that as the workforce ages, experienced workers with a lot of specialized knowledge and expertise will be retiring, so we need to find a way to take over their knowledge to next generation. Does it mean that we use AI to replace those workers? I don't think we're at that stage yet. But we can use the knowledge of AI to observe what the workers do and learn from them so that we can build a system that claim the novice title of new workers but help to improve these processes as well. My point is that we don't need to wait until self-learning autonomous systems are complete, and still now there is a lot of value that can be and is provided from them.
Please let me touch on another example in healthcare. We should look at not just this patient’s history origins and gene record, but also the history of the patients so similar to them. And that is very difficult for one doctor because their experience is based on the patients they have seen before or what they learned from textbook. But if you could gather information about similar patients living in various parts of the world that the doctors never visited, we can take all of that and learn from that, then we can provide much better diagnostics and therapy for patients. We've done this experiment in our labs. So, here is how it works. When patients come into hospitals, they can now look at their history, and learn from these and choose the best right hospitals or treatment. One of the most critical things we're trying to do is to understand which of these patients may need to be admitted to the ICU because ICU needs to be open for those who are in urgent need of special treatment. So, we could actually build successful predictive models to understand which patient could go to the ICU.
Again, they're not replacing the human, they are augmenting the human because now we can look at a lot more data, process that data, and particularly data from wider areas, a coverage that the individual human cannot access and process, especially process in a timely fashion.
So to me, that's where the excitement is these days, how do we build these very smart systems, I guess they used to call them “the support systems” but that's kind of an old-fashioned term, but really much smarter assistants that are informed through data and these learning processes.
Schilling: Technology depends on how you accept and use it.
Dayal: That is exactly right. Another example, as you know, food problem of the world is becoming a critical social issue. If you are in a sort of developed societies, a lot of farming processes have already been seen mechanized. But if you look at the vast number of farms in the rest of the world, they're really family farms and have been practicing the same kind of techniques for centuries, and new technology just hasn't reached them. So, we became farmers. We bought land and said “let's put sensors,” and we helped the farmers understand both through agronomic models, trying to understand what crops they should be growing. This is because they grow the same crops over and over again, and if there's a bad yield, they will end up with wiping out all the earnings of that season or that year. So, we did a Proof of Concept building crop models, fertilization models, and irrigation models, and all of these are data-driven models, leading to help them understand how to access the wider market to be more profitable.
Schilling: Really Interesting.
I write a lot lately about clean meat or lab meat and plant-based meats. Because a few years ago I started reading about the impending meat crisis that we're supposedly facing. The discussion was, as more developing countries grow, meat consumption in those areas will increase as people add more meat to their daily diet. But we can't grow meat stocks that fast and if we did, it would be like a planetary disaster because cows and pork produce as much gas emissions as cars in the world do. That would be a terrible crisis as well so I'm getting really interested in this lab meat discussion.
However, the interesting thing is when I started writing stuff about or tweeting about this, you get a really angry backlash from people who would like to defend the traditional farmers. This is no longer a technology feasibility discussion but more a cultural or a philosophical problem to solve.
So, as you said earlier, for the new technology be accepted to a society, people’s mindset or common sense of social mainstream need to change not just the realization of the technology itself.
Dayal: Yes exactly. So, this goes back to the point we were discussing earlier, sometimes it's not the technology -yes, the breakthrough is required for some of these that will enable rapid development, but how do we address these kinds of social issues?
In our lab at Santa Clara, we bring in members of the community and so on to try to help us understand what the values to new technology are or what would actually enable the adoption of it or even just what the future looks like. As a company, we should be listening to those who oppose while seeking possibility. There is always a naysayer, but we have to make a progress and that is really an exciting moment.
Release Date：December 2019
Production：NewsPicks Brand Design