Illustrating the Benefits of Systems Thinking and Early Stage Product Design

By Eric Smalley

Maria Yang

Much of the design and development of products focuses on engineering. But Assistant Professor Maria Yang believes that to get the most out of the product design process, it pays to linger over pencil and paper right at the beginning.

Research led by Yang shows that the design sketches generated during the earliest part of the design cycle can have a big impact on outcome. Yang, who teaches courses in MIT's System Design and Management (SDM) program, is on the faculty of both the Mechanical Engineering Department and the Engineering Systems Division. She also leads the Ideation Lab, which studies the processes and behaviors involved in early phases of product development.

The preliminary stage of design is challenging to study because of its ambiguous nature. Design concepts can change rapidly, and this evolution can be difficult to capture and assess. Yang studies these so-called "informal" representations, which can take the form of sketches, models, prototypes, or even words. She said that for many engineers and designers, the act of sketching itself is way of thinking.

"You'd be hard-pressed to find a designer or engineer who doesn't sketch out ideas when they are trying to think about new concepts," said Yang. "But there is little understanding of how such representations subsequently impact a design outcome."

Research suggests that the early stage decisions designers make have a significant impact on the cost and performance of the final product, said Yang. The Ideation Lab examines how design teams generate concepts, decide what to design, and, most importantly, how representations of these designs drive the early design process. They study when in the design cycle designers construct informal representations, what they represent, and the quantity and quality of these "sketches."

Research indicates that higher sketch quantity early in the design process correlates with good design outcomes, while sketching later correlates negatively. The simplicity of a prototype also correlates positively; complexity tends to lead to more negative outcomes.

With this in mind, it's clear that the conceptual phase of the design process often gets short shrift. "People have a tendency to jump to a design quickly," she said. "There's incredible pressure—because of deadlines, cost, competitiveness—to select a 'final' concept as soon as possible. However, you can't skip thorough the early exploration phase of design concepts, whether through visualization or other low-overhead modeling."

Taking a holistic view of the design process shows the benefit of thorough exploration in the conceptual phase. "Often when we think of engineering, we think in terms of creating and applying technology," said Yang. "Systems thinking means you consider broader issues, in addition to the technology—preferably at the very beginning of design. These questions must be answered: 'Who will use this product? How will it be used? What's the business case? How can it be produced?' These concepts should subsequently be rendered visually, through a drawing or model that a design team can refer to as a representation of their overall vision."

Sergey Naumov, SDM '11, Co-Leads MIT's Career Fair

By Cody Romano

Sergey Naumov

Sergey Naumov, SDM '11, was one of just two graduate students chosen to lead MIT's 2011 Career Fair. Volunteering as the event's employer relations director, he arranged for 300 organizations—including McKinsey, Exxon Mobil, Pixar, and the CIA—to visit the MIT campus during a week in late September—all while fulfilling his responsibilities as a father and a full-time System Design and Management (SDM) Fellow.

To build relationships with employers, Naumov spent eight months emailing, calling, and meeting face to face with company representatives. At the same time, he contributed to the career fair's website and participated in weekly planning sessions with a team of other students.

"Sergey is a natural leader who's always smiling and has a way of making people feel comfortable," said SDM Director of Career Development Helen Trimble, who identified Naumov as a candidate for the leadership role.

In addition to co-leading the MIT Career Fair team, Naumov personally addressed some of the concerns expressed by students and employers. For company representatives who wanted to size up attendance, he compiled a report of demographics from past years. For students worried about vying for an employer's attention, Naumov designed an online networking system that allowed job-seekers to make an impression without waiting in line.

He also worked to meet the specific needs of SDM job-seekers, who face a different challenge than most undergraduate students: as experienced technical professionals, most are seeking mid- to upper-level positions, while employers at career fairs typically recruit for entry-level jobs. Aware of this dynamic, Naumov lobbied for SDM students to moderate company panels. This allowed 18 students to network one on one with executives from such industries as aerospace and business consulting.

Naumov attributes the success of the career fair to effective teamwork and systems thinking. By collaborating regularly with others in his cohort via Skype and social media, he enhanced his leadership and communication skills and used them to build important relationships for MIT and SDM. Meanwhile, the holistic mindset that Naumov developed in classes like Systems Architecture allowed him to prioritize tasks more easily and strategically.

