About

The Water Research Foundation is currently accepting proposals for its project namely Data Management Best Practices: Integrating Data Sources for Treatment Optimization and Efficiency.

Objectives

• Review and compile case studies of tools that utilities use to combine data sets, organize and validate data, and maintain data security before carrying out treatment optimization efficiency projects.

• Create a best practice guide for utilities wanting to integrate their datasets to prepare for Machine Learning (ML) Artificial Intelligence (AI) based treatment optimization efficiency projects. 

Research Approach

• The precise research approach will be determined by the selected research team, but the following key aspects should be considered by the researchers:

  • Conduct a literature review, including ML AI work, to document the software and tools being used to organize data across multiple systems (SCADA, CMMS, financial, etc.), validate data, and maintain data security. Tools can include programming languages used to combine information from multiple databases into a single source, in addition to data development for convenient machine reading.

  • Interview survey utilities that are combining and organizing datasets across multiple platforms, validating data, and maintaining data security before carrying out treatment optimization efficiency projects.

  • Inventory and characterize the quality reliability of typical utility data sets (SCADA data historian, customer billing, capital assets such as service line material, LIMS, etc.) and expected level of effort to prepare the data sets for ML, AI, or reliable data mining.

  • Identify which data sets might be of greatest use in the future so utilities can prioritize investments in improving data quality reliability.

  • Conduct a workshop with utilities to discuss tools and document their challenges and successes.

  • Compile a best practice guide for utilities to use as a reference for successfully combining and organizing datasets across multiple sources in a way that can be used for ML AI applications.

Funding Information

• The maximum funding available from WRF for this project is $250,000. The applicant must contribute additional resources equivalent to at least 33% of the project award. For example, if an applicant requests $100,000 from WRF, the applicant must contribute an additional $33,000 or more. Acceptable forms of applicant contribution include cost share, applicant in-kind, or third-party in-kind that comply with 2 CFR Part 200.306 cost sharing or matching. The applicant may elect to contribute more than 33% to the project, but the maximum WRF funding available remains fixed at $250,000.

• The anticipated period of performance for this project is 24 months from the contract start date

Expected Deliverables

• Deliverables from this project include:

  • Literature review of existing methods for data collection, validation, and security

  • Case studies report based on utility interviews with a minimum of five case studies (one case study per utility interviewed)

  • Data quality and reliability review

  • Utility workshop materials and summary

  • Best practice guide based on previous research activities and deliverables

  • Utility report-out video

Eligibility

• Proposals will be accepted from both U.S.-based and non-U.S.-based entities, including educational institutions, research organizations, governmental agencies, and consultants or other for-profit entities.

• WRF’s Board of Directors has established a Timeliness Policy that addresses researcher adherence to the project schedule. Researchers who are late on any ongoing WRF-sponsored studies without approved no-cost extensions are not eligible to be named participants in any proposals.

Post Date: September 30, 2024