U.S. Department of Energy - Energy Efficiency and Renewable Energy
Building Technologies Office
National Residential Efficiency Measures Database Unveiled (text version)
Below is the text version of the Webinar titled "National Residential Efficiency Measures Database Unveiled," originally presented on January 18, 2011. In addition to this text version of the audio, you can the presentation slides and a recording of the Webinar (WMV 47 MB).
Welcome to today's webinar. Ms. Werren, you may begin.
Hello. My name is Gail Werren, and I'd like to welcome you to today's webinar entitled National Residential Efficiency Measures Database Unveiled. This webinar is presented by the Buildings Technologies Program at the U.S. Department of Energy.
We're excited to have with us today one of the database's principal developers who will talk about how it functions, what progress has been made to date, and planned enhancements for Version 2.1 and beyond.
But before I start, I have some housekeeping items to cover. First, I want to mention that everyone today is on listen-only mode. We will have a Q&A session at the end of the presentation.
You can participate by submitting your questions electronically during the webinar. To submit a question, click on the Q&A link on the top bar of your screen, type the question in the box, and click ask. Please be sure to click ask and not the symbol of the raised hand. Our speakers will address as many questions as time allows after the presentation.
Also, I wanted to point out the URL on the screen, www.buildings.energy.gov/webinars. On that webpage is a link to see today's slides. Today's presentation is being recorded, and a video of the presentation will be posted in the near future. You can also view past webinars on the archives page.
Finally, we have a few quick questions to ask you to help us learn more about the audience and target future presentations. We will start with two questions now and then have more questions at the end of the presentation before the Q&A session. Please click on your screen to indicate the appropriate response.
The first question should be on your screen now. Please review the question and the answers, and click the appropriate response. How many people at your site are participating in today's webinar? We're about to close this question, so please vote now.
Now the next question. We are looking for information about what best describes you, your organization or affiliation. We're about to close this question, so if you have not voted, please vote now. Thank you so much for your participation.
And now I'll introduce our speaker Dave Roberts. Dave is a senior engineer at the National Renewable Energy Laboratory. He currently supports DOE's Building America Program technical activities, including the database project we're discussing today and the BEopt software development efforts.
During today's webinar, Dave will outline the current status and future direction of the database project. Sean Casey and Noel Merket, two other NREL engineers who helped develop the database, will also be on hand to field questions during the Q&A session. And with that, I'll turn the presentation over to Dave.
Thank you Gail. Good morning everybody. Thanks for taking time out of your busy day to join us and learn a little bit more about this project. This is a photograph of our campus here in Golden, Colorado, and this is where Sean and Noel and I are sitting at the moment. And it's just as sunny here right now as it looks in this photograph.
So the way I've got this little presentation organized today is I'm going to go over kind of the initial concept of the project, the genesis of the project, give you a little background on what we've done, tell you where we are today, and then kind of look ahead at our next release and beyond.
So the initial project was conceived — or the project was initially conceived at Building Technologies Program at DOE, and NREL was first tasked with this project in October of 2009. And the project supports the current administration's goal of retrofitting the existing housing stock in this country.
I'm sure all of you know we've got over 100,000,000 homes in the ground, a lot of those need work, and this is all part of the current administration's goals to create jobs and help with green — reduce greenhouse gas emissions.
The primary idea here is to integrate some existing databases that DOE has been funding and supporting over the years into a single database, single unified national database, and make those — make that database available across all the DOE labs and to the general public.
The databases that we're trying to unify across the labs include Lawrence Berkeley National Laboratory's Home Energy Saver Software Database. The Home Energy Saver Software has been around for a number of years. It's a Web-based application. It's a consumer-faced software where homeowners can go and do an assessment of their home.
More recently, there's been a professional version of that software that's been made available, the HESPro Software, and it is also the software that underlies DOE's Home Energy Score Pilot Program that Vice President Biden and Secretary Chu announced several months ago. And you can find more information about that on the Internet if you look around.
NREL has a software — has had a software tool that we use primarily internally within NREL and share it with our Building America research teams. And this is a residential research program that's been in place for about ten years. And it's an optimization tool that helps steer our research toward very efficient optimized, energy-efficient optimized homes. And so it steers the research program and finds us where the — finds where the next gaps are that need to be as we try and make new and existing homes more efficient.
And then finally, there's the Oakridge National Laboratory tool, the Weatherization Assistant, which is new in DOE's Weatherization Assistance Program, which is the low-income weatherization program.
