Technologies for accident prevention in forklift operations and safety in human–machine interaction: a scoping review

Abstract

Introduction  Accidents involving machinery and forklifts are a major cause of worker fatalities. The Fourth Industrial Revolution and Health 4.0, with technologies such as wearables and the Internet of Things, offer solutions to prevent accidents.

Objective  To map scientific evidence on the implementation of collision-prevention technologies in forklift operations, with a focus on reducing accidents and promoting a safe work environment.

Methods  A scoping review was conducted with searches of 11 databases. The research protocol was registered on the Open Science Framework (OSF) (). Articles available in full text, in any language, and without time restrictions were included.

Results  The initial search yielded 390 articles, of which eight were selected for this review. The main approaches analyzed included the Industrial Internet of Things, computer vision, automation, 3D image capture, and data analysis.

Conclusion  The studies indicate that the adoption of these technologies in industrial settings represents a significant advance in creating safer working conditions, with great potential to shape the future of occupational safety.

Keywords
Wearable Electronic Devices; Occupational Health; Industrial and Manufacturing Facilities; Workplace Safety; Accident Prevention

Resumo

Introdução  Os acidentes com máquinas e empilhadeiras são uma importante causa de morte de trabalhadores. A Quarta Revolução Industrial e a Saúde 4.0, com tecnologias como wearables e Internet das Coisas, oferecem soluções para prevenir acidentes.

Objetivo  Mapear evidências científicas sobre a implementação de tecnologias de prevenção de colisões em operações com empilhadeiras, com foco na redução de acidentes, na promoção de um ambiente de trabalho seguro.

Métodos  Revisão de escopo baseada em buscas em 11 bases de dados. O protocolo de pesquisa foi registrado no Open Science Framework (OSF) (). Foram incluídos artigos disponíveis na íntegra, em qualquer idioma e sem restrição temporal.

Resultados  A busca inicial retornou 390 artigos, dos quais oito foram selecionados para compor esta revisão. As principais abordagens analisadas incluíram Internet Industrial das Coisas, visão computacional, automação, captura de imagens 3D e análise de dados.

Conclusão  Os estudos indicam que a adoção dessas tecnologias nos ambientes industriais representa um avanço importante na criação de condições de trabalho mais seguras, com um grande potencial para moldar o futuro da segurança ocupacional.

Palavras-chave
Dispositivos Eletrônicos Vestíveis; Saúde do Trabalhador; Instalações Industriais e de Manufatura; Segurança no Trabalho; Prevenção de Acidentes

Introduction

Workplace safety is viewed as a continuous and dynamic process that goes beyond the prevention of failures and accidents, focusing on the adaptability and resilience of the work system. According to Hollnagel¹, this model emphasizes the importance of understanding how processes function well most of the time and how workers and organizations can adapt and respond to unforeseen events. This proactive approach to safety is particularly relevant given the alarming numbers of accidents and fatalities in the workplace¹.

According to the International Labour Organization (ILO), approximately 330,000 deaths from workplace accidents occur worldwide each year². In 2022, in Brazil, the Observatory of Occupational Safety and Health reported that 612,900 work-related accident reports (CAT) were recorded, including 2,500 deaths. The state of São Paulo accounted for approximately 34.6% of these incidents, followed by Minas Gerais (10.8%), Rio Grande do Sul (8.56%), and Santa Catarina (7.93%)³.

From 2012 to 2021, in Brazil, a large proportion of workplace accidents (15%) occurred during operations involving machinery and equipment. In 2021, this percentage remained high, reaching 16% of the total⁴. According to the Occupational Safety and Health Administration (OSHA)⁵ of the United States of America (USA), one in six fatal workplace accidents in that country involves forklifts, and in 80% of these cases, pedestrians are involved⁵. In 2024, OSHA estimated that between 35,000 and 62,000 forklift-related injuries occur annually in the USA, resulting in approximately 75 to 100 worker deaths, with an average of 87 fatalities per year⁶.

In Brazil, in 2024, 2,635 accidents involving forklifts were reported, 347 of which occurred in the state of Santa Catarina³. That same year, Brazil recorded six forklift-related fatalities: three in Santa Catarina, one in Bahia, one in Goiás, and one in Minas Gerais. In the same year, the fatality rate from forklift accidents was 1.85% in Bahia, 1.33% in Goiás, and 0.82% in Santa Catarina⁷. Therefore, accidents involving forklifts are frequent and can result in serious injuries and fatalities, raising constant concerns regarding occupational safety and health.

In recent years, there has been a transformation in work and communication methods, known as the Fourth Industrial Revolution (FIR). This new paradigm is characterized by the adoption of processes that utilize machines powered by artificial intelligence⁸. Digital transformation is a global phenomenon that has redefined culture, structures, and socioeconomic systems, generating significant impacts on both society and businesses⁹.

Industry 4.0 represents a strategy focused on the development of advanced and automated manufacturing systems, marking a new stage in the manufacturing industry. Its goal is to drive the evolution of smart industry using data and digital technologies¹⁰. This concept of digital transformation, based on automation and data integration, can be extended to areas beyond manufacturing, such as healthcare. Health 4.0 applies the advancements of Industry 4.0 in institutional and social settings, using the same digital technologies to improve and preserve the population’s health¹¹.

Health 4.0 drives significant transformations in the medical, sociological, and psychological fields, reflecting the innovations of FIR¹². Through this technological integration, it enables the identification of improvements and promotes more informed decisions, facilitating the transition from a reactive, fee-for-service model to a system focused on value, outcomes, and proactive prevention¹¹.

Among the technologies that have emerged in this context, the potential role and application of wearables as solutions aimed at the safety and health of workers in various sectors stand out¹³. Wearables are technological devices that can be worn on the human body, which marked the beginning of Health 3.0, designed to fit the body and allow for the practical performance of daily activities. These devices offer a variety of features, integrating technology into daily life in an intuitive and efficient manner¹⁴, ¹⁵.

