Distributed Sensing in Networks: The Example of Traffic

Lynne Kiesling

Yesterday I attended a conference at Carnegie-Mellon on valuation of distributed monitoring and sensing technologies. One of the primary benefits of distributed sensing is that it reduces transaction costs and information costs, thus leading to more efficient transactions and to the developement of new products, services, and markets. Sadly, this benefit is really, really hard to quantify.

So in the “when you have a broken leg you see lots of them” department, I was very interested to see article on traffic congestion websites in the WSJ today (subscription required). I think there are some lessons from traffic congestion information commerce that are directly applicable to electric power. According to the WSJ:

Several online traffic-information companies have made major improvements to their direct-to-consumer services. Using upgraded street-level monitors installed by government agencies, these Web-based reports give live traffic conditions — often with video — for commuting routes, calculate emissions, money and hours saved taking alternate routes, and even send an email or text message telling drivers what to expect when they get behind the wheel.

The federal government is part of the push to make these services more mainstream. It has helped launch traffic-detection Web sites in more than 20 states, most recently Florida and Idaho, and is giving states more discretion in how they spend their transportation budgets, an incentive that could allow states to vastly improve the network of cameras and sensors used by traffic sites.

In this case, the remote sensors are a government initiative, and the website information services are private and commercial. Another service that follows a similar model is weather forecasting (although the Accuweather folks think that NOAA goes too far in providing basic information, undercutting their ability to profit from providing weather services).

Is this a good model for the electric power network? Is the public benefit of reduced transaction costs high enough (or the entrenched barriers to doing it privately too high) that the installation of remote sensors should be a government initiative? I think this is an open question. Clearly there are substantial private benefits, but they may be so diffuse, and so obscured by the benefits to incumbent interests of maintaining the information-lite status quo, that a concentrated investment in sensing technology could contribute to more competitive markets. I offer this idea in the spirit of my long-standing belief that the role of public policy should be to reduce transaction costs that inhibit the private, mutually beneficial exchange of economic agents. But I am leery about recommending such government involvement in capital investment.

More on this as I think it through.

Oh, and of course, traffic sensing technology would create even more value if coupled with a road pricing system. What if the sensors communicate both to the toll gate and to customers when toll prices change dynamically because of varying congestion conditions? How cool would that be?

6 thoughts on “Distributed Sensing in Networks: The Example of Traffic”

  1. What I’ve always wanted to see is the marriage of GPS road navigation systems and these real-time sensors, so that your car can alert you to traffic conditions on the fly and re-route you automatically. If such a system were in widespread use (and a significant fraction of people actually took advantage of it), traffic flows would be ideally distributed.

    Of course, as usage of such a tool increased, the utility from using the tool would decrease… so there’s probably an equilibrium where only some drivers use it and there’s always a little congestion.

    What do you have in mind for electricity markets? A government mandated and funded real-time pricing upgrade?

  2. What I’ve always wanted to see is the marriage of GPS road navigation systems and these real-time sensors, so that your car can alert you to traffic conditions on the fly and re-route you automatically. If such a system were in widespread use (and a significant fraction of people actually took advantage of it), traffic flows would be ideally distributed.

    Of course, as usage of such a tool increased, the utility from using the tool would decrease… so there’s probably an equilibrium where only some drivers use it and there’s always a little congestion.

    What do you have in mind for electricity markets? A government mandated and funded real-time pricing upgrade?

  3. Is the government involvement in the traffic sensing merely a result of the government ownership of the streets and road right-of-way? If so, is government involvement in privately-owned transmission systems likely, or even possible?

  4. Or we could provide better training for drivers so we don’t have so many idiots on the road. You know, teach them how to merge onto freeways so they don’t stop and block up traffic for miles???

    In reading your blog, I thought you may be interested in a new blog by the Center for International Private Enterprise at http://www.cipe.org/blgo.

    Cheers.

  5. ENERGY MARKETS AND POLICIES: 2006-02-06

    As we noted in last Thursday's technology and innovation-focused New Energy Currents post, US energy policy has been a slightly hotter topic than usual after President Bush claimed that the US is 'addicted' to oil…

Comments are closed.