So what we're shooting for here is to create a singular database and a standardized format, and so that we have technical consistency across these different tools and other private players, software providers that want to use or other programs and organizations that want to use the data, we're all working from the same data source. We also want to have these data in a place where they can be publicly vetted, and we can all work together to improve the data through public transparency.
The data that are in there right now include performance parameters for retrofit measures. And we'll talk more about this, but performance parameters include things like U values of windows and SEERS of air-conditioners. It also includes a range of estimated costs that one might expect to find for these different measures. And the database is being hosted here at the National NREL Energy Laboratory. Our development team here at NREL consists of engineers, building scientists and software developers.
And outside the laboratory, we pulled together an Advisory Group when we first kicked off this project. The Advisory Group includes the other two laboratories, it includes a representative — we're lucky to have a representative from California's Database for Energy Efficient Resources, which is a very similar project, although it's California centric.
We have a couple of software — audit software providers that are helping us. We have an executive director of Affordable Comforts, and which is a home performance professional organization. So they represent home performance contractors, by and large.
We have two home performance companies, Re-Curve and Greenhouse — I'm sorry — GreenHomes America helping us. And finally, we have a couple of the DOE Building America Research Teams, Building Science Corporation and Steven Winter Associates.
The Advisory Group helped us to establish initial design and content of the database, and they continue to provide feedback and help us establish the future project.
This is an overview of the project as it sits right now. We have the retrofit Measures Database residing within NREL firewall. The NREL administrators, Sean and myself and some of our IT folks have access to it, and we're the ones that populate and mange the data.
We've put together a public-user interface that's a Web-based interface, and that's a place where you can interact with the data, look at the data, and provide public comment and ask questions about the database.
We've also created a couple of data feeds, XML feeds. XML is extensible markup language, which is a standard data transfer protocol. And we've provided those feeds to the laboratories and any private sector software tools that might want to download these data electronically and use them in their applications.
And then finally, we're trying to collect data from historic existing, past retrofit programs. And going forward, we'll be collecting data from programs that are just getting started and ramping up. And those data will help inform the Measures Database and help us do research into existing home retrofits as we go forward.
We're working with a classic software-development cycle here. We're trying to shoot for a release every six months. I think our first release was in February, that was Version 1, and Version 2 was released in the fall. And right now we're working on Version 2.1, and we're kind of in the design/development part of the cycle here. So it's a good time to get feedback from you folks and help us with Version 2.1 and beyond.
So the database is up. It's hosted here at NREL. Version 2 is the current version that's up on the website. It was released in November. It's got the Web UI, it's the XML export capability, and we have approximately 2,800 measures in the database right now.
We started with a concept and design developed here at NREL in concert with the Advisory Group, and we tried to develop a database schema that's flexible and expandable, so as we move forward, we can adapt and add data and add concepts.
We also started with the Lawrence Berkeley Laboratory's Home Energy Saver Data, as I mentioned. That particular tool has been around for a number of years, and they had measures data that they were using in that software. And that was kind of our starting point.
As we moved forward, we developed data collection and processing tools that allowed us to collect data beyond the Home Energy Saver Data, process those data and prepare them for uploading to the database.
At this point, LBNL aligned HES, HESPro, Home Energy Score. There's actually three tools that are all closely related. They've aligned the data in those tools with some of the measures in the database. Not all 2,800, but a subset. And likewise, NREL has aligned the BEopt Version 1 Measures Library with the measured database or a subset of the measures in the measure database.
Lots of boxes, circles, lines. I won't belabor this. I just want to point out the main concepts here. This is our database schema for Version 2. And the two primary objects here in the database that you should focus on, in the middle of the page are actions and components. A measure, in our mind, is made up of components and actions.
A component, in our little world here, are basically — is basically a material. So examples would be walls, windows, air-conditioners. An action is basically a labor object. So it would be replace, install, remove. And so a measure — a typical measure might be replace an air-conditioner. So that's made up of two components, the old air-conditioner and the new air-conditioner, and an action to replace.
If you follow the arrows from the component box out to the right, you see it points to costs. So components hold cost. And the cost for a component are basically just the material cost. The cost of the installing; the cost of the air-conditioner.
If you follow the arrow down from the component box, you see it's pointing to a box called properties. Now, properties include three basic categories. We have performance characteristics. For example, the SEER of an air-conditioner, the U factor of a window. It includes performance standards.
So what we tried to document here is when a particular component meets the performance standards of either the federal standards for like appliances or the code, the International Energy Conservation Codes for wall insulation, attic insulation. Or we also have Energy Star performance standards in there, so you can see if a particular component meets the Energy Start Home — or the Energy Star requirements. And then finally, we have a lifetime for the measure. That's another property of the component.
