by David Jaber
Benchmarking doesn’t require site assessments, as the Orchard Hotel and Xanterra examples in Part 1 show. We just need the data.
For example, a state in the Western US wanted to look at the comparative environmental performance of five plants in one of its most prominent industry sectors. They already had data on a host of air and water quality indicators, as well as data on production. Looking at the trends was insightful.
Here we have wastewater flow per amount of production, graphed over nearly 60 months. Four of the plants show relatively stable wastewater generation per amount of product. There’s some fluctuation, but that’s almost always going to be the case, and since we’re technically not controlling against the fact that some months have more days than others, we shouldn’t expect this graph to be horizontal lines, even if the plants’ wastewater generation was exactly the same each day.
However, you see one of the plants is generating significantly more wastewater than the others, even when we’ve taken the amount of production into account. There’s also a lot of volatility (don’t be thrown by the big dip in the middle – that’s a missing data point or a plant shutdown). It makes one wonder what water efficiency practices, or internal water recycling, or other measures from any of the other companies may be transferred to this one to get those Factor 3 water savings. It also makes one wonder what the difference is between the peaks and the valleys, and if there are management techniques that can bring the peaks back down to the valleys. Are staff more effective in managing water in those valley months? Is the plant manufacturing more water-intensive products in peak months? Or does an irrigation line periodically blow, and it takes a month to discover the problem? In any event, we find insight and get help identifying problems and finding solutions.
Care in Benchmarking
You also need to be careful in benchmarking with what you “normalize” to, that is to say, the denominator that places measurements of different scales on the same footing. E.g. looking at total energy use per square foot, rather than simply total energy use, so you can compare buildings of different sizes. The industry sector shown above is very energy intensive, and one would expect a high correlation between amount of energy used and the amount of production.
In the hotel/resort sector, one would expect some correlation between energy use and “production” (which in the hotel case, we might use room occupancy levels, since occupancy is a measure of activity at the site). However, particularly in resorts with many smaller buildings, the energy use will start to be driven by quality of building shell and how well those building are managed. I.e. energy use starts to depend less on how many people you have present, and more on how well the building is managed, because there are so many more opportunities for losses and the energy use for the “production” process is not so much higher than the baseline building energy use.
So, for benchmarking of this sort, you do need comparable organizations to look at. You do need to think through how you’re going to compare different organizations and buildings. And the results do not tell you how to improve. But, when done correctly, the results do let you know where to investigate for improvement, and let you start to ask some really interesting questions.
David Jaber is a Principal at inNative, where he helps organizations get better intelligence on where their significant environmental impacts lie, and how they can improve. http://www.innative.net