Total Cost of Ownership Prediction in Chilled Water Plants: Contributing Factors and Role of Artificial Intelligence

This study investigates key parameters and applications of artificial intelligence (AI) in predicting the total cost of ownership (TCO) for chilled water plants (CWPs). Forecasting the TCO of CWPs is challenging due to the diverse and dynamic factors and parameters that influence it, necessitating understanding their complex correlations and causations. While AI and non-AI approaches have improved parameter prediction accuracy in different engineering applications, comprehensive literature reviews on chiller TCO prediction methodologies and their influencing factors are limited. This systematic review addresses three objectives: (1) to identify the key parameters in estimating TCO of CWPs, (2) to examine the existing techniques employed in TCO forecasting and their benefits in energy and cost savings, and (3) to evaluate how AI enhances TCO prediction accuracy and robustness. Following the preferred reporting items for systematic reviews and meta-analyses (PRISMA) guidelines, this review analyzed studies from 2017 to 2024 sourced from the Web of Science and Scopus databases. This study identifies several key parameters influencing TCO, including cooling load, energy consumption, chiller capacity, and the Coefficient of Performance (COP). The review shows that AI-driven models, such as deep learning and machine learning algorithms, have improved the accuracy and robustness of TCO predictions, and it further demonstrates scenarios where AI outperforms conventional prediction and forecasting methods. Notably, the current review shows that AI techniques are predicted to be capable of reducing total life cycle costs by up to 18%, based on modeling estimates.