With the advancement of the Internet of Things (IoT), entire value chains are being reshaped by digital technology, either gradually or through disruptive changes⁹. The function of the Industrial Internet of Things (IIoT) is to monitor, collect, process, and analyze data¹⁶, enabling devices to adjust their behavior autonomously or instruct other devices to do the same, without the need for human intervention¹⁷,¹⁸.

Computer vision, a subfield of artificial intelligence and machine learning, uses computational systems to analyze and interpret visual data in detail¹⁹. This technology enables the detection and monitoring of risks in real time by recognizing objects and identifying dangerous situations early on. Furthermore, its ability to process large volumes of visual data enables continuous, automated surveillance, especially in industrial and high-risk environments²⁰,²¹.

Accidents involving forklifts are classified as occupational hazards, with potentially serious consequences for workers⁵,⁶. Although collision-prevention technologies are on the rise, scientific evidence regarding their effectiveness and implementation remains scattered throughout the literature. Mapping this knowledge is therefore important for guiding evidence-based decisions, as well as for identifying gaps that require future research.

Therefore, the objective of this article is to map scientific evidence on the implementation of collision-prevention technologies in forklift operations, with a focus on reducing accidents and promoting a safe work environment.

Methods

Protocol and Registration

This is a scoping review that followed the steps recommended by the Joanna Briggs Institute (JBI)²² and was reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses: Extension for Scoping Reviews (PRISMA-ScR)²³ guidelines. This type of review seeks to explore the main concepts of the topic in question, assessing the scope, reach, and nature of the research, synthesizing and publishing the data, and thereby identifying gaps in the existing research. The research protocol was registered with the Open Science Framework (OSF) (doi: 10.17605/OSF.IO/8K39X and URL: https://osf.io/8k39x/).

Eligibility Criteria

Quantitative, qualitative, and mixed-methods primary studies were eligible for inclusion in this review, as well as secondary studies²³, such as systematic, scoping, integrative, narrative, and literature reviews, provided they were available in full in any language, with no time restrictions. Articles that did not meet the established criteria for the objective and research question were excluded, as were gray literature and those not available in full, studies in the planning phase, or those without results.

Information sources

The following databases were consulted: Nursing Database (BDENF), Cumulative Index to Nursing and Allied Health Literature (CINAHL), Compendex, Computers and Applied Sciences, Embase, IEEE Xplore Digital Library, Latin American and Caribbean Health Sciences Literature (LILACS), United States National Library of Medicine/Medical Literature Analysis and Retrieval System Online (PubMed/MEDLINE), Scopus, Scientific Electronic Library Online (SciELO), and Web of Science. These databases were selected for their comprehensiveness and broad coverage in the fields of engineering, technology, health, and safety.

Search

To support the development of the search strategy, the Population, Concept, and Context (PCC) mnemonic²² was employed. The study population was defined as workers who operate forklifts; the concept refers to the implementation of collision-prevention technologies; and the context encompasses the industrial environment where forklifts are used. The search strategy was designed using the Boolean operators AND and OR and combinations of terms from the Health Sciences Descriptors (DeCS) and Medical Subject Headings (MeSH) in all languages, as described in Table 1. No time or language filters were applied, as this would have reduced the number of articles retrieved from the databases. The search was conducted in January 2025.