Although these lessons benefited Naumov, his duties as a full-time student, a researcher, and a dad made managing the project an even greater challenge. In addition to taking 14 courses since the spring, Naumov has worked as a research assistant for MIT's Chrysler Leaders for Manufacturing Professor Charles Fine, focusing on technological competition and developing computer models.

Originally from Moscow, Naumov and his daughter moved to Lexington, MA, several years ago. Before joining SDM, he was head of information technology for Akku-Vertrieb Ltd., a Russia-based battery manufacturer. After he graduates, he plans to work as an information technology project manager for a firm such as Google or Microsoft, where he will apply lessons learned from SDM and from planning the MIT Career Fair.

"My experience planning this event reinforced the necessity of carefully managing large projects and the importance of what I learned in SDM about project management, leadership, and teamwork," explained Naumov. "If the time is short and the project is really large, the success of your team depends on strong knowledge in all of these areas."

Competitive Advantage through Commonality: Finding Sustainable Benefits in Execution Challenges

By Eric Smalley

Bruce Cameron

Commonality seems like an obviously good thing. Why incur the cost of making different parts for different products if the parts do the same thing? As it turns out, however, commonality is not always the right thing to do. And even when it is right, it can be difficult to achieve. SDM's Bruce Cameron is using a healthy dose of systems thinking to tease out when commonality makes sense and how to get companies to pull it off.

Cameron, a lecturer in the Engineering Systems Division at MIT who teaches System Architecture (ESD.34), led a study of 16 companies' platforming strategies. Platforms are collections of technologies, components, and manufacturing processes that are used to produce sets of related products. One of the major benefits of platforming is commonality: using as many of the same parts and processes as possible across multiple products. This both reduces costs and makes it easier to go after niche markets, which can increase revenue.

"There's a lot of evidence out there that firms that do this very well see a lot of competitive advantage," said Cameron.

So an important question is, why do so many companies fail when they attempt to implement commonality strategies? A classic example is the Joint Strike Fighter program. The program target was to share 80 percent to 90 percent of parts across three variants of the aircraft, but in the end only 30 percent to 40 percent were shared, said Cameron. "That type of behavior and phenomenon is seen in studies that we did in automotive, consumer products, and transport," he said.

If commonality is such an obvious cost saver, why do companies struggle with it? First, it isn't right for everyone. It has "a downside as it turns out," said Cameron.

Commonality may inhibit the development process and make it hard to adjust to changes in the market that call for greater product differentiation, said Cameron. In other words, it's important that companies avoid having commonality become a straitjacket that limits the flexibility of their product development process.

Second, even when commonality is warranted, achieving it is more difficult than it appears.

One common counterproductive force is competition between product lines within a company over resources and control, said Cameron. A second counterproductive force is the intentional pursuit of uniqueness: the tendency of engineers and designers to come up with different results for the same task. Even slightly different designs can mean a company must stock more parts than if the designs were identical, he said.

Systems thinking helps tease out whether commonality is the right thing to do and how best to do it, said Cameron. Systems thinking allows companies to take a holistic view of the benefits of commonality, said Cameron. "This is much more difficult in a corporation because some departments will see big advantages, while other departments will be forced to invest in costly parts management schemes."

Once a company determines that commonality makes sense, systems thinking can help companies achieve it through platforming. Platforming is classic systems thinking, said Cameron. "If you look at each product line individually, you would make a different set of decisions. But, if you look at them globally and allow trade-offs between them, then you start to see how the benefit emerges," he said.

Cameron's current research examines how aware companies are of the positive and negative aspects of commonality and how systematically they make trade-offs. He's also studying the control mechanisms and incentive schemes available to companies for finding the right level of commonality.

Cameron is slated to deliver an SDM webinar in March on platforming and commonality. The presentation will include examples from the 16 case studies in Cameron's platforming study.

Quality Progress Features SDM Student Asgar Rahman

Read about Asgar "Oz" Rahman, SDM '09, in the Quality Progress article, "The New Faces of Quality."

Systems Thinking Conference Highlights Practical Applications in Healthcare, Education, Product Development

By Lois Slavin, SDM Communications Director

Whatever your profession, systems thinking is critical for success in the global economy, according to speakers at the 2011 MIT SDM Conference on Systems Thinking for Contemporary Challenges.

