Turn Your Data Into a Crystal Ball: How Big Data Analytics Can Target Energy Savings Opportunities as Well as Predict Impending System Failures to Achieve Improved Energy Efficiency, Comfort, and Sustainability at Laboratory Facilities

Julianne Rhoads, Cimetrics Inc.

Fault detection and root cause analysis using big data provide a strategic approach to energy reduction and ongoing maintenance in high-performance laboratory buildings. Insidious HVAC faults are often superseded by reactive maintenance. By analyzing building data, large scale operational issues can be mitigated and persistent alarms can be minimized. Furthermore, predictive fault detection algorithms identify potential faults and equipment failures before they occur, allowing for predictive rather than reactive maintenance. By using state-of-the-art big data approaches that apply predictive algorithms to system models, the effectiveness of predictive maintenance is greatly increased. The economic impact associated with these issues can be used to quantify building performance improvement potential. BMS data from representative healthcare, pharmaceutical, and university laboratory facilities will be used to highlight the procedure for fault detection and predictive analysis. This presentation will discuss typical high-value controls, mechanical, and operational faults while stressing the benefits of utilizing big data for predictive maintenance such as increased equipment life, improved reliability and lower labor cost. Using aggregated facility data, I will explore fault identification and failure prediction, root cause analysis, and issue remediation. The presentation will then focus on quantifying the energy savings that result from the appropriate corrective actions. Lastly, I will cover the impact of data quality and reliability on our outcomes. Through the examples provided, this presentation will demonstrate how a methodical approach to BMS analysis and design can result in high-functioning, energy efficient lab building operation.

Learning Objectives

  • Understand how to evaluate big building management data as a tool for identifying faults;
  • Understand how control strategy optimization can counter reactive maintenance;
  • Identify the economic and environmental benefits of big data analysis; and
  • Recognize and appreciate high-value data points and reliable data collection.


Julianne Rhoads joined Cimetrics in 2017 and is responsible for energy analysis and reporting on more than 35 buildings, including over 4.5 million square feet of facilities in the healthcare, higher education, federal, and pharmaceutical research sectors throughout the United States. She has identified and helped to implement more than $4 million in annual energy savings. Prior to joining Cimetrics, Ms. Rhoads was responsible for designing Energy Performance Contracting projects at Siemens.


Note: I2SL did not edit or revise abstract or biography text. Abstracts and biographies are displayed as submitted by the author(s).