Table 1
Search strategy for the scoping review

Database	Search Strategy
BDENF	(Forklift* OR “Carretilla Elevadora” OR “Carretillas elevadoras” OR Forklift* OR “Lift Truck” OR “Lift Trucks”) AND (“Technology” OR Tecnologi* OR “Technological Development” OR “Digital Technology” OR “Biomedical Technology” OR “Assistive Technology” OR “WearableElectronicDevices” OR “WearableComputer” OR “WearableElectronicDevice” OR “WearableDevice” OR “WearableDevices” OR “TechnologicalDevelopment” OR “SmartPhone” OR “Self-HelpDevices” OR “WearableElectronicDevices” OR “Wearable Computers” OR “Wearable Electronic Device” OR “Wearable Electronic Device” OR “Wearable Device” OR “Wearable Device” OR “Wearable Electronic Devices” OR “Wearable Devices” OR “Wearable Devices” OR “Technology” OR “Technolog*” OR “Technological Development” OR “Digital Technology” OR “Biomedical Technology” OR ”Self-Help Devices” OR ”Wearable Electronic Devices” OR ”Wearable Device” OR ”Wearable Devices” OR ”Wearable Computer” OR ”Wearable Computers” OR ”Wearable Device” OR ”Wearable Devices” OR ”Wearable Electronic Device”)AND(“Safety” OR “Industrial Safety” OR “Safety Management” OR “Accident Prevention” OR “Seguridad” OR “Seguridad Industrial” OR “Administración de la Seguridad” OR “Prevención de “OR”Safety” OR ”IndustrialSafety” OR ”SafetyManagement” OR ”AccidentPrevention”ORPreven*)
Compendex	(Forklift* OR “Lift Truck” OR “Lift Trucks”) AND (“Technology” OR Technolog* OR “Technological Development” OR “Digital Technology” OR “Biomedical Technology” OR “Self-Help Devices” OR “Wearable Electronic Devices” OR “WearableDevice” OR “WearableDevices” OR “WearableComputer” OR “WearableComputers” OR “WearableDevice” OR “WearableDevices” OR “WearableElectronicDevice”) AND (“Safety” OR “IndustrialSafety” OR “SafetyManagement” OR “AccidentPrevention” OR Preven*)
Computers & Applied Sciences Complete	(Forklift* OR “LiftTruck” OR “LiftTrucks”) AND (“Technology” OR Technolog* OR “TechnologicalDevelopment” OR “DigitalTechnology” OR “BiomedicalTechnology” OR “Self-HelpDevices” OR “WearableElectronicDevices” OR “WearableDevice” OR “WearableDevices” OR “WearableComputer” OR “WearableComputers” OR “WearableDevice” OR “WearableDevices” OR “WearableElectronicDevice”) AND (“Safety” OR “IndustrialSafety” OR “SafetyManagement” OR “AccidentPrevention” OR Preven*)
CINAHL	(Forklift* OR “LiftTruck” OR “LiftTrucks”) AND (“Technology” OR Technolog* OR “TechnologicalDevelopment” OR “DigitalTechnology” OR “BiomedicalTechnology” OR “Self-HelpDevices” OR “WearableElectronicDevices” OR “WearableDevice” OR “Wearab leDevices” OR “WearableComputer” OR “WearableComputers” OR “WearableDevice” OR “WearableDevices” OR “WearableElectronicDevice”) AND (“Safety” OR “IndustrialSafety” OR “SafetyManagement” OR “AccidentPrevention” OR “Preven*”)
Embase	(Forklift* OR “LiftTruck” OR “LiftTrucks”) AND (“Technology” OR Technolog* OR “TechnologicalDevelopment” OR “DigitalTechnology” OR “BiomedicalTechnology” OR “Self-HelpDevices” OR “WearableElectronicDevices” OR “WearableDevice” OR “WearableDevices” OR “WearableComputer” OR “WearableComputers” OR “WearableDevice” OR “WearableDevices” OR “WearableElectronicDevice”) AND (“Safety” OR “IndustrialSafety” OR “SafetyManagement” OR “AccidentPrevention” OR Preven*)
IEEE Xplore	(Forklift* OR “LiftTruck” OR “LiftTrucks”) AND (“Technology” OR Technolog* OR “TechnologicalDevelopment” OR “DigitalTechnology” OR “BiomedicalTechnology” OR “Self-HelpDevices” OR “WearableElectronicDevices” OR “WearableDevice” OR “WearableDevices” OR “WearableComputer” OR “WearableComputers” OR “WearableDevice” OR “WearableDevices” OR “WearableElectronicDevice”) AND (“Safety” OR “IndustrialSafety” OR “SafetyManagement” OR “AccidentPrevention” OR Preven*)
LILACS	(Forklift* OR “Carretilla Elevadora” OR “Carretillas elevadoras” OR Forklift* OR “Lift Truck” OR “Lift Trucks”) AND (“Technology” OR Tecnologi* OR “Technological Development” OR “Digital Technology” OR “Biomedical Technology” OR “Assistive Technology” OR “WearableElectronicDevices” OR “WearableComputer” OR “WearableElectronicDevice” OR “WearableDevice” OR “WearableDevices” OR “TechnologicalDevelopment” OR “SmartPhone” OR “Self-HelpDevices” OR “WearableElectronicDevices” OR “Wearable Computers” OR “Wearable Electronic Device” OR “Wearable Electronic Device” OR “Wearable Device” OR “Wearable Device” OR “Wearable Electronic Devices” OR “Wearable Devices” OR “Wearable Devices” OR “Technology” OR “Technolog*” OR “Technological Development” OR “Digital Technology” OR “Biomedical Technology” OR ”Self-Help Devices” OR ”Wearable Electronic Devices” OR ”Wearable Device” OR ”Wearable Devices” OR ”Wearable Computer” OR ”Wearable Computers” OR ”Wearable Device” OR ”Wearable Devices” OR ”Wearable Electronic Device”)AND(“Safety” OR “Industrial Safety” OR “Safety Management” OR “Accident Prevention” OR “Seguridad” OR “Seguridad Industrial” OR “Administración de la Seguridad” OR “Prevención de “OR”Safety” OR ”Industrial Safety” OR ”Safety Management” OR ”Accident Prevention”ORPreven*)
PubMed/MEDLINE	(Forklift* OR “Lift Truck” OR “Lift Trucks”) AND (“Technology” OR Technolog* OR “Technological Development” OR “Digital Technology” OR “Biomedical Technology” OR “Self- HelpDevices” OR “WearableElectronicDevices” OR “WearableDevice” OR “WearableDevices” OR “WearableComputer” OR “WearableComputers” OR “WearableDevice” OR “WearableDevices” OR “WearableElectronicDevice”)AND(“Safety” OR “Industrial Safety” OR “Safety Management” OR “Accident Prevention” OR Preven*)
SciELO	(Forklift* OR “Carretilla Elevadora” OR “Carretillas elevadoras” OR Forklift* OR “Lift Truck” OR “Lift Trucks”) AND (“Technology” OR Tecnologi* OR “Technological Development” OR “Digital Technology” OR “Biomedical Technology” OR “Assistive Technology” OR “WearableElectronicDevices” OR “WearableComputer” OR “WearableElectronicDevice” OR “WearableDevice” OR “WearableDevices” OR “TechnologicalDevelopment” OR “SmartPhone” OR “Self-HelpDevices” OR “WearableElectronicDevices” OR “Wearable Computers” OR “Wearable Electronic Device” OR “Wearable Electronic Device” OR “Wearable Device” OR “Wearable Device” OR “Wearable Electronic Devices” OR “Wearable Devices” OR “Wearable Devices” OR “Technology” OR “Technolog*” OR “Technological Development” OR “Digital Technology” OR “Biomedical Technology” OR ”Self-Help Devices” OR ”Wearable Electronic Devices” OR ”Wearable Device” OR ”Wearable Devices” OR ”Wearable Computer” OR ”Wearable Computers” OR ”Wearable Device” OR ”Wearable Devices” OR ”Wearable Electronic Device”)AND(“Safety” OR “Industrial Safety” OR “Safety Management” OR “Accident Prevention” OR “Seguridad” OR “Seguridad Industrial” OR “Administración de la Seguridad” OR “Prevención de Accidentes” OR “Safety” OR “Industrial Safety” OR “Safety Management” OR “Accident Prevention” OR “Preven*”)
Scopus	(Forklift* OR “LiftTruck” OR “LiftTrucks”) AND (“Technology” OR Technolog* OR “TechnologicalDevelopment” OR “DigitalTechnology” OR “BiomedicalTechnology” OR “Self-HelpDevices” OR “WearableElectronicDevices” OR “WearableDevice” OR “WearableDevices” OR “WearableComputer” OR “WearableComputers” OR “WearableDevice” OR “WearableDevices” OR “WearableElectronicDevice”) AND (“Safety” OR “IndustrialSafety” OR “SafetyManagement” OR “AccidentPrevention” OR Preven*)
Web of Science	(Forklift* OR “LiftTruck” OR “LiftTrucks”) AND (“Technology” OR Technolog* OR “TechnologicalDevelopment” OR “DigitalTechnology” OR “BiomedicalTechnology” OR “Self-HelpDevices” OR “WearableElectronicDevices” OR “WearableDevice” OR “WearableDevices” OR “WearableComputer” OR “WearableComputers” OR “WearableDevice” OR “WearableDevices” OR “WearableElectronicDevice”) AND (“Safety” OR “IndustrialSafety” OR “SafetyManagement” OR “AccidentPrevention” OR Preven*)

