Here’s the map of who gets what.

My state of Michigan has two awards:

The Whirlpool project is focused on consumers, but will the DTE project include business facilities in the 660k smart meter rollout?

Smart meters are cool, cutting edge, and promising. But they are also complex socio-technical systems.

So getting the service design right is critical (interface, business model, technical functionality, integration, etc.).

Here’s a passage from a response to the smart meter ordinance in California that was drafted by Jared Blumenfeld (Director, San Francisco DOE) and Ed Harrington (GM, San Francisco PUC):

While I agree “it is difficult to estimate with certainty either the costs or the benefits” that is no justification for not doing so. Approximations with high and low case sensitivity analysis – including guesses about the qualitative factors that may drive solutions to either end of the ROI scale – can be very helpful.

I also agree that “proposed Smart Grid features are unproven on a large scale.” IS researchers can help by examining system design issues, service design issues, cost-benefit issues, adoption issues, competitive pricing issues, etc. That’s what we’ve done for the past three decades and that informs policy making.

Finally, I agree that “poor technical choices can result in significant cost escalations.” However, attributing a key problem to the technical domain masks the often bigger issues on the people and process end of things (see Nelson’s paper here, especially their conclusion that IT project mistakes tend to be “people or process related”). Again, IS reseachers have developed a wealth of knowledge about what works and what doesn’t and how to avoid failure.

“Connected Urban Development (CUD) demonstrates how to reduce carbon emissions by introducing fundamental improvements in the efficiency of the urban infrastructure through information and communications technology (ICT)”

Examples

Connected Bus in San Francisco

Smart Work Centers in Amsterdam

Personal Travel Assistance in Amsterdam and Soeul

EcoMap in San Francisco

Smart meters – the voice of the customer:

When I wake up, and there’s nothing on apart from things like my radio alarm clock and the fridge freezer, it’ll be registering about 0.9p an hour. But when I put the kettle on for a cup of tea it’ll immediately bump up to 29p. And I’m still amazed at the oven. When you switch it on and the meter immediately goes straight from green to red, you think maybe you shouldn’t cook that roast beef.”

GHG emission broken down by source (transportation, energy, waste) and zip code (click for bigger picture) in the Urban EcoMap.

Question of the day: What’s driving the differences across zips? 

Kudos to Cisco and The Climate Group’s Connected Urban Development Alliance.

The UK has plans to install smart meters in all homes by 2020, but focus groups suggest graphical displays are critical to changing behavior:

Most people preferred a graphic indicator of their real-time rate of energy consumption, expressed in terms of expenditure per day.

A gauge of cumulative spend was also a popular idea. Consumers thought that these features would help them track consumption and pinpoint areas of high usage.

The focus group members also felt that the ability to access historical data on consumption and compare it with current levels was essential.

The trust’s energy efficiency strategy manager, Ben Castle, told BBC News that only stand alone in-house display monitors could offer the immediate, direct and real time feedback on consumption that consumers need.

He said: “With a display, you have it in your house. It moves and it can catch your eye and it is very little hassle for you to check how you are doing. It gives a much greater certainty that people will receive this feedback.

“Other alternatives, such as online services and mobile phone technology, don’t give that immediate feedback that people need to inform their behavioural choices.

“I think having good, well designed displays accompanying smart meters is key to the objective of saving energy, saving carbon, saving money for householders.”

The goal of smart meters in all homes by 2020 is laudable as it provides a key infrastructure for information-enabled energy efficiency – using less electricity and lowering GHGs. But there’s a problem here. “IT systems” have a long history of ineffectiveness precisely because of small-scale focus groups and researcher/software engineer/(fill in the blank) bias (I have been guilty of this myself early in my career as a software entrepreneur). It goes something like this: “I think it’s a good system, makes sense, gets the job done, well designed, therefore others will.” 

There are likely to be unarticulated user needs that are best discovered via design thinking: observation, story, ethnography, customer journey map, etc. Moreover, the framing of this design problem is critical. Is this a means for energy efficiency, lowering my carbon footprint, enabling big brother to monitor my energy use, a way to cut my monthly energy bill? In sum, the design problem is not about technology or systems or monitors or displays – this is a service design problem. Maybe the headline should have read

Smart Meters (In) Need (of) Service Design

OECD Working Party on the Information Economy released “Measuring the Relationship Between ICT and the Environment” (2009).  The report contains a useful framework for asking important questions from supply (producer) and demand (in use) perspectives concerning positive and negative environmental impacts (click on diagram to see larger version).

Reducing energy used by home computers and TVs by 50% would save much less energy (and GHG) than using IT to lower energy used by washing machines, dryers, and other household energy hogs (e.g., by demand response) by 20%. Suggestion: do both. 

Times article implies digital age is feeding a hunger for energy:

But CO2 emissions due to computers and related is just a few percentage points (2.8% in 2020 BAU scenario) of all energy use (though it is growing three times faster than emissions from other sources):

And IT’s potential indirect impact on lowering CO2 emissions in other sectors (.8 GtC02) dwarfs its own use (180 million metric tons):

The Times Green Inc. blog reports today on a study of energy use in 100 business and residential sites after a demand-response system was initiated:

A smart grid pilot project in Fayetteville, N.C., has resulted in an initial 20 percent decline in average electricity consumption, according Consert, a Raleigh, N.C. technology company.

Those numbers are based on the first month of the project, a joint effort between Consert and I.B.M. that installed energy management systems for 100 residential and business customers of the Fayetteville Public Works Commission, the local utility.

Consert attached controllers on hot water heaters, air conditioners and pool pumps and then let customers go online and set targets for their monthly electricity bill. Smart meters and a wireless communications system provide real-time electricity consumption data to allow the utility to cycle appliances on and off to achieve the savings and help it manage peak demand.

The customer sets up a profile detailing when they wake up in the morning, go to work, return home and what temperature they’d like in their home.

“The consumer can say ‘I want my utility bill to be not to be greater than $200 a month,’ and then we’ll look at their past bill history to see if that’s achievable and ask what they want to do to achieve their goals,” said Jack Roberts, Consert’s chief executive.

This is a good start and suggests that demand-response has potential. I’d be more confident in the results if this were designed as an experiment with randomized participants assigned to groups (e.g., no demand-response, demand response) and controls for residential and business characteristics.