Computational model offers insight into mechanisms of drug-coated balloons.
"Smart" fettuccine that tells a microwave how it should be cooked? Vials of medicine that alert the consumer when they've expired? An office that can understand -- and respond to -- its occupant's verbal commands?
Welcome to "Smart World," where everyday objects communicate with each other, the Internet and us via embedded computer chips and other technologies.
As became clear at an MIT conference April 12-13, we are at the threshold of such an era; indeed, some applications are already here. In an agreement reported in the Wall Street Journal during the conference, International Paper and Motorola are creating smart packages that will be on the market by the end of the year. Among other benefits, tiny chips embedded in these boxes will allow manufacturers to automatically determine exactly what is in a given box without human intervention.
"This calendar year I think we will start to see a greater public awareness of [smart technology]," said Guy Mason, vice president of CHEP (the world's leading pallet manufacturer) and a panelist at the conference. "I'm betting my career and a large chunk of my company's money on the fact that it's here."
"Smart World in the New Millennium" featured talks by representatives from MIT and industry on the technologies and applications that will drive this new environment. It was sponsored by the Industrial Liaison Program and the MIT Auto-ID Center, where researchers are developing "automatic identification," or the bar code of the future.
"For three or four decades we have been entering the computer's world," said Dr. Victor Zue, associate director of the Laboratory for Computer Science (LCS). In systems now under development, "the computer enters the human world. Cars, houses and rooms will be instrumented rather than your having to carry those instruments with you."
Dr. Zue described an ambitious project to that end involving some 200 researchers in the Artificial Intelligence Laboratory and LCS. The Oxygen Project aims to "bring information technology to the people," he said. It will do so via handheld devices that communicate with sensors and other systems embedded in rooms. Everything is supported by an overarching computer network.
People will interact with the system verbally and visually. For example, "say I walk into [a co-worker's] office, and he asks me if I can join him on a trip to Hewlett-Packard," Dr. Zue said. "I pick up the [handheld device] on his desk, which recognizes my face and begins to personalize itself to me." In other words, the device automatically caches data it thinks the user might need, like an address book or personal calendar.
"I then say, 'arrange a trip to HP on May 14 and returning a few days later.'" After delegating the travel arrangements, "I put the [device] down and return to my office."
Oxygen is a playground in which the researchers are integrating technologies they've developed over several years. "We want to be able to stitch these things together in a graceful way," Dr. Zue said. The goal is to have a prototype system in five years, though the first version will come out in 2001.
The Auto-ID Center is another group working to make our world smart. Researchers there are focusing on the "bar code of the future," which will be composed of electronic tags on everything from laundry detergent to pet food. Electronic grids built into the environment and connected to the Internet will automatically "read" the e-tags.
The current bar code, or Universal Product Code (UPC), defines groups of objects. "We're trying to go beyond that to define each object itself," said Dr. David Brock, co-director of the Auto-ID Center. The Electronic Product Code (EPC) under development is a numbering scheme that will allow "trillions upon trillions of unique identifiers," Dr. Brock said.
The new EPC system "will enable real-time remote monitoring and tracking of every tagged physical object," said K.Y. Sunny Siu, research director of the Auto-ID Center and associate professor of mechanical engineering. As a result, among other benefits products could be tracked from manufacture to disposal, and information on their use could be accessed from the Internet.
The ability to track an item through its life cycle could cut costs in industry by making the supply chain, or movement of products from manufacturer to consumer, more efficient. Right now it suffers from the "bullwhip effect," in which perturbations at the retail end are magnified at the manufacturer's end.
This is largely because "manufacturers are trying to guess at lots of things," said Sanjay E. Sarma, a co-director of the Auto-ID Center and an associate professor in mechanical engineering. Better information flow, however, could help solve the problem -- and cut costs. For example, Professor Sarma and mechanical engineering graduate student Yogesh V. Joshi have found that "once manufacturers have confidence in how much a retailer needs, they can scale down their safety stocks."
At the panel discussion on the use of smart technology in industry, Bruce Lynes, director of research and development at International Paper, ticked off some other benefits of such technology. For example, smart chips embedded in, say, a perfume package will assure that the consumer "gets the real thing and not a knockoff," he said.
In the pharmaceutical industry, smart technology will allow us to "track [a medicine] throughout its life cycle to make sure it's in the right environment, or is not getting bad," said Dirk Heyman, global head of life science and consumer product industries for Sun Microsystems. Like Dr. Lynes, he also noted the application in authentication. "For example, is this blue pill really Viagra from Pfizer, or a fake? That's a big problem right now."
In five to 10 years, smart technology "will percolate into our homes," predicted Chris Luebkeman, director of research and development at Ove Arup Engineering and a founder of house_n, MIT's Intelligent Home of the Future project. He envisions "buildings that are aware, that adapt to us."
The panel was moderated by Kevin Ashton, executive director of the Auto-ID Center.
A version of this article appeared in MIT Tech Talk on April 26, 2000.