Selection of Evidence Sources

After removing duplicates, three reviewers screened the studies (BPB, GMS, and GA). This process was conducted blindly and independently among the reviewers.

Initially, the titles and abstracts were reviewed. Two reviewers (BPB and GMS) read the abstracts of the selected articles to assess their relevance to the study and verify whether they met the inclusion and exclusion criteria.

The preselected articles were read in full by three reviewers (BPB, GMS, and GA) to confirm their relevance to the study and their compliance with the established criteria.

Data Extraction Process

Data extraction was conducted independently by author BPB from the selected databases. The data were organized in a spreadsheet created in Google Sheets® (gap table), providing a structured and organized visualization of the information extracted from the studies included in the review.

Data Items

The following relevant information was extracted: (1) study characteristics, including author, journal, year, title, and study type; (2) clinical applicability; (3) type of technology used; and (4) main results and limitations.

Summary of Results

The information was organized into tables and accompanied by narrative content, using Google Sheets®.

Results

A total of 390 records were retrieved from the following databases: CINAHL (n = 4); Compendex (n = 82); Computers and Applied Sciences (n = 63); Embase (n = 25); IEEE Xplore Digital Library (n = 49); PubMed/MEDLINE (n = 21); Scopus (n = 104); and Web of Science (n = 42). No eligible studies were found in the BDENF, LILACS, and SciELO databases.

After removing duplicates (n = 131), 259 documents remained for title and abstract screening. Of these, 248 were excluded for not meeting the scope of the study, resulting in the selection of 11 articles for full-text review, of which three were excluded for not aligning with the objective and the research question. In the end, the review consisted of eight articles. The search and selection procedure for the studies in this review is illustrated in the flowchart (Figure 1).

Figure 1
Search flowchart, according to recommendations, adapted from PRISMA-ScR

Regarding country of origin, two studies were conducted in China (25%), two in Spain (25%), two in Italy (25%), one in the United States (12.5%), and one in Thailand (12.5%). Regarding language, all studies were in English (n = 8; 100%). The distribution of articles by year of publication was as follows: 2014 (n = 2; 25%), 2021 (n = 1; 12.5%), 2024 (n = 4; 50%), and 2025 (n = 1; 12.5%). A detailed description of the studies, including title, objective, authorship, year, country, and journal of publication, is provided in Table 2.

Table 2
Characteristics of the studies included in the scoping review

ID	Title	Objective	Study design	Number of participants and outcome	Authors	Year
A120	Lightweight Object Detection Model for Frozen Food Warehouses	To propose a new object detection model, YOLOv8-RSS, for the intelligent management of frozen food warehouses.	Experimental study on the development and validation of the proposed deep learning model and testing of the YOLOv8-RSS model on a proprietary dataset, including an ablation study and comparison with existing models	Not applicable. The study used an image dataset consisting of 4,463 photographs captured by three surveillance cameras at a cold storage warehouse in Yuxi, China. Outcomes: Object detection accuracy (mAP@0.5 and mAP@0.5:0.95); number of model parameters (M); computational cost (FLOPs); detection of workers with and without appropriate PPE (protective clothing); and detection of three types of forklifts	Yang et al.	2025
A217	Occupational Health and Safety Digitalization in Thailand: A Framework and Case Study	To describe digitalization in Thailand and the digitalization of occupational health and safety in the country.	Narrative mini-review + descriptive case study. The review describes the landscape of digitalization in occupational health and safety (OHS) in Thailand; the case study reports on the implementation of an IoT system for forklift fleet management at a manufacturing company, with data collection over 6 months (June–November 2023), following the PDCA cycle.	Not applicable to the narrative review. Operators’ driving speed; impact force on forklifts (driving behavior indicator); drivers’ working hours; distance traveled; engine temperature (preventive maintenance); occurrence of near misses and accidents; energy consumption and maintenance costs.	Boonruksa, Nontapa, and Theppitak	2024
A324	Connected and Intelligent Framework for Vehicle Automation in Smart Ports	To design and validate a scalable collaborative framework that meets the specifications and functions of port services, using Industry 4.0 technologies and advanced wireless communications through Connected Intelligent Transportation Systems (C-ITS).	A research article on the development and technological validation, implementation, and testing of a connected and intelligent framework for forklift automation in a smart port (Smart-Port).	Not applicable. The study does not involve human participants. The system was tested with three forklifts. Feasibility of V2X (Vehicle-to-Everything) communication between forklifts and port infrastructure; accuracy of autonomous trajectory tracking; ability to detect and respond to obstacles; synchronization between vehicles and smart traffic lights; interoperability of the V2X-MQTT gateway with the port management system (PMS); operational safety through the automation of repetitive and high-risk tasks	Vaca-Recalde et al.	2024
A421	A few-shot learning methodology for improving safety in industrial scenarios through universal self-supervised visual features and dense optical flow	To develop a methodology for the detection, identification, and tracking of moving objects in generic industrial scenarios, with the aim of assessing situations that pose potential risks to worker safety.	Study on the development and validation of the DINOFSAFE methodology for detecting moving objects in industrial environments, using videos obtained under real operational conditions.	Not applicable. The study used a dataset of approximately 6,500 annotated images of industrial environments, without direct human involvement. The training of the classifiers was based on 45 manually annotated instances. The study evaluated object classification and detection performance, computational efficiency, and annotation effort, as well as the system’s ability to detect hazardous situations involving workers, forklifts, and loads in industrial environments.	del Olmo et al.	2024
A525	Investigating the two-by-two relationships between situational awareness, eye movements, and performance in forklift drivers	To investigate the relationship between situational awareness, eye movements, and performance in forklift drivers.	An experimental field study with forklift operators, using eye tracking, situational awareness assessment, and performance analysis. The data were analyzed using Pearson’s correlation and machine learning algorithms.	15 forklift operators. The following were evaluated: situational awareness, operational performance, and eye-tracking metrics, as well as the associations among these indicators across different forklift operation tasks.	Kang et al.	2024
A626	Forklift tracking: Industry 4.0 implementation in large-scale warehouses through UWB sensor fusion	To determine the location of pallets in a warehouse by associating their position with the forklift’s trajectory and orientation during unloading.	Experimental study with validation via computer simulation (numerical analysis using Monte Carlo + real-world testing).	Not applicable. Technological study conducted with a forklift on nine distinct experimental routes in a 4,560 m2 warehouse. The accuracy of forklift localization was evaluated through the fusion of data from UWB, IMU, and optical flow sensors, using an Unscented Kalman filter, as well as the ability to track pallets in real time in an industrial environment.	Motroni, Buffi, and Nepa	2021
A727	Methods for Improving Visibility Measurement Standards for Powered Industrial Vehicles	The study aims to improve the visibility of forklifts to reduce accidents involving pedestrians by developing an automated system to measure blind spots and guide manufacturers on how to design or adapt vehicles with safety sensors.	An experimental study comparing four advanced methods for evaluating forklift visibility based on digital modeling and laser scanning techniques.	Not applicable. Technological study conducted using a forklift. The study evaluated forklift blind spots, the compliance of visibility measurement methods with current standards, and the optimal placement of safety sensors to reduce operational risks.	Bostelman et al.	2014
A828	Application of RFID Technology to Support Effective Risk Management in Chemical Warehouses	The study analyzes trends in chemical supply chains and their impact on risks to warehouse operators. It presents a case study of a chemical warehouse, dividing the facility into areas for storage, loading/unloading operations, blending and filling of bulk products, and commercial offices.	Single-case study with quantitative risk analysis at a 25,000 m2 chemical warehouse, supplemented by a technical and economic feasibility assessment of an RFID/RTLS system for operational risk management.	Not applicable. The study was conducted at a warehouse with more than 100 workers, distributed among administrative and operational roles, within the context of organizational and technological analysis. The risks of serious accidents involving forklift operations in a chemical warehouse were quantified, along with their economic costs, and the feasibility and cost-benefit of RFID/RTLS systems for risk mitigation were evaluated.	Bragatto, Pirone, and Gnoni	2014