The annual event, sponsored by MIT's System Design and Management (SDM) program, drew almost 300 attendees from across MIT and around the world on Oct. 24 and 25. This year's conference, which highlighted SDM's 15th anniversary and featured several SDM alumni speakers who are now senior executives, focused on addressing complexity and innovation in healthcare, education, and product development.

Dr. Katharine Frase, VP Industry Solutions and Emerging
Business, IBM Research, discussed Watson and its
implications for industry and society.
Photo: Dave Schultz, SDM

Speaking live from Moscow, Russia, via remote videoconferencing, keynote presenter and SDM co-founder Edward Crawley first defined systems in order to give attendees a common understanding of the term. "A system is a set of interrelated entities that perform a function," said Crawley, the Ford Professor of Engineering in the Department of Aeronautics and Astronautics. The function that emerges, he said, is greater than what could come from any single entity — and the overall system's "emergent properties" are what produce value.

Crawley then explained that systems thinking is a way of looking at problems in context, in order to more successfully predict what will emerge to ensure value. "This is the real art and the real goal of systems thinking — training yourself in the domain in which you work to look at an unprecedented system, predict outcomes, and add value," Crawley said. In essence, using systems thinking helps make complex challenges less complicated.

Several speakers subsequently outlined complex challenges in healthcare. In Tuesday's keynote, Dr. Julian Goldman, who directs the Program on Medical Device Interoperability at Massachusetts General Hospital, described some of the problems hospitals have in managing incompatible medical devices.

The result, Goldman explained, is that most medical devices are not interoperable and cannot be synchronized or networked. This can cause "alarm fatigue" in medical professionals who must monitor multiple devices simultaneously, often while managing a medical crisis. The consequences can be dire. "We have many well-reported incidents of adverse results and patient deaths due to alarm fatigue," he said.

Goldman believes that the root of this problem is the failure of manufacturers to consider each device as part of a wider system. A main reason is the complexity involved in developing devices that connect to those produced by other manufacturers, because this would complicate a wide range of considerations at every level of the system, from user instructions to liability.

Other conference speakers highlighted applications and the need for systems thinking in fields as diverse as aeronautics, engine control, flu prevention, and food safety.

For example, John Helferich, SDM '10, former senior vice president of R&D for Mars Inc. and currently a PhD student in MIT's Engineering Systems Division, noted that although thousands of deaths from food-borne illnesses occur annually in the United States, the food industry does not have consistent safety standards along the supply chain from farm to table. "Think about all the things we do to fly safely," he said. "We don't have that stringency in food safety."

A panel titled "Watson, Analytics, and the Implications for Industry and Society" brought together IBM Vice President Katharine Frase, SDM alumnus and Bank of America executive Doug Hague, and healthcare consultant David Hartzband, who discussed computing in banking and healthcare.

The conference also emphasized education, from MIT Professor Richard Larson's discussion of technology-enabled learning around the world to new universities launching in Singapore and Moscow led by SDM co-founders and conference keynotes Institute Professor Thomas L. Magnanti and Crawley.

"The need is for people who understand systems and people who understand information technology," said Magnanti, the founding president of Singapore University of Technology and Design, which will open in April. "Future success will depend on educating more people in systems thinking."

Visit sdm.mit.edu to view videos of presentations delivered at the 2011 MIT SDM Conference on Systems Thinking for Contemporary Challenges.

Next year's conference will be held at MIT on Oct. 22 and 23, 2012.

Armando Hurtado, SDM '11: Managing the Complexity of Global Products

By Eric Smalley

Armando Hurtado
Photo by
Kathy Tarantola Photography

At first glance, you might think that someone who sells diapers has little need for systems thinking. Look a little closer and you'll see that in today's world of global brands, localized markets, and globalized supply chains, a deceptively simple object like a package of Pampers embodies a wealth of complexity.

Armando Hurtado's seven years in product development at Procter & Gamble have taught him that developing global products involves a host of challenges that the term "supply chain" only begins to hint at. Hurtado, SDM '11, is a senior engineer in product development. He worked on Pampers for emerging markets and now works on Gillette razors for Latin America and Asia.

