Deadline: 13-Dec-22
The Water Research Foundation (WRF) is seeking applications for Quantifying the Impact of Artificial Intelligence/Machine Learning-Based Approaches on Utility Performance.
This project is funded by The Water Research Foundation (WRF) as part of WRF’s Research Priority Program.
Project Objectives
- Provide quantifiable results, real case studies, and an implementation demonstration study to gain insight into the capabilities of vendor-driven, academic-driven, or home-grown solutions for process performance evaluation.
- Evaluate artificial intelligence/machine learning (AI/ML) techniques and compare them with conventional alternatives such as heavily instrumented processes where the instruments can be expensive and hard to maintain.
- Demystify AI/ML to enable all utility staff to understand, trust, and embrace AI/ML via provable outcomes and value-driven key performance indicators (KPIs). Provide guidance on how to assess which AI/ML solutions are best for specific use cases.
- Identify impact types (e.g., efficiency, effectiveness, cost-saving, etc.) and how the benefits from the initial AI/ML implementation could endure with only minimal upkeep (i.e., how sustained are the initial impacts over the long run; what financial and effort investments may be required in the future to at least sustain the initial benefits).
- Identify and discuss risks and mitigation requirements that may apply when an AI/ML approach is used.
Research Approach
This RFP is intentionally flexible in the research approach to encourage creativity and originality from proposers. Proposers should describe how they will conduct the research to meet the objectives listed above. The following approach is intended as a starting point:
- Evaluate KPI baselines using historical data and then test AI/ML-based solutions ranging from general (such as “AI Solution in a Box” vendors) to specific (such as specific performance-based coagulation models) at one or more utilities.
- Determine data requirements and availability to ensure a positive outcome.
- Investigate long-term benefits of AI/ML-based solutions.
- Compare benefits with respect to cost and effort of implementation across multiple areas of operation and interest.
- Incorporate an AI/ML demonstration study to support evaluation protocols and utility benefits (required).
Funding Information
Applicants may request up to $350,000 in WRF funds for this project.
The anticipated period of performance for this project is 24 months from the contract start date.
Expected Deliverables
- A framework on the extent to which AI/ML can be used to support process optimization
- Protocol for effectively measuring and reporting the change in KPI when implementing AI/ML
- Demonstration/case studies with successes and pitfalls described, and technical discussion of the merits of general AI/ML approaches versus specific AI/ML approaches
- Recommendations for where AI/ML-based approaches could have positive impacts across water and wastewater utilities based on the results of the evaluation, including a ranking of utility areas that would benefit the most from AI/ML applications with the least level of implementation effort (e.g., predictive maintenance, operational response to disasters, overall utility health [staffing, financial, assets, knowledge], prioritization of competing strategic initiatives, etc.)
- Recommendations on how to consider risk and responsibility where AI/ML is either guiding or controlling processes that have regulatory requirements
Eligibility Criteria
Proposals will be accepted from domestic or international entities, including educational institutions, research organizations, governmental agencies, and consultants or other for-profit entities.
For more information, visit https://www.waterrf.org/open-rfps
