Title:
Causality Versus Correlation in the Ready-Mix Industry Through use of IOT Systems
Author(s):
Tregger
Publication:
Web Session
Volume:
ws_F23_Tregger.pdf
Issue:
Appears on pages(s):
Keywords:
DOI:
Date:
10/29/2023
Abstract:
Artificial intelligence and machine learning have permeated nearly every industry from medical insurance to art. The concrete industry is no exception. Concrete lends itself well to machine learning techniques as there is a tremendous amount of variables and interacting factors affecting the performance. This poses a real challenge in developing tools that help at the practical level. More specifically, the number of factors makes it hard to determine correlation versus causation, a deficiency of most machine learning algorithms that is often under-appreciated. This talk will demonstrate methods to determine causal factors of performance issues faced by the ready-mix producer. This allows practical solutions to be taken instead of guessing at solutions that at the end, may just be correlated instead of causal. Real-life examples will be presented.