Please visit to see the new & improved site.
Please visit to see the new & improved site.

Wednesday, July 30, 2008

Sensing Stock Levels

More on my theme of sensing, but, rather than sensing water levels, this example is about sensing stock levels. It comes from implementation of a wireless-based system that BT implemented for one of our customers, Britvic, a softdrink producer and distributor. Wireless devices were installed in Britvic’s soda machines. Previously, drivers would have to drive around to the machines to check and restock those that required it.

Collecting and processing data from the wireless devices in each machine, Britvic was able to eliminate the journeys to those machines that didn’t need restocking. That reduced the mileage logged by their fleet by 10%. Knowing in advance what stock was required in the machines allowed them to stock their vehicles more efficiently and thereby reduce their loads by 30%. Assuming a mileage reduction of 10% equates to an emissions reduction of 10%, and that 30% reduction in vehicle load equates to an emissions reduction of 10%, we have about a 20% reduction in carbon footprint for the fleet as a whole.

Of course, we also have to consider the emissions of the wireless devices, the network and application hosting and perhaps the screen and computer at the customer’s operations center. But, we only get an allocation of most of these components as they are also being used by other applications. Added together, the footprint of the solution represents only a very small proportion of the emissions saved from a 20% reduction in driven miles.

And, the beauty of a solution like this is that it makes commercial sense even without the emissions reduction benefit -- just in terms of the cost savings and increased revenue from improved vending machine stock levels.

That said, as per my previous post, “RFID: A Double-Edged Sword in Sustainability,” it is always valuable to consider emissions benefits in your solution design, or what seems like a good solution can backfire on you.

Thursday, July 24, 2008

To Host or Not to Host? That is the question

Network-hosted solutions surely offer a reduction in the energy consumption on the users’ premise, but do they reduce overall consumption or do they just move it from the users premise to the carrier’s premise? I just spoke about this with Dean Parker, president and CEO of Callis Communications . Callis is a hosted telephony and managed IP communications service provider serving small and medium-sized businesses throughout America’s Gulf coast region and has tried to quantify the difference.

Callis estimates the energy consumption of an individual premise-based IPT solution at 20-30 amps. A hosted solution removes this consumption from the customer’s premise to the carrier’s and reduces the customer’s direct consumption due to telephony by about 74%. In comparison, a partitioned hosted solution running within Callis’ network operates at 80 amps, but with a capacity of up to 500 customers. Breakeven comes at about 20 customers.

Additionally, moving the IPT phones and the premise switch from independently powered to powered over ethernet reduces the consumption from 12 amps to 3 amps for a 24-phone Cisco system, taking the switch and the handsets into account.

Combining the energy benefits of a hosted solution and power over ethernet and assuming Callis puts 250 customers on their hosted solution, that represents more than an 80% reduction in overall energy consumption, compared to a premise-based, independently powered solution. Sounds good to me and plenty of justification to consider multi-tenanted hosted solutions.

Friday, July 18, 2008

This Year’s Greenest Conferences

We are a little over halfway through the year, and I have spoken on sustainability-related issues at several industry conferences - links to most of these events are noted in the “on the road” section of this site.

Comparing subjectively the extent to which each conference sponsor attempted to mitigate the environmental impact of their event, CERES and Freedom to Connect stand out.

CERES took a comprehensive and sophisticated approach and went to great efforts to work across their vendors, the hotel, transport and catering providers to do so. But, CERES is a devoted sustainability organization, so it should come as no real surprise that they did this so effectively.

In contrast, Freedom to Connect has no specific environmental sustainability mission. The event is organized by David Isenberg and is a meeting of self proclaimed ‘netheads’ engaged with “Internet connectivity and all that it enables…shaped by universal connectivity and the plunging capital requirements of information production…changing many of our fundamental economic and social assumptions.”

However, there was plenty of room on the agenda to cover environmental issues; there was no conference literature at all in hard copy - agenda and all other conference details were provided in soft copy; the conference was connected on the web and so it was possible to attend remotely (the quality of the webcast was poorer than expected, but it was a great first step).

Most importantly, taking up most of a giant screen was a chat session. Attendees had their laptops open and were chatting - reflecting on what the speaker was saying. This provided the potential for an attendee to avoid travel and participate remotely almost as actively as many of the in-person attendees by watching the broadcast and joining in the live chat.

I am sure it helped to have a highly technically literate audience and a limited budget. It certainly left an impression on me of what is possible if we rethink our approach to conferences.

Monday, July 7, 2008

Sensing the Environment with ICT

A short program ran on NPR a couple of weeks ago that illustrated the social and economic disadvantages faced by farmers in Ethiopia, unable to get frequent and reliable weather forecasts. The reason – the network of weather sensors is not nearly as granular as is available to forecasters in industrialized nations (West Africa 16 countries, 28 stations, UK 1 country, 31 stations). Common wisdom has it that, as a result, the forecasts are no more reliable than looking out the window. Listen to the report 'Developing countries and weather forecasts’ (it's only 4 minutes).

The World Technology Environment Center produced a report in 2006 called Sensors for Environmental Observatories. It is a comprehensive review of the application of networks and sensing devices to monitor environmental activity from water levels in river beds to pollution. “Sensor networks will produce a revolution in our understanding of the environment by providing observations at temporal and spatial scales that are not currently possible”.

To make more extensive use of these devices both to support farmers in a developing nations and scientists trying to make more sense of climate change patterns, we need to look towards the development of very low cost wireless devices that can be distributed widely, last indefinitely and run independently of the electricity grid. We also need a widely accessible mechanism of networking, that will allow a sensor anywhere to communicate – even if only intermittently. Symbiotic networks is one conceptual solution. If we can receive weather reports from Mars, we must be able to find a way to receive them from Central Africa.

I am exploring a couple of real life solutions for a future post. Please describe any you are aware of.