Actions. If you follow the action box, it goes to — look at the action box and follow the lines to the right, it also points at costs, and these are labor costs, the cost to install, replace, remove.
And then finally, in the lower right-hand corner there you see references. And there's not a whole lot of data there right now. Moving forward, we plan to populate that with links to best practices and measured definition documents so we're all on the same page on what a particular measure entails and how it's implemented in the field.
So to give you an idea of what we — how we populated this database, we built a system of tools to facilitate bringing what we call raw data and processing it into the final data that shows up in the database.
So we collect data for any particular component or action, we collected data from a variety of sources and brought those data into our tools and aggregated the data and showed those on the database.
So for example, the cost data we analyzed statistically. And we'll see when we get to the database that we present an average cost for a measure, and we present a range of those costs, expected costs for the measure.
And the range — the limits of the range, kind of the simple to the complex installation are represented by what we calculated to be the 10 and 90th percentile of all the cost data we collected.
We collected data from a variety of sources, including home performance contractors, the database — the California DEER Database, which is the construction cost database, and from retailers for things like appliances; we can just look at the retailers.
And generally we normalized the costs before we uploaded them to the database, so they could be applied across a range of homes. So for example, insulation is — insulation costs are reported in dollars per square foot of treated surface area. Air-conditioning, heating is in dollars per 1,000 BTUs per hour of heating and cooling capacity, that kind of thing.
The — it's important to realize as you look at this that he limitation of trying to use an average cost in any given home, you know, we got consistent feedback from our Advisory Group and others that have provided comment that it is very dangerous to apply a singular cost in a home without actually doing an assessment of that home.
The ranges that are reflected in the database are meant to encapsulate all the types of things that you might encounter as you look across lots of homes. So access, location of the home, you know, there's regional cost differences. If it's a rural property, you're going to do a lot of driving. If you can't get into the attic, you got to cut a hole. The crawl space is a foot high or three feet high. All those things dramatically influence the cost of implementing the measure in any given home.
So just a word of caution that as you look at the data in the database, the averages really need to be used carefully. And it's the range of costs that one might expect to find as you move across homes and move across the country.
We also built some testing and quality assurance tools, some data visualization tools that once we have all the data up into the database, it allows us to look at the data visually. And as you compare across different efficiencies and different sizes, you can look at the performance of the measures, and you can look at the performance of the components, and you can look at the cost of the measures visually, and this helps us to find outliers and problems and fix them before we publish the final database.
So as I mentioned, there's a public Web interface, and the idea of this is to allow you to review data. The XML feeds are there. We've got all our documentation loaded there, and we provided a couple of pages for submitting comments and submitting data.
I'll give you a quick tour, just real quickly. This is the homepage. This is what it would look like when you land here. On the left here, this is the navigation menu. Click on any of these, that's where you'd move to.
Down below here — I'll just go quickly through these supporting resources. We have a data dictionary. The data dictionary shows every table, every field and every table, defines what kind of data you're going to find in those fields, and a quick description of those.
As I mentioned, we have some XML feeds that can be downloaded here, and I'll talk more about those. There's two XML feeds. I have a slide to discuss that. We provided a glossary, so we're all talking the same language. If you have questions about how we are using a particular term or what it means to us, you can go to the glossary.
We put together a guide that's a PDF document specifically for application developers. So these are the software developers who may want to download the data and utilize them in their applications.
And then finally, there's the develop document, the document that tells you how we collected the data, how we processed the data, what our data sources are. So it describes in detail sort of the underlying method that was used to populate the database.
Click on the view data now button, it takes you to this page, which shows all the measured categories that we have in the database. And it's broken up by groups that — or they're grouped in a way that hopefully makes sense to everybody.
And these are all links. So just as an example, I'll click on the attic and ceiling measure list. Our definition of a measure is a before component and after component and our cost to get there. So if we look at the very first one here, I'm looking at an un-insulated attic, and I'm ending up with an R-19 cellulose attic, and I have a range of costs to get there that ranges from 20 cents to 91 cents per square foot with an average value of 61 cents per square foot.
Under each component you see a list of properties. In this case, for a ceiling, we have the insulation thickness, the insulation type. We have a lifetime. In this case, it's essentially infinite. And we have the R value of the insulation that's been installed.
