Yazar "Gunduz, Tulin" seçeneğine göre listele

Listeleniyor 1 - 2 / 2
  • Küçük Resim Yok
    Öğe
    A proactive approach to ergonomics: an evaluation of multi-method risk assessment in warehouse operations
    (Taylor & Francis Ltd, 2026) Ugurtay, Oguzhan; Korkmaz, Sonnur; Dalfidan, Derya; Gunduz, Tulin
    The proactive use of ergonomic assessment methods enhances worker performance and health while improving task quality and reducing operational costs through human-centred work design. This study was conducted in a logistics warehouse, where 197 tasks across nine locations were evaluated using REBA, RULA, BAuA, and LMM-ZS. Data were collected through on-site observations and task-specific video recordings. Risk scores were normalised to a 0-100 scale to enable cross-method comparison. Multi-method evaluations create two main challenges: selecting the most suitable method for each task and integrating outcomes derived from incompatible scales. To address these issues, a harmonised framework is proposed that standardises score ranges and categorical zones across the four methods.
  • Küçük Resim Yok
    Öğe
    The effect of cognitive emotional states on physiological productivity
    (Taylor & Francis Ltd, 2025) Dalfidan, Derya D.; Gunduz, Tulin
    Emotional states are fundamental attributes distinguishing humans from machines, and productivity represents one of the primary life objectives for this emotionally driven being. However, existing research on productivity and job performance frequently underestimates the impact of underlying emotional mechanisms. Thus, a systematic examination of the emotion-productivity interface is essential to clarify the psychophysiological processes that regulate work efficiency. In this study, emotional induction was achieved through a curated video stimulus set designed to evoke positive (happiness) and negative (sadness) responses in 39 participants, followed by a computer-based Stroop task. Electroencephalography (EEG) was employed to capture emotional states within the two-dimensional valence-arousal framework. During task performance, parameters related to productivity metrics were recorded. Three machine learning models - Artificial Neural Networks (ANN), Support Vector Machines (SVM), and Random Forest (RF) - were implemented to predict productivity levels. For positive emotions, mean absolute error (MAE) values were 0.1031 (ANN), 0.0760 (SVM), and 0.0721 (RF). For negative emotions, the values were 0.1165, 0.0902, and 0.0659, respectively. Results demonstrated that productivity levels increased during tasks performed after the induction of positive emotions. Overall, this study provides empirical evidence that productivity can be predicted from emotional states, emphasizing their integral role in cognitive processes and their potential utility for optimizing workplace performance.