All this talk about artificial intelligence (AI) is fine, but how do you actually make it useful to you? Yes, you can put your emails into ChatGPT or Anthropic and ask it to make the emails better, but what about something a little more impactful? In this presentation, after a brief review of the fundamentals of AI in public works, we will go through the process of selecting a real-world, public works problem that AI can address, and then take you through the steps of developing an AI tool to solve the specified problem. Choosing the problem in public works operations that is best served by an AI-based tool requires that you identify issues that are well addressed by data-driven decision-making, or that are clearly customer-service focused, or that are candidates for process automation. Sometimes, more than one of these areas may be involved in a project. This presentation will discuss selecting your “target” and indicate how the example we will use was chosen. The presentation will then take you step by step through the tool creation process. First, objectives must be clearly defined. Second, sources of data must be identified and, if needed, prepared to be ingested by the AI tool. Third, appropriate algorithms must be selected and coded. And finally, the whole tool must be rigorously tested and then deployed.

After this session, participants will be better able to:
• Review the fundamental concepts of AI and how it can be applied to solve common challenges in public works, such as infrastructure management, maintenance, and data querying.
• Identify specific problems within public works operations that AI can address, including data-driven decision-making, customer service, and process automation.
• Review insights into the step-by-step process of developing an AI tool, from defining objectives, gathering and preparing data, choosing algorithms, and testing and deploying the tool.

Contributor/Source

Eryn Stone-TSMO Engineer, Wilfrid Nixon-Presenter

Claim CEUs

Submit this form to claim CEUs.