The New World of Smart Networks Created by "New Combinations"

In a keynote speech at the 2014 Techno-Mall of the Faculty of Science and Technology, Heizo Takenaka, former Minister of State for Economic and Fiscal Policy and a professor at Keio University's Graduate School of Media and Governance, stated that the key to creating innovation, as advocated by Austrian economist Joseph Schumpeter, lies in "new combinations." As I understand it, this means that new innovative technologies emerge by combining multiple, seemingly unrelated technologies. Last year, the Keio Faculty of Science and Technology launched KIF (Keio Innovation Foundry) as a place for industry, government, and academia to gather. Here, research has begun with KIF serving as a "Watering Hole" for technology—a place where different technologies can meet. At KIF, we have now established the Smart Network Research Center, where we are proposing the concept of an EVNO (Energy Virtual Network Operator).

A conventional smart grid is like a pool of electricity roughly the size of Shibuya Ward, controlled to keep the pool's "depth" (voltage and frequency) constant by ensuring that the amount of power generated and consumed is simultaneously equal. In the figure, two companies, EVNO-1 and EVNO-2, conduct business without owning any power transmission and distribution network infrastructure. These two companies adhere to the rule of simultaneous and equal amounts, each implementing control in its own unique way. For example, EVNO-1 has a power plant with a relatively large capacity margin, allowing its users to use electricity freely. On the other hand, when short-term demand arises, such as from a hairdryer, EVNO-2 increases efficiency by controlling nearby air conditioners and refrigerators (negawatts) to make it seem as if no demand has occurred, thereby bringing the demand peak-to-trough ratio close to 100%. In this way, each EVNO can vary its prices and service quality, which in turn can generate competition.

As shown in Figure 2, the actual control matching (pairing supply with demand) is performed with scalability using P2P technology (the same technology that was once popular for distributing music and movie content). This is automatic control between machines, known as an M2M platform. Since performing matching via P2P also poses security risks, virtual agents are placed on the cloud. Once a match is made, these agents have the functionality to perform 1) authentication, 2) transaction management, and 3) billing. P2P allows for flexible agent query responses. For example, to a query like, "Is 500W of power available for 25 yen?" responses could be, "It is for 30 yen," or "It's available in Osaka." To a request for power "from 10:00," a response might be, "It's cheaper if 11:00 is okay." While this is commonplace in the human world, the world of power control is also moving in this direction.

Let's not go into the technical details. This marketplace for buying and selling electricity is called an M2M platform. Let's consider Rakuten, a service many of you may use. A buyer who wants the cheapest electric guitar, preferably with no shipping costs (i.e., within pickup distance), is matched with a seller who wants the highest possible price. Since they are strangers, there is some hesitation in both paying the money and shipping the guitar. Similarly, the M2M platform mentioned earlier matches the buying and selling of electricity, monitors the actual generation and consumption, and charges each party accordingly. In other words, participants trade under the impression that they are "buying from the EVNO" and "selling to the EVNO," but in reality, the EVNO does not buy or sell electricity itself; it only manages the matching process. While this example deals with energy, in the future, it will be possible to trade all kinds of "M" (machine) resources, such as content, sensor data, and computing resources. For instance, if someone wanted to offer consulting to a convenience store on predicting customer traffic and managing inventory using sensors (like street monitors), it might be possible to charge for sensor usage at a rate of, say, 0.1 yen per use. This is a future that requires new "combinations," not only of communication and ICT technologies but also of fields like business, law, and even policy.