• PPE: Personal Protective Equipment; FLOPs: Floating point operations per second; OSH: Occupational Health and Safety; PDCA: Plan, Do, Check, Act; UWB: Ultra Wideband; IMU: Inertial Measurement Unit; RFID/RTLS: Radio Frequency Identification / Real-Time Location System.

The eight included studies were grouped into four technological categories, according to the primary nature of the solution employed: (1) Industrial Internet of Things (IIoT); (2) radio frequency tracking and location; (3) connected systems and vehicle automation; and (4) computer vision and deep learning. Additionally, two studies addressed situational awareness and operator visibility as complementary dimensions to technology adoption.

Industrial Internet of Things (IIoT)

IIoT-based applications have demonstrated the potential to improve safety and operational efficiency. Boonruksa, Nontapa, and Theppitak¹⁷ presented a case study on forklift fleet management (n = 5) at a continuously variable transmission manufacturing plant in Thailand, using an IoT system composed of five integrated components: a GPS unit (to record speed, operating hours, distance traveled, and parking location), card readers (to monitor operator behavior), an engine temperature sensor, an impact force indicator, and an automated shutdown system.

The core mechanism for improving safety was based on individually tracking each operator’s driving speed via the GPS unit: by making each driver’s speed behavior visible and measurable, the system raised operators’, leading to a voluntary reduction in speed. After implementation, 60% of the operators (3 out of 5) reduced their average driving speeds. However, the reduced speeds still did not reach the safety limit established by the company (≤ 7 km/h), which highlights a limitation of the model based solely on data visibility: individual awareness alone proved insufficient to ensure full compliance with safety standards. Considering this, the authors¹⁷ planned to enhance the system by adding real-time automatic alerts when the speed limit is exceeded, moving from a passive monitoring approach to an active intervention targeting operator behavior.

In addition to speed management, monitoring operating hours made it possible to identify imbalances in the distribution of workload among operators, ranging from 105 to 360 hours over a six-month period, and to inform the redesign of travel routes, reducing the distance traveled from 336.5 to 264.5 meters, which had a direct impact on fatigue prevention, reduced energy consumption, and lower maintenance costs. During the IoT system’s implementation period (June through November 2023), no accidents or near-misses were recorded, in contrast to the historical average of 0.97 incidents per semester over the previous nine years¹⁷.

Tracking and Location via Radio Frequency (RFID, UWB, and RTLS)

Motroni, Buffi, and Nepa²⁶ proposed a tracking system for forklifts in a 4,560 m² warehouse, combining Radio Frequency Identification (RFID), Inertial Measurement Unit (IMU), Optical Flow Sensor (OFS), Ultra Wideband (UWB), and the Nonlinear Kalman Filter (UKF) algorithm. This sensor fusion enabled precise tracking of the forklifts even during movement and unloading operations. The use of UKF to determine the vehicles’ orientation improved the accuracy of pallet localization and the efficiency of the goods-handling process. The achieved localization performance, combined with low processing load, qualifies the system not only for real-time monitoring of warehouse conditions but also as a candidate for collision-prevention applications²⁶. Although the direct impact on promoting safer environments has not been measured, the capabilities offered by the technology have the potential to contribute to workplace safety.