If we look at the second measure here, you'll see in the component here, we have an R-30 attic. And you'll see — in addition to the properties, you'll see performance standards. And so what we're reporting here is that an R-30 ceiling meets the 2009 IECC for the climate zones that are listed there.
Look at another example. I'll go over to refrigerators. Here you see a lot more properties for refrigerators that we had for attics, including the volume and the rated annual energy consumption. We have — shown, again, performance standards. This one meets federal standards. If I scrolled down, we'd start to see Energy Star requirements showing up under performance standards.
The other thing we did is added a filter here at the top. It allows you to filter on the starting point. So if you walk into a home that has an un-insulated attic and that's all you're interested in, you can filter and look at only those measures.
Just below that we report the total number of measures for this group, this category. And because it's filtered, I'm showing you how many are here. So if we scrolled down, we'd see three refrigerators.
We also added a page to allow you to submit your questions and comments. So if you don't get them in today during the presentation or if you have — if you spend time with the database later and you want to submit comments or questions at that point, you just click on this link. We also submitted a page — or added a page to submit data. We're soliciting data, and I'll talk a little bit more about that later.
So back to the XML feeds. As I mentioned, there are two, showed that they're available view the Web interface. We put two on there because one of them reflects basically what you see on the website. All the measures that you see on the website are in that XML feed, and they're organized just as they are on the data — on the website. So it's a measure, and each measure owns a before component and an after component and the cost to get there.
We also made available a more flexible, larger feed that includes all the data in the database and follows the database schema that I put up earlier with the all the boxes, circles, lines. So it's a loose collection of components and actions. And this gives more flexibility, more responsibility to whoever's downloading it, but allows them to basically assemble measures as they choose to rather than, you know, getting what we've decided are measures that we wanted to put on the website.
Each of the feeds comes with two data files. One is the XML itself, which is the data in an XML format, and the other one is what they call a schema. It's a schema document in an XML format. So it doesn't mean much to most of it. Just the programmers will understand what that means.
So looking ahead, we're going to do a little revision here in the spring. Not a huge overhaul, but there's a few holes in our database. We need to add some measures for some of the more popular things that it has. And so they haven't quite made it into the database yet.
And then we're going to — you know, because of the range in such an important concept here, the cost range, we're going to add some information to each page. Again, data driven, so it'll be in the database as well. But a list of factors that kind of drive that range so that folks that are looking at the prices can be cognizant of those things that drive the range.
And then we're going to start expanding some of our measure properties. For example, rather than just having a SEER for an air-conditioner, we'll start adding some performance curves. And, you know, that kind of information is useful and important to those that are modeling these air-conditioners and software.
Further down the road, I mentioned already, we're going to hook up some best practices and other resources to the measures so that we're all understanding, you know, the best way to implement these measures and that we're doing the right thing in the field.
We're going to add some operational data for the programmers and software developers out there. You know, a lot of the assessments that are being done are what we call asset ratings. So the idea is to reflect the performance characteristics of the home rather than the way a particular occupant happens to be driving the home at the time.
So what we need is everybody to be on the same page regarding thermostat set points and the way the blinds are used and how often the lights are on and all those kinds of things so that we're all producing asset ratings using the same underlying assumptions. And a lot of that research is going on right now regarding, you know, what's the best assumptions to use.
We also want to get a lot more detailed on the cost data and break out and present labor and material costs separately, provide regional multipliers, so as you're, you know, either working in, you know, a rural part of a state versus, you know, New York City, the labor costs can be adjusted to account for that.
And then the other drivers I was mentioning as well. Access, drive time, complications, you know, lead paint, asbestos. Those kinds of things that really can drive the cost, have those broken out, so, you know, as you're putting together or trying to estimate cost, you know, you have something to work with and more detail.
And then finally, we're working to pull together retrofit project data, both historical data and data going forward as these programs continue to ramp up. And basically, what we're looking for, you know, are data that — basically, audit data, utility billing data and retrofit cost data. That's basically what we're looking for.
And the audit data, the characteristics for the home and the billing data will help us to do — to test our software programs and do other kinds of research. And the cost data, obviously, will help inform the cost data that are presented in the database.
So the basic concept here is to create a sandbox that researchers can utilize to answer important questions and help the retrofit industry grow and evolve. You know, remember, our goal here is to retrofit 100,000,000 homes, you know, kind of scale the industry up to 5,000,000 a year. So this is going to be going on for a while. There's a lot of work ahead of us.
