Life-cycle Cost Analysis with Machine Learning

Beware of little expenses; a small leak will sink a great ship.

—— Benjamin Franklin

Project 1: Developing machine learning models for facility life-cycle cost analysis

In this project, the research team aimed to develop a comprehensive framework for creating machine learning models to predict facility life-cycle costs (LCC). We first conducted a literature review and a questionnaire survey to identify key variables affecting facility LCC. Utilizing this information, they proposed a structured approach for developing LCC analysis models using machine learning. This approach included data collection, model training, and evaluation steps. To demonstrate its practical application, we conducted a case study on Georgia Tech campus, developing predictive models for estimating the LCC of facilities. The study concluded that existing building systems contain sufficient data for effective LCC analysis using the proposed framework.

Project Team: Dr. Ray Gao and Dr. Pardis Pishdad‐Bozorgi

---