Bragatto, Pirone, and Gnoni²⁸ analyzed the application of RFID and Real-Time Location Systems (RTLS) to reduce accidents and failures in complex industrial systems, with a focus on chemical warehouses. The researchers demonstrated that, in addition to reducing accidents, these technologies are cost-effective. The implementation of RTLS was identified as a viable solution for tracking operations in highly complex environments, potentially reducing the likelihood of errors during maneuvers and cargo handling²⁸.

Connected Systems and Vehicle Automation

Vaca-Recalde et al.²⁴ investigated the use of Industry 4.0 technologies in a Smart-Port environment, employing Connected Intelligent Transportation Systems (C-ITS), Automated Guided Vehicles (AGVs), and on-board units (OBUs) for collaborative maneuvers among forklifts, obstacle detection, and integration with the port’s connected infrastructure. The framework demonstrated feasibility in logistics control, improving cargo handling efficiency and increasing workplace safety through collaborative maneuvers among vehicles. The integration of AGVs into the logistics system contributed to a smarter and safer operation²⁴.

Computer Vision and Deep Learning

Research focused on computer vision highlights advances in the use of algorithms and smart sensors for risk detection and operational monitoring. Yang et al.²⁰ proposed the YOLOv8-RSS model, an adaptation of YOLOv8 for object detection in frozen-food warehouses, trained using a dataset collected by surveillance cameras. The improvement aimed to reduce computational demands while maintaining accuracy and effectiveness in detections, essential characteristics in logistics operations, where rapid response times are critical for safety and operational efficiency²⁰.

Del Olmo et al.²¹ applied the DINOFSAFE methodology, which combines dense optical flow with the DINOv2 model and the Vision Transformer (ViT), to identify hazardous conditions in industrial environments. The methodology was developed and tested in warehouse areas using a custom dataset of approximately 6,500 images, combining motion segmentation and visual features to identify risk patterns. This approach reduced the need for human training and optimized the use of computational resources, making it a scalable solution for various industrial environments. The integration of deep learning and computer vision techniques represents a significant advance in the automation of workplace monitoring²¹.

Situational Awareness and Operator Visibility

Kang et al.²⁵ conducted a field study with 15 forklift drivers using the Tobii Glasses 2 device to track eye movements. The study assessed the operators’ situational awareness, an essential skill for safe driving in dynamic environments, and found that operators with greater situational awareness tend to keep focus on the direction of travel, broadening their perception of the environment and reducing the risk of accidents. In contrast, operators with lower situational awareness diverted their attention to the instrument panel, increasing the risk of distraction. The results reinforce the need for technologies aimed at improving situational awareness²⁵.

Bostelman et al.²⁷ proposed the use of CAD modeling, laser scanning, and panoramic photography to measure forklift operator visibility and identify potentially hazardous blind spots. Panoramic photography emerged as a cost-effective and efficient solution for assessing the operator’s field of view, while laser scanning and CAD modeling provided more detailed assessments of visible and non-visible areas. The results informed adjustments to vehicle design and the placement of safety sensors, with the aim of mitigating operational risks²⁷.

Discussion

The studies analyzed focus on three technological areas applied to safety in the internal movement of cargo. In the field of IIoT and tracking, Boonruksa, Nontapa, and Theppitak¹⁷ showed a significant reduction in accidents using IIoT in forklifts, recording zero incidents during monitoring compared to a historical average of 0.97 per semester. Complementary results were obtained by Vaca-Recalde et al.²⁴ through the integration of AGVs and C-ITS systems in a Smart-Port, and by Motroni, Buffi, and Nepa²⁶, who combined RFID, IMU, UWB, and the UKF algorithm for precise tracking with the potential to prevent collisions. Bragatto et al.²⁸ reinforced the economic viability of these solutions by demonstrating the affordable cost and positive return on investment of RFID and RTLS systems in complex logistics environments.

In the field of computer vision, Yang et al.²⁰ improved the YOLOv8 model for hazard detection in warehouses, while Del Olmo et al.²¹ developed the DINOFSAFE methodology, combining optical flow and Vision Transformer for scalable identification of hazardous conditions. Bostelman et al.²⁷ contributed by mapping operators’ blind spots using CAD modeling and laser scanning, supporting the redesign of safer vehicles. Additionally, Kang et al.²⁵ tracked operators’ eye movements and found that greater situational awareness is associated with a lower risk of accidents, showing that operational safety depends as much on automated solutions as on the enhancement of human perception. Overall, the findings reinforce the trend toward technological convergence as a strategy for enhancing safety and operational efficiency, although the predominance of studies in controlled environments points to the need for research in larger-scale real-world scenarios.

Safety management is based on the principle of responding to adverse events or identifying risks classified as unacceptable, with the aim of eliminating their causes and strengthening prevention and mitigation barriers. However, the Safety II approach expands this concept by emphasizing the importance of understanding the factors that enable successful operations under normal conditions, to elucidate the circumstances that may eventually lead to failures¹.

The results obtained from the application of IIoT-based technologies in industrial settings are promising and indicate a trend toward evolution in occupational safety practices.

In the study by Boonruksa, Nontapa, and Theppitak¹⁷, the implementation of these solutions improved forklift management, promoting greater safety and operational efficiency by enhancing operators’ perception of speed and reducing average speed, thereby lowering the risk of accidents. The authors¹⁷ also proposed monitoring forklift operating hours and speed, the use of card readers to track operator behavior, temperature sensors to check engine conditions, impact indicators, and automated control mechanisms.

This integrated approach highlights a trend observed across the analyzed studies: different technologies tend to prioritize distinct aspects of occupational safety. Technologies such as RFID and UWB stand out for their ability to optimize operational flows, control access to hazardous areas, and monitor the location of equipment and workers in real time, contributing primarily to the efficiency and traceability of processes. In contrast, computer vision-based solutions are predominantly geared toward active risk detection, identifying unsafe behaviors, proximity between operators and equipment, and situations of imminent danger. This distinction is relevant from the perspective of occupational safety management: while the former act in a more structural and preventive manner on the environment, the latter respond more dynamically to real-time events. Combining these approaches, as suggested by the study by Boonruksa et al.¹⁷, can enhance results, since the complementarity between efficient tracking and risk detection tends to produce more robust and comprehensive safety systems.

