MODELING, ESTIMATION, AND OPERATIONAL POLICIES FOR HUMAN PERFORMANCE WITH TEMPORAL MOTIVATION

Open Access
- Author:
- Kim, Ji Eun
- Graduate Program:
- Industrial Engineering
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- June 15, 2017
- Committee Members:
- David Nembhard, Dissertation Advisor/Co-Advisor
David Nembhard, Committee Chair/Co-Chair
Andris Freivalds, Committee Member
Gul E. Kremer, Committee Member
Stanley Gully, Outside Member - Keywords:
- Time-pressure reactivity
time management
human factors
temporal motivation - Abstract:
- In the contemporary workplace, companies, trainees, and workers are under increasing time pressure as organizations respond to broader competition, shorter staffing, and greater customer expectations. One observable time management device found in practice is the ubiquitous use of deadlines for both cognitive and manual work. Deadlines are known to increase productivity by helping companies and workers manage time efficiently, although many unaddressed questions remain regarding individual differences in performance, measurement, and appropriate deadline settings. For example, compared to individuals with longer deadlines, those who work with frequent and shorter deadlines exhibit longer work times and more completed tasks, while very short deadlines are known to reduce work performance and quality. Few studies have quantitatively modeled individualized pacing behavior and applied it to the workplace with a focus on conceptualization. The purpose of this dissertation is to model and estimate individual behavior related to deadlines, then to propose an improved design for time management policies. To do so, several experimental (an air-traffic control setting and online learning settings with eye-tracking measurements), field (Bayesian estimations to the course website data), and simulation (queueing simulations) investigations were conducted. The dissertation concludes with four major findings. First, the dissertation examines effective factors in generating models for aligning time pacing in the presence of deadlines, finding that task complexity and group size affect individual performance with temporal motivation in an air-traffic control setting. Second, the dissertation improves the quality of estimation for individual time-pressure reactivity by using a parametric Bayesian estimation approach. Third, it models the relationships between individual pacing with work productivity. Both the quality (e.g., GPA and class scores) and quantity (e.g., task completion time) of performance are found to be related to individual time pacing based on course website and eye-movement data that was analyzed using a structural equation model. Fourth and lastly, the dissertation proposes designs and policies for time management that improve productivity; prioritizing tasks via early due dates is recommended to increase individual productivity, based on the results from a queueing simulation. The result of this research contributes to generating policies or designing personalized learning by considering individual differences in time-pressure reactivity that had been previously ignored or oversimplified in their practical application.