And to the extent that we can collect and organize these real-world empirical data into a single — in a single place and make these available to the researchers in the labs and outside the labs to test their software, to test new procedures, new audit procedures, to see if we can figure out which homes need retrofitting when we're looking at, you know, huge databases of utility bills, those are the kinds of things that we're trying to facilitate through this project. And then, of course, the cost data to continue to improve our — the estimates in the database.
So as part of that effort, we're looking — we're planning to evolve our data intake process going forward. Right now we're taking data in basically any format, you know, through the public UI, and then we're manually looking at those data, assessing their value, organizing those data.
As we move forward and we see more and more data, we're hoping to standardize the data format and automate the kinds of processing and tools that we need to interface with those data in the long term. So as we evolve and mature the data-collection process, it'll help us to deal with more data, which will help us to answer grander questions.
So having said that, if anybody has any data now, it could be very useful to the DOE research — retrofit research goals. And as I mentioned, you know, you can either upload it via the website, or you can contact us here at NREL, and we'll help you help us.
So thank you very much. I see Sean and Noel here are rapidly looking through lists of questions. But if you have questions and you haven't uploaded them yet, feel free to do that now. Do you have any questions?
Thank you Dave. As I mentioned, we have some additional polling questions before the Q&A session begins.
The first question is up now, asking what you were hoping to learn from today's webinar. Thank you.
And the final question will ask how satisfied you were with today's webinar. Okay. We'll be closing this question now, so please vote if you haven't done so. Thank you.
Now we'll get into the Q&A. As we mentioned, we ask everyone to submit their questions online. Dave, Sean and Noel will address as many questions as time allows. Sean, would you like to start?
Sure. Thanks Gail. Hi, everyone. This is Sean Casey. Thanks so much for all the questions. Noel Merket, Dave Roberts and I will answer as many as we can. And if we don't address your questions, feel free to send it via the Measures Database website.
So with that, the first question I'd like to address, sort of a package of questions, but they evolve around whether or not energy savings per measure are included in the database. And sort of a little further is are there payback estimates for particular measures?
And estimates for typical energy savings are not listed in the database. The database is designed to provide consistent measure definition in a standardized format to improve the technical consistency and accuracy of results from software programs as a whole. So ultimately, those energy-savings predictions would come from software programs outside of the database.
So ok, here. All right. This is Noel Merket, and another question we got several times is about the costs and how — whether we're taking into account regional factors and things like that into the cost.
Right now all those are included in the range. And in the future, as the database continues to be developed, we plan on including regional factors such as labor and that and splitting up the labor and material costs in the database to better allow calculation of the costs, depending on where you are.
Another question we received is, is climate data collected? Now, I think that depends on the definition of climate data. Currently, the database does not contain weather-related climate data. But properties like performance levels that are climate specific such as IECC code compliance or Energy Star compliance, are covered in the database structure.
Now, for example, if you were to look under attic insulation measures, with a particular R value installed, you would see whether or not that measure would qualify for IECC 2009 in that specific climate zone.
Hi. This is Noel Merket again. There's another question here of does the cost vary by date? When we received the data points that we put all together to make the cost data, we — each one of those has a data associated with it. And as the dates get farther into the past, we'll either correct them for inflation or just drop them out completely if they're really old cost data.
And so right now the cost information in the database represents costs for those measures right now. And as the database continues to be updated, those costs will be updated to bring the costs to what they are at the time of release of the database.
Okay. And while we're on that, too, there's another question here regarding cost. How do you deal with the fact that there's a non-zero intercept for all costs? And that's a very good question. It's another thing we're looking to add in a future release of the database. That's something we've been working on quite a bit.
So right now it works on the assumption that there is a zero intercept. So as things get very small there may be some inaccuracies there, but look for that in the future release of the database, that we will be including that kind of a slope intercept model for costs.
I see another question about access to the data. And when will it be available? So the data is available publicly and accessible to everyone right now as we speak. And it's available in two formats. One would be going to the website listed on Dave's slides. And from that database, you can also download the contents of the database from the website in an XML format.
So this is Noel Merket again. One of the — we've had a couple questions about how to access the database and that. And what we use for the database on our end, it's a Microsoft Sequel Server database, but the way that it's being distributed is through these XML feeds.
One of the questions we had was whether there's information about the XML feeds, and I think the best way to answer that would be through — the information about how the XML feeds is structured is on the page where you download it. There are schema files for the XML feeds, which give direction for how the XML feeds are structured, and can be used in many different software environments for importing that information into your software for whatever needs that you may have.