These measures contributed to increased safety, reduced fatigue, route reorganization, energy savings, and preventive maintenance planning.

These findings are consistent with the literature, which indicates that the IIoT enhances visibility into the operational environment, facilitating early risk identification and the implementation of more agile corrective actions¹⁸.

Complementarily, Motroni, Buffi, and Nepa²⁶ demonstrated that the IIoT is essential for tracking forklifts in industrial environments, enabling real-time data collection and transmission, reducing route errors, and improving the accuracy of equipment location. This application contributes to the optimization of logistics processes and increased safety in loading and unloading operations²⁶. Similar results are observed in studies that combine tracking and automation technologies, such as the integrated use of RFID and connected sensors, which reinforce the potential of these solutions to reduce accidents and improve operational control²⁹,³⁰.

Similarly, the framework developed by Vaca-Recalde et al.²⁴ demonstrated feasibility for logistics control in a Smart Port environment, promoting greater efficiency in cargo handling and increased safety through collaborative maneuvers between vehicles. In line with this, studies that applied real-time monitoring to control safety procedures in high-risk areas also revealed improvements in safety management and the mitigation of occupational risks³¹. Furthermore, RFID technology, as highlighted by Bragatto, Pirone, and Gnoni²⁸, has proven effective for tracking assets, tools, and workers, strengthening organizational safety practices through the control of material flow and access to restricted areas.

In this context, progress has been made in the application of IIoT in the industrial sector, establishing it as a strategic ally in incident mitigation by enabling continuous tracking and greater predictability of operations, which represents a new perspective for safety management in industry.

Computer vision has been widely applied for real-time detection and monitoring of risks. In the studies by Yang et al.²⁰ and del Olmo et al.²¹, the relevance of this technology for object recognition is highlighted, enabling the early identification of risky situations and rapid decision-making. The ability to process large volumes of visual data enables continuous, automated surveillance in industrial environments and high-risk work activities²⁰,²¹. In this context, model-based approaches such as YOLOv8 have been used not only to detect people and equipment but also to monitor safe distances and compliance with safety standards, such as OSHA guidelines⁶,³², reinforcing the preventive potential of these solutions.

The YOLOv8-RSS model, used by Yang et al.²⁰ in the monitoring of frozen food warehouses, demonstrated improvements in resource allocation and increased operational safety. The system stood out for its ability to simultaneously detect people and forklifts, as well as to identify risks, such as the improper use of personal protective equipment. Similarly, other applications of YOLOv8 have incorporated distance estimation and postural analysis modules using Convolutional Neural Networks, enabling the automatic classification of safe and unsafe operations and the identification of risky operator behaviors in real time³²,³³. Furthermore, the lightweight design of these models facilitates their implementation in environments with limited computational resources²⁰, broadening their practical applicability.

In summary, studies demonstrate that the application of computer vision – especially through deep learning-based models such as YOLOv8 – has proven useful in detecting risks, monitoring behaviors, and automating safety in industrial environments. When integrated with the IIoT, this technology enhances the capacity for incident prevention and response, contributing to safer, more efficient, and smarter operations.

In industrial environments, del Olmo et al.²¹ proposed a methodology geared toward contexts with intense interaction between people and objects, applicable to various manufacturing sectors. Based on universal visual features and self-supervised learning, the proposal reduces the need for manual training and, consequently, computational costs. In addition, the authors developed a dataset containing approximately 6,500 images, contributing to the advancement of research in the field. However, they emphasize that the creation of safer operational environments depends on the development of specific computational rules capable of identifying and predicting hazards according to the reality of each process²¹.

Furthermore, Vaca-Recalde et al.²⁴ validated a modular system for port automation, which ensures safer operations and efficient communication. According to the authors, automation increases productivity, reduces repetitive tasks, and requires a skilled workforce. Although laboratory tests have confirmed their feasibility, challenges remain regarding adaptation to new types of vehicles, the complexity of operational scenarios, and cybersecurity. The study also highlighted the role of data analysis algorithms, which process and interpret information captured by sensors, identifying patterns and predicting potential incidents²⁴.

Another relevant aspect is the application of eye-tracking and data analysis technologies, as demonstrated by Kang et al.²⁵, who used Eye-Tracking combined with the Situation Awareness Rating Technique (SART) to measure operators’ situational awareness and identify lapses in attention that could potentially cause accidents. The study showed that operators with greater situational awareness maintain their focus on the frontal field of view, while those with lower SART scores tend to concentrate on panels and consoles, increasing the risk of incidents. The technology provides objective, real-time data to optimize situational awareness, support driver training, and improve the design of human-machine systems, thereby reducing the likelihood of accidents in environments with multiple interactions²⁵.

Finally, 3D modeling and simulation tools have proven to be strategic for improving workplace safety. Bostelman et al.²⁷ demonstrated that the use of CAD models and three-dimensional simulations enables the creation of virtual scenarios to test, validate, and optimize operations, allowing for the early identification of risks and the adoption of corrective measures prior to practical implementation, which contributes to greater safety and operational efficiency. In line with this perspective, Building Information Modeling (BIM) applications have also demonstrated potential in preventive safety management by facilitating the visualization of risks, the detection of structural failures, and the advance planning of interventions³⁴. Such approaches reinforce the importance of digital modeling technologies as tools to support decision-making and integrated safety management.

Overall, studies are consistent in showing that the use of emerging technologies, such as IIoT, computer vision, intelligent automation, eye tracking, and 3D modeling, represents an advance in industrial safety management by enabling the precise identification of risks, the anticipation of failures, and the promotion of safer and more efficient work environments.

The integration of these solutions establishes a data-driven approach capable of enhancing safety management and optimizing operational processes. Thus, it is evident that incorporating these technologies into industrial routines not only enhances the efficiency and reliability of operations but also redefines the paradigms of incident and accident prevention and control, constituting a decisive step toward a safer, smarter, and more resilient industry.