One of Hurtado's tasks is cost engineering to make the products affordable in developing countries. He has to ensure that his global suppliers are not only inexpensive but will remain inexpensive for at least three years. This involves assessing political stability and currency exchange issues, as well as the usual measures of good suppliers.

Hurtado also analyzes consumer needs to tune products for different markets. He has to find the right balance between producing products that appeal to both the Brazilian and Turkish markets, for example, and producing products that can be aimed at different markets without reinventing the wheel for each one.

The key is thinking in terms of product platforms. "You're not just designing a product, you also need a manufacturing technology for that product," Hurtado said. "Plus, you need that manufacturing technology to be flexible enough to be able to change and make products that you don't even know you're going to need."

The complexity of managing a global product platform isn't well understood, which is a key reason Hurtado came to SDM. "I saw that in order to progress as a good technical leader, I needed to learn so much more than what I could just get from work experience."

Hurtado was looking for a program that linked business and engineering, and that could expand on his engineering training to help him manage complexity. He was also looking for a program that could help him reach his ultimate goal: becoming a CTO or vice president of research and development. "I saw that SDM was unique in that," he said.

The SDM program has given Hurtado tools that he's able to bring back to his job. He's been able to use statistical tools for predicting a design's robustness, and he's learned techniques for managing large complex projects. And simply being at MIT has helped him keep up with new technologies and trends, he said.

Being able to continue working while in the SDM program is a major advantage. "I didn't have to quit everything I was doing for a year or two," said Hurtado. "I'm still very much involved with my work. I love developing products and bringing them to market. And I plan to continue to do that in a leadership position, and influence the product development chain for Procter & Gamble," he said.

Conference Panel: IBM's Watson and the Rise of Analytics

By Eric Smalley

Irving Wladawsky-Berger

When IBM's Watson supercomputer bested the top two human competitors in a widely viewed bout of the Jeopardy TV game show in January, many observers saw an example of technology displacing people. Irving Wladawsky-Berger, however, saw the dawning of an age when powerful analytical tools like Watson will augment virtually every facet of human intellectual endeavor.

Wladawsky-Berger, vice president emeritus at IBM and a visiting lecturer in the Sloan School of Management and MIT's Engineering Systems Division (ESD), is scheduled to lead a panel discussion on information analytics at the 2011 MIT SDM Conference on Systems Thinking for Contemporary Challenges this month. The discussion, titled "IBM's Watson, Analytics, and the Implications for Industry and Society," will explore the impact such highly advanced tools are likely to have on a range of disciplines, including healthcare, finance, and education.

In the last few years we have seen incredible advances in information analytics, which involve processing large amounts of information with sophisticated algorithms running on powerful supercomputers, said Wladawsky-Berger. "The implications of having these powerful new analytic tools are really deep, and they go everywhere," he said.

When any powerful new technology arises, the key questions are how people use it and how it changes what people do, said Wladawsky-Berger. "How can we leverage these incredible advances in technology for business value, to raise the standard of living and the quality of life in our societies, to empower individuals so they can do a better job?"

These tools can have a big impact in healthcare. "Capture the world's top medical centers' expertise and distribute it via tools like Watson to a large number of physicians and healthcare workers, and you can significantly improve the overall quality of healthcare," said Wladawsky-Berger.

Katharine Frase, vice president of Industry Solutions and Emerging Business at IBM Research, will begin the panel discussion with an overview of Watson, including its application in healthcare.

Visiting ESD scholar David Hartzband will talk about advanced analytic tools in healthcare. Hartzband has a long history in academia and industry. His current work focuses on healthcare information technology.

SDM alumnus Doug Hague, small business analytics executive in Consumer and Small Business Banking at Bank of America, will talk about advanced analytic tools in finance. Hague leads a team that analyzes business performance, client behaviors, and strategic initiatives.

Financial services companies have long had an interest in analytic tools. "Risk management, identity management, security, those are all incredibly important questions to the world of finance," said Wladawsky-Berger.

Building these complex tools, not surprisingly, requires systems thinking. "The only way this can truly work is to take a holistic view of the problems we're trying to attack," he said.

Building these tools is as much about people as technology, concluded Wladawsky-Berger. "A lot of the design is about the interplay between the technology and the people who use it. What kinds of technologies and capabilities are the most useful? What kinds of tools would help [people] do a much better job?"