GE Plugs Second Internet
A second internet, otherwise commonly referred to as “The Internet of Things,” one on which machines are allowed to communicate and “learn” from each other may be in the works relatively soon.
By spending a billion and a half dollars over the next three years in technological research and development, GE is hoping to land in the forefront of companies hoping to use analytics predictively and reactively.
A significant portion of that R&D investment will go to a new software center in San Ramon, California, where machine-learning researcher Amil Varma is working on, among other things, determining which data is valuable regarding the maintenance of airplane engines.
GE’s interests span several sectors, from aviation to energy. In particular, one of GE’s goals is to place sensors which collect and store data from their 20,000-plus jet engines in an effort to predict when those engines will need maintenance. Currently, according to Varma, data is only taken for three statistical means: takeoff, flight, and landing. Any additional information, such as specific minute-to-minute temperature and pressure readings, are often discarded if the flight runs smoothly.
On the one hand, it makes little practical sense to keep data that serves little purpose. In aviation, however, it is slightly difficult to determine exactly which data is useful unless it is kept and further studied. According to Varma, GE’s goal is to have each GEnX engine (to be used in the making of Boeing 787s) equipped with a sensor that can track and store all of the possibly necessary data. Ideally, an airline would eventually be able to monitor those conditions on real time, keeping an eye on irregularities as they happen mid-flight.
The problem is that this constant stream of data coming from GE’s 20,000 engines alone creates bandwidth and storage issues. Thus, the idea of the “Industrial Internet” was born. The Industrial Internet—a term coined by project leader and former Cisco executive William Ruh, would serve as a platform for machines to do their communicating and learning without other traditional internet devices interfering and taking up space.
The San Ramon R&D Center, which has already been responsible for aiding Canadian utility companies visualizing energy grids during a storm, would be crucial in building GE’s second internet.
The result would benefit more than the While detecting maintenance problems before they become problems is a priority, so too is increasing energy efficiency. According to Ruh, the end is coming for how much more efficient individual devices can get (he is likely referring the diminishing of Moore’s Law), and thus it is essential for the machines to better use their power collectively.
“We know, operationally, that we can change 1%,” says Ruh. “It’s not going to be done anymore through better devices, because we are reaching the end of what physics can give us.” According to Ruh, even a 1% improvement would save them $2 billion.
The Industrial Internet or Internet of Things may someday help software makers and the silicon industry deal with the impending death of Moore’s Law. For now, though, GE is hoping it will let their engines live longer.