Paper Title
APPLICATION OF MACHINE LEARNING METHODS TO IDENTIFYING EFFICIENT AND INEFFICIENT WORKERS

Abstract
For many manufacturing companies that still perform manual tasks such as hand assembly, improving worker performance remains crucial. As there is a distinct lack of research investigating the impact of worker aptitude on assembly operation performance, we conducted a series of experimental studies to examine this relationship. Our findings revealed that two-thirds of the variance in assembly times was determined by the workers themselves, with aptitude exerting a stronger influence than experience. It is therefore crucial to measure worker aptitude and allocate workers to roles based on it. This study uses assembly time as a quantitative measure of performance. Five Factor Personality Questionnaire (FFPQ) scores were used to measure worker aptitude. Data were collected from 249 workers, and machine learning methods such as decision trees and random forests were employed to identify efficient and inefficient workers based on their FFPQ scores. The aim of this study is to clarify the effectiveness of applying machine learning methods to identify efficient and inefficient workers and to provide practitioners with key insights to improve the effectiveness of assembly operations. Keywords - Workers’ Aptitude; Five Factor Model; FFPQ-50; Classification