Improving Work Performance Through Mobile Sensor Technology

Posted: July 26, 2017
ECE Professor Emre Ertin
Mobile sensors can already monitor a person's health and wellness levels, but the same technology may now be applied to help improve performance and productivity in the workplace.

A University of Memphis-led six-university team, will develop and test a system of mobile sensors and software, called mPerf, to objectively assess everyday job performance. The $13.8 million project is sponsored by the Intelligence Advanced Research Projects Activity (IARPA)’s Multimodal Objective Sensing to Assess Individuals with Context (MOSAIC) program.

Ohio State University Electrical and Computer Engineering Professor Emre Ertin will develop wearable sensors for continuous monitoring of markers of activity and behavior relevant to performance at the work place. Approximately $1.3 million of the overall funding will support work performed by Ertin and his team.

Current workforce evaluation tools, such as interviews, cognitive assessments and questionnaires, do not always capture how an individual performs on a day-to-day basis. To address this challenge, mPerf will build upon an open-source software platform developed by the NIH-supported Center of Excellence for Mobile Sensor Data-to-Knowledge (MD2K), also headquartered at the University of Memphis. This platform allows researchers to gather, analyze and store high-frequency mobile sensor data to discover and validate mHealth biomarkers. mPerf will extend this platform to model and predict work performance based on passively collected sensor-based markers of activity, behavior, and context.

The project is directed by University of Memphis Professor Santosh Kumar, Lillian and Morrie Moss Chair of Excellence in Computer Science. Kumar, widely known as one of the nation’s leading scientists in mobile health, leads a team including some of the nation’s top researchers in work performance (Deniz Ones, Minnesota), sensor design and signal processing (Ohio State’s Ertin), interpersonal communications (Eugene Buder, Memphis), stress (Mustafa al’Absi, Minnesota), mobile sensing (Tanzeem Choudhury, Cornell), mobile computing (Deepak Ganesan, UMass Amherst and Mani Srivastava, UCLA), and machine learning (Benjamin Marlin, UMass Amherst).

Ertin is also a key research partner in the Memphis-led MD2K consortium. Three of his biosensors contribute prominently to MD2K’s work to apply new technologies to reduce congestive heart failure patient hospital readmissions and prevent relapse among people who have quit smoking.

“Through MD2K, we have already developed many novel ways to monitor health and wellness using mobile sensors,” Kumar said. “The mPerf project allows us to expand MD2K’s offerings to help assess work performance and productivity using the same mobile sensors.”

To develop and evaluate its models, the mPerf team will collect data from 600 employees at 5 to 10 different organizations in the U.S. and abroad. Leveraging their decade-long experience in this field, mPerf researchers will develop unique sensor-based markers. They will then apply novel sensor data analytics to create a library of sensor-based indicators to measure work performance

“This unique project will add a new dimension to our sensor research on unobtrusive monitoring of human behavior and activity” Ertin said. “Our team will employ wearable sensors developed at Ohio State to unravel major indicators of performance and productivity at workplace.”

Find out more about the mPerf project here:

About IARPA’s MOSAIC program:

MOSAIC is a new program looking for new ways to measure and predict individual job performance using unobtrusive, passive and persistent sensor-based measurements. Its goal is to improve the Intelligence Community’s capabilities to evaluate its workforce throughout their careers. For more information, go to

About MD2K:

The MD2K Center at the University of Memphis is one of 13 national Big Data Centers of Excellence awarded by the National Institutes of Health as part of its Big Data-to-Knowledge initiatives. MD2K’s goal is to generate generalizable theory, methods, tools, and software to address major barriers to processing complex mobile sensor data. MD2K brings together top minds in computer science, engineering, medicine, behavioral science and statistics, drawn from 13 universities (Cornell Tech, Georgia Tech, Harvard, Northwestern, Ohio State, UCLA, UC San DIego, UC San Francisco, the University of Massachusetts Amherst, the University of Memphis, the University of Michigan, the University of Utah and West Virginia University). To learn more about MD2K, go to