The BPD Data Contribution Process.
BENEFITS OF CONTRIBUTING DATA
When contributors provide data to the BPD, Lawrence Berkeley National Laboratory (LBNL) cleanses the data set to remove errors and outliers, and provides you this cleansed version of your dataset. This also includes translation into the standard Building Energy Data Exchange Specification, identification of errors and outliers, and some basic analysis. Recognition as a data contributor is optional.
HOW TO CONTRIBUTE DATA
- Contact Us: Contact us at BPD@ee.doe.gov.
- Identify Data: Ensure that your data meets the minimum field requirements for contribution:
• Gross floor area
• Location information (i.e. postal code and/or ASHRAE climate zone)
• One complete year of energy consumption information
• Dates of energy data collection (either: start & end date, or start/end date & metering interval, etc.)
• Activity area information and/or building type
It will also be necessary to review and accept the BPD Data Privacy Agreement.
Building records can contain any data field defined by the Building Energy Data Exchange Specification. For those just beginning data collection, please refer to our BPD Priority Fields list, which enumerates fields that were found to have the greatest impact on energy consumption.
- Transfer Data: Send your data to the BPD team. Data can be contributed in any electronic format including a secure file transfer through a File Transfer Protocol website or email. Additionally, we can accept data through Energy Star Portfolio Manager account shares. The BPD accepts all electronic file formats, including .csv, .xls and .accdb. The data must be labeled “Proprietary Data” in the file name or email body. Please see our Data Contribution Best Practices guide for more details.
- Cleanse Data: To anonymize data prior to entry into the BPD, we strip it of all personally identifiable information such as owner’s name and street address.
Data is made anonymous to protect the privacy of data contributors and allow the sharing of data that would otherwise not have been made public, such as data collected to satisfy benchmarking and disclosure laws.
You may request a copy of your cleansed data set.
- Access the BPD: The BPD team will let you know when your data has been uploaded. The data is stored under stringent privacy and security guidelines.
No data is entered directly into the BPD. All data contributions are cleansed and mapped to a standard specification by the Lawrence Berkeley National Laboratory (LBNL) before entry to ensure data quality and robustness in the BPD. Where necessary, derived values are calculated before data is entered into the BPD. Our full data preparation procedures can be found here, and are summarized below.
- Before accepting data, information is gathered regarding the origin of the data to ensure that data is measured and that all building records correspond to real, operational buildings.
- The raw data fields are then translated into a standard data specification, the Building Energy Data Exchange Specification. This process is referred to as data mapping, and involves matching raw data field names with the schema field names, translating building and asset types to schema enumerated types, converting data types, parsing text descriptions, and other processes that encode raw data from providers into a uniform schema. This process ensures that every data field in the BPD holds the same significance for every building record, to enable one-to-one comparisons of data fields across buildings.
- Cleansing the data includes reviewing consistency within and across data fields. Cleansing is done using various “checks” to verify that data fields are within the realm of possibility based on high level building information, and that database records are internally consistent.
- The calculation of derived values involves assigning primary facility type, dominant equipment types and calculating energy totals from the data provided. The primary facility type, equipment types and energy totals are values that are used in the BPD tool algorithms and filtering.
Data is analyzed in aggregate to ensure that no one building will ever be able to be identified. Filters allow the user to narrow the peer group down to a minimum of 10 buildings below which no results are shown.
Data contributed to the BPD will never be exposed externally. It is protected by law from release under the Freedom of Information Act (FOIA), and will not be shared at the federal level internally. The Database’s security policies align with DOE policies and Information Security best-practices. The website itself employs Secure Sockets Layer (SSL) certificates with 2048-bit RSA encryption, which are internationally recognized as the standard for website security.
For more information, see the BPD Security document.