So that's currently how this database is being distributed. And so it should be available for any platform that can understand XML, which is just about every platform out there. And be interpreted through these schema files. And we'll continue to develop the schema files as the feeds are developed, and include more documentation in those.
Another good resource for how to use the data is in that development document, but also on the data dictionary page, it says what each and every field means and how it's related to other fields. So from the application-developer's standpoint, it — that's a good way to get started on how to use the data.
Thanks Noel. So there's another question. Will the database provide incremental costs? Currently, all of the costs listed on the website are for total installed measure, meaning a combination of labor and component.
One could calculate the measures if one wanted to by downloading the XML feed and programmatically calculating incremental costs or just by hand, taking the existing unit and calculating incrementally what the more efficient unit installation would cost.
Also, related to that question, I see another — let's see. Another one about — let's see — oh, managing the Community Energy Retrofit Program. Is there a set way to get data to DOE without much labor cost to you?
If you wanted to examine labor cost and material cost separately, again, you would want to download the XML feed and examine that question that way instead of going through the website. Because again, the website lists total installed costs for both labor and materials.
I see a question about the upload process. How does the database ensure data — duplicate data is not entered? And could two providers submit data for the same project? How do we control this?
I think that's a process that we are going to establish guidelines for controlling, quality control, making sure that the data we get is used in a consistent format and that we are not sort of counting the same data point twice from either the same source or different sources. So again, whatever detailed information you can provide to improve the overall data is requested, and is much appreciated.
All right. Another question we had is has any consideration been given to developing these considerations into cost benefit from a homeowner's perspective? Probably the best — the purpose of the database isn't to make that judgment call. It's to provide information to inform those judgments.
One of our first users of this database is the Home Energy Saver Web service that Lawrence Berkeley Lab is putting out. And that is an excellent tool to provide those sorts of cost-benefit considerations. And there are other tools that we're using, and this database will inform making those decisions.
Another question we had is do you plan to expand the database outside of the residential sector? No, there are not currently plans to do so.
Thanks Noel. I'm seeing a question about clarity for costs and whether they're specific to a state or a region. Currently, the costs are to be nationally representative. But certainly, in future releases of the database, we will be incorporating specific regional considerations for costs.
I see — this is Dave. I see a question here. Do you see the financing industry being a user of this data? And I guess my short answer would be no, I don't, but the research that — some of the research that we — that DOE plans to conduct, that the labs would conduct, some of the conclusions that are drawn from the retrofit program data that we are collecting will probably find its way into reports analysis and reports that the financing industry might use in assessing the risk associate with lending against retrofit projects going forward.
So this is one of the key questions right now in the financing industry is, you know, how real are the savings that are being predicted from the software and the audits? And one of the overarching goals here is to collect as much data as we can to try and answer that question.
Another question I see here is a somewhat specific one about refrigerators, but I think applies to many different component types. The question is, is the pre-usage number for refrigerators used in your example representative of all fridges or does it represent some specific population of fridges, older units, for example?
With the refrigerators, in particular, we based our refrigerator selection on the units that were considered in the standard-making procedure for refrigerators. So the ones where they came up with the federal standards for refrigerators.
So they are generally representative of a specific population of refrigerators, but I think the question comes down to — and it's one that's kind of been alluded to several times here — is are there — are the — all the options in the database representative of everything out there? And the answer is most certainly not.
There — we're trying to get a representative sampling of the units available, but there's really no way to have every possible refrigerator that could exist in somebody's home in the database.
So the idea is to get a representative sample and to show where, you know, the — for instance, if I upgrade to an Energy Star refrigerator, how much will I be saving? And what's the performance increase of that?
I see a question about incorporating data from existing programs. So currently, as shown in Dave's slide, we have a certain number of databases that we're leveraging, one of which is DEER, and the database from Home Energy Saver that formed the nucleus or the kernel for the current database.
Now, clearly, we would like to expand as many sources as we can use. In terms of existing databases, we'd like to leverage whatever exists out there for — to improve the quality of our database quotes now. So we certainly encourage you to contact us. And hopefully we can discuss about how to leverage those going forward.
And with that, I'll turn it back over to Gail. Thank you very much.
So that's all the time we have today for questions. As Dave mentioned, if you have additional questions or your questions were not addressed, please go to the database webpage and submit them online.
We'd like to thank our speaker Dave and Noel and Sean for their time today. We'd also like to thank all of you for participating.
Please visit buildings.energy.gov/webinars to download a copy of the slides. Also, please check this webpage for information on future Building Technologies Program webinars. This concludes our presentation. Thank you, and goodbye.
[End of Audio]