Order picking is one of the most critical and resource-intensive tasks in modern supply chains and warehouses. While previous research has concentrated on increasing productivity by introducing technological advancements or procedural improvements, managerial decision impacts on operator well-being have been insufficiently explored. This thesis places itself in the context of human-oriented warehousing and addresses how managerial decisions between two popular picking strategies, i.e., order picking and batch picking, affect both performance measures and human factors, such as physical, cognitive, perceptual, and psychosocial dimensions. In striving for this aim, a multi-dimensional approach was applied. Controlled laboratory experiments were conducted simulating a parts-to-picker setting, combining objective measurements of productivity and accuracy with biometric measurements (EEG, ECG) and validated subjective ratings (NASA-TLX survey). The experimental results were subjected to rigorous statistical analyses, such as paired t-tests, Wilcoxon tests and correlation, to identify significant differences between picking policies. The findings exhibit a clear trade-off between productivity and accuracy: batch picking improves throughput and lowers physical demand but raises errors and cognitive workload, while order picking provides greater accuracy and lower cognitive effort at the expense of lower productivity. In addition to these aggregate results, the analyses also point to significant individual heterogeneity. Complementary, cluster and simulation analyses were subsequently used to generalize the findings, exploring team-level configurations and task assignment strategies based on productivity, accuracy, preference, and perceived workload. The cluster analysis suggests that both perceived workload and the tendency to make mistakes are relatively stable individual traits. By contrast, alignment between task and personal preference statistically reduces perceived effort and improves accuracy. Simulation analyses illustrate that managerial policies reflecting operator preferences or perceived operators' workload can achieve more balanced performance than policies exclusively maximizing throughput or minimizing errors. This research thus offers a pilot example of integrating biometric and perceptual data in warehouse studies and illustrates the managerial importance of preference-sensitive and human factor–based task assignment. Finally, implications are applied to an actual case study, and suggestions for improvement in warehouse operations are provided.
L’attività di picking (ricerca e prelievo degli oggetti in magazzino per soddisfare gli ordini del cliente) è una delle più critiche e dispendiose in termini di risorse nelle moderne supply chains. Mentre le precedenti ricerche si sono concentrate prevalentemente sull'aumento della produttività attraverso l'introduzione di progressi tecnologici o miglioramenti procedurali, l'impatto delle decisioni manageriali sul benessere degli operatori non è stato ancora sufficientemente esplorato. Questa tesi si colloca nel contesto di un magazzino orientato all'uomo e analizza come la scelta tra le due più diffuse strategie di picking – prelievo a singolo ordine, single order picking, o a lotti di ordini, batch picking - influenzino sia le misure di performance sia i fattori umani, come le dimensioni fisiche, cognitive, percettive e psicosociali. Per raggiungere questo obiettivo, è stato adottato un approccio multidimensionale: esperimenti in ambiente controllato sono stati condotti simulando un contesto "parts-to-picker", combinando misurazioni oggettive di produttività e accuratezza con misurazioni biometriche (EEG, ECG) e valutazioni soggettive validate (sondaggio NASA-TLX). I risultati sperimentali sono stati poi sottoposti a rigorose analisi statistiche, come t-test appaiati, test di Wilcoxon e analisi di correlazione, al fine di identificare differenze significative tra le politiche di picking. Analisi complementari di cluster e simulazioni sono state successivamente utilizzate per generalizzare i risultati, esplorando configurazioni a livello di team e strategie di assegnazione delle attività basate su produttività, accuratezza, preferenze e carico di lavoro percepito. I risultati mostrano un chiaro trade-off tra produttività e accuratezza: il batch picking migliora la produttività e riduce la domanda fisica, ma aumenta gli errori e il carico di lavoro cognitivo, mentre l’order picking offre una maggiore precisione e un minore sforzo cognitivo a scapito di una minore produttività. I risultati dei cluster indicano che il carico di lavoro percepito e la propensione a commettere errori rappresentano tratti individuali relativamente stabili, mentre l'allineamento tra preferenza dell’operatore e task svolta riduce lo sforzo percepito e migliora la preisione. Le analisi di simulazione illustrano inoltre che le politiche gestionali che riflettono le preferenze degli operatori o il carico di lavoro percepito dagli operatori possono raggiungere prestazioni più equilibrate rispetto alle politiche che massimizzano esclusivamente la produttività o minimizzano gli errori. Questa ricerca offre quindi un esempio pilota di integrazione di dati biometrici e percettivi negli studi sulle attività di magazzino e illustra l'importanza manageriale di un'assegnazione dei compiti basata sulle preferenze e sul fattore umano. Infine, i risultati vengono testati su un caso di studio reale e vengono forniti suggerimenti per migliorare le operazioni di magazzino.
Integrating human factors into managerial decision making in order picking: the role of personal characteristics and preferences
MARCIANO, MATTEO;Tardugno, Flavio
2024/2025
Abstract
Order picking is one of the most critical and resource-intensive tasks in modern supply chains and warehouses. While previous research has concentrated on increasing productivity by introducing technological advancements or procedural improvements, managerial decision impacts on operator well-being have been insufficiently explored. This thesis places itself in the context of human-oriented warehousing and addresses how managerial decisions between two popular picking strategies, i.e., order picking and batch picking, affect both performance measures and human factors, such as physical, cognitive, perceptual, and psychosocial dimensions. In striving for this aim, a multi-dimensional approach was applied. Controlled laboratory experiments were conducted simulating a parts-to-picker setting, combining objective measurements of productivity and accuracy with biometric measurements (EEG, ECG) and validated subjective ratings (NASA-TLX survey). The experimental results were subjected to rigorous statistical analyses, such as paired t-tests, Wilcoxon tests and correlation, to identify significant differences between picking policies. The findings exhibit a clear trade-off between productivity and accuracy: batch picking improves throughput and lowers physical demand but raises errors and cognitive workload, while order picking provides greater accuracy and lower cognitive effort at the expense of lower productivity. In addition to these aggregate results, the analyses also point to significant individual heterogeneity. Complementary, cluster and simulation analyses were subsequently used to generalize the findings, exploring team-level configurations and task assignment strategies based on productivity, accuracy, preference, and perceived workload. The cluster analysis suggests that both perceived workload and the tendency to make mistakes are relatively stable individual traits. By contrast, alignment between task and personal preference statistically reduces perceived effort and improves accuracy. Simulation analyses illustrate that managerial policies reflecting operator preferences or perceived operators' workload can achieve more balanced performance than policies exclusively maximizing throughput or minimizing errors. This research thus offers a pilot example of integrating biometric and perceptual data in warehouse studies and illustrates the managerial importance of preference-sensitive and human factor–based task assignment. Finally, implications are applied to an actual case study, and suggestions for improvement in warehouse operations are provided.| File | Dimensione | Formato | |
|---|---|---|---|
|
2025_10_Marciano_Tardugno_Executive_Summary.pdf
accessibile in internet per tutti
Descrizione: Executive summary
Dimensione
890.49 kB
Formato
Adobe PDF
|
890.49 kB | Adobe PDF | Visualizza/Apri |
|
2025_10_Marciano_Tardugno_Tesi.pdf
accessibile in internet per tutti
Descrizione: Tesi
Dimensione
5.28 MB
Formato
Adobe PDF
|
5.28 MB | Adobe PDF | Visualizza/Apri |
I documenti in POLITesi sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.
https://hdl.handle.net/10589/243330