.png){kind=icon}
.png){kind=icon}
Return
Predicting human reliability based on probabilistic mission completion time using Bayesian Network
.png){kind=icon}
PRE
AI
DOI：10.1016/j.ress.2022.108324.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
This study considers the characteristics of a worker performing a sequence of tasks in a mission by developing a Bayesian Network model to predict reliability and mission completion time, the two measures of overall performance. The mission is broken down into tasks of different types, some of which may not be repeated back-to-back. A orker's initial task performance, learning, fatigue, and stress are the factors that affect the overall performance, and they vary by worker and task type'. Those characteristics and the task sequence plan are incorporated into a Bayesian Network to measure the performance of each task and, subsequently, the mission. Taking the task sequence plan into account adds a new dimension to the Bayesian Network as it counts the number of repetitions performed for each type of task, which allows linking the performance, learning, fatigue, and stress levels of a preceding task to a succeeding one. The developed model is general and can be applied to different real-life settings that are stressful and labour intensive. A numerical analysis is conducted to study how a worker's characteristics affect her/his reliability and the mission duration. The results are discussed, and managerial insights are presented.
Keywords:
Bayesian Network (BN)
Human reliability
Completion time
Task sequence
Human factor
Journal
R
IF:
11
Papers:
9.0K
Citations:
4.2W