(110d) How to Implement AI in a Data Lean Environment to Deliver Business Value.

The promise of AI used to be contingent on the need for big volumes of data. However, AI is not as data-hungry as once thought. New techniques in industrial AI and expert practices in data pre-processing combined with minimum good quality data are proven to deliver true benefits. In this presentation, Canvass AI will discuss the key data requirements that engineers need to get started with AI to derive real value. We’ll explore how to assess data sufficiency, including quantity, quality, and descriptiveness. In addition, we’ll address how to integrate experience into the data, particularly as it regards the institutionalization of knowledge and advanced enablement.

This presentation includes real-world case studies of how chemical engineers are extracting value from their existing data to simplify their decision-making and fast-track their day-to-day problem-solving. The presentation will include a deep dive into one of the case studies to share how data and model building steps, and specialized Industrial AI capabilities can analyze process dynamics and narrow down the problem size to create AI models that are highly accurate and reliable.