Implementation
IWS FRAMEWORK BUILDING BLOCKS
An IWS requires investment and advancement of many different building blocks that work together to help better manage sewersheds. This study presents the key building blocks in the proposed IWS framework and recommends approaches any utility can take within each building block to support a successful implementation of IWS. These building blocks belong to three main categories of the IWS framework – technical, organizational, and foundational. The categories and each of their building blocks are presented in the figure below.
The technical building blocks cover the important technological practices needed to advance the current practices in sewershed management. Technical building blocks help improve understanding of IWS definition and outcomes, sewershed system-of-systems, data collection, data management, modeling, cybersecurity decision centric and implementation aspects of sewershed management.
Organizational building blocks cover aspects that need to be addressed at the utility level to allow effective implementation of the technical and foundational building blocks. Specifically, these cover the development of strategic plans, governance structures, workforce improvement and communication strategies that can support implementation of IWS at the sewershed scale.
The foundational building blocks describe the core values and vision needed to ensure the right mindset and long-term sustainability of the IWS.
PILLARS OF AI FRAMEWORK FOR THE WATER SECTOR
Successful implementation of AI in the water sector requires the utilities to work on the 7 pillars. A brief description of the 7 pillars is provided below:
The first pillar of understanding AI and its benefits focuses on briefly explaining the different foundational concepts of AI like its main characteristics and techniques. These concepts are important for water managers and AI developers in utilities to consider. The information provided for this pillar is not exhaustive and interested readers may need to look at other literature for specific topics.
The second AI application goals pillar describes how AI should be applied and discusses topics like the application at different levels of the studied system, how to deploy AI in various categories, and the different modes of building AI.
The third pillar of data readiness describes how to evaluate the quality and quantity of collected data and how to preprocess to prepare AI-ready data.
The fourth pillar of knowledge integration explains how the knowledge in the minds of utility experts can be integrated into AI modeling frameworks to build more robust models.
The fifth pillar of model development discusses how to develop accurate and reliable AI models.
The sixth pillar of decision support explains methods to improve the trustworthiness of AI models and how to develop ‘human-in-the-loop’ models.
The seventh pillar on AI implementation explains methods for ensuring successful AI applications in the real-world and continual improvement of models.