Conclusion

The results of this study indicate that technologies applied to the industrial environment show promising potential for promoting workplace safety. The solutions analyzed – IIOT, computer vision, and tracking systems – showed results that are potentially beneficial for improving operational safety, reducing human error, and optimizing processes in specific contexts. IIoT, for example, proved to be a promising approach for forklift management, with the integration of sensors and connected devices contributing to efficiency and safety gains in the scenarios investigated.

However, it is important to note that these findings derive primarily from case reports and technical feasibility tests, based on a relatively small number of studies included in this review (n = 8), which limits the generalizability of the results. No statistically robust evidence of accident reduction on a real-world scale was identified in the studies analyzed. Thus, although the technologies examined appear to be promising candidates for logistics and industrial environments, their effectiveness in broad operational scenarios still lacks empirical validation, reinforcing the need for future studies with greater methodological rigor and scope.

Computer vision, on the other hand, stood out in the early detection of risks and automated surveillance, contributing to real-time safety. In addition, eye-tracking technology was instrumental in improving operators’ situational awareness, promoting a safer work environment by reducing the likelihood of accidents.

The implementation of these technologies not only improves safety conditions but also contributes to operational efficiency, enabling real-time monitoring and intervention, as well as more accurate and informed decision-making. The integration of systems such as IIoT, RFID, and data analysis algorithms offers a holistic approach to safety management, with positive impacts on both worker protection and operational productivity. In short, the adoption of these technologies in industrial environments represents a significant step forward in promoting safer working conditions, with great potential to shape the future of occupational safety.

However, it is essential to deepen our understanding of the phenomenon under study to establish guidelines that more precisely define the risks to workers and the mechanisms for predicting and mitigating them. Furthermore, this review highlights a gap in applied research, especially in Brazil, where no studies on the topic were identified.

References

• ¹ Hollnagel E. Safety-I and Safety-II: the past and future of safety management. Londres: CRC Press; 2014.
• ² International Labor Organization. Segurança e saúde no trabalho. Geneva: ILO; 2023 [citado 5 jun 2025]. Disponível em: https://www.ilo.org/pt-pt/resource/news/quase-3-milhões-de-pessoas-morrem-devido-acidentes-e-doenças-relacionados
  » https://www.ilo.org/pt-pt/resource/news/quase-3-milhões-de-pessoas-morrem-devido-acidentes-e-doenças-relacionados
• ³ Smartlab. Observatório de Segurança e Saúde no Trabalho: promoção do meio ambiente do trabalho guiado por dados. Brasília: Ministério Público do Trabalho; 2022 [citado 5 jun 2025]. Disponível em: https://smartlab.mpt.mp.br/sst
  » https://smartlab.mpt.mp.br/sst
• ⁴ Nações Unidas no Brasil. Acidentes de trabalho e mortes acidentárias voltam a crescer no Brasil em 2021. Brasília: ONU Brasil; 2021 [citado 5 jun 2025]. Disponível em: https://brasil.un.org/pt-br/178950-acidentes-de-trabalho-e-mortes-acident%C3%A1rias-voltam-crescer-no-brasil-em-2021
  » https://brasil.un.org/pt-br/178950-acidentes-de-trabalho-e-mortes-acident%C3%A1rias-voltam-crescer-no-brasil-em-2021
• ⁵ Occupational Safety and Health Administration. Commonly used statistics. Washington: OSHA; 2023 [citado 5 jun 2025]. Disponível em: https://www.osha.gov/data/commonstats
  » https://www.osha.gov/data/commonstats
• ⁶ Occupational Safety and Health Administration. Forklift safety: hazards & prevention. Washington: OSHA; 2024 [citado 13 set 2025]. Disponível em: https://www.osha.com/blog/forklift-hazards-prevention
  » https://www.osha.com/blog/forklift-hazards-prevention
• ⁷ Instituto Nacional do Seguro Social (BR). Comunicações de Acidente de Trabalho - CAT: Plano de Dados Abertos: jun. 2023 a jun. 2027. Brasília: INSS; 2025 [citado 13 set 2025]. Disponível em: https://www.gov.br/inss/pt-br/acesso-a-informacao/dados-abertos/plano-de-dados-abertos/pda-inss-2023-2027.pdf
  » https://www.gov.br/inss/pt-br/acesso-a-informacao/dados-abertos/plano-de-dados-abertos/pda-inss-2023-2027.pdf
• ⁸ Pauliková A, Babelová ZG, Ubárová M. Analysis of the impact of human-cobot collaborative manufacturing implementation on the occupational health and safety and the quality requirements. Int J Environ Res Public Health. 2021;18(4):1750. https://doi.org/10.3390/ijerph18041927
  » https://doi.org/10.3390/ijerph18041927
• ⁹ Schwab K, Davis N. Aplicando a quarta revolução industrial. São Paulo: Edipro; 2019.
• ¹⁰ Meindl B, Ayala NF, Mendonça J, Frank AG. The four smarts of Industry 4.0: Evolution of ten years of research and future perspectives. Tech Forecast Soc Change. 2021;168:120784. https://doi.org/10.1016/j.techfore.2021.120784
  » https://doi.org/10.1016/j.techfore.2021.120784
• ¹¹ Chanchaichujit J, Tan A, Meng F, Eaimkhong S. Healthcare 4.0: next generation processes with the latest technologies. Singapura: Palgrave Macmillan; 2019.
• ¹² Cwiklicki M, Klich J, Chen J. The adaptiveness of the healthcare system to the fourth industrial revolution: a preliminary analysis. Futures. 2020;122:102602. https://doi.org/10.1016/j.futures.2020.102602
  » https://doi.org/10.1016/j.futures.2020.102602
• ¹³ Santos JLS, Marques MAJ, Sell D, Trierveiler HJ. Potenciais aplicabilidades de tecnologias digitais: um estudo sobre wearables na indústria de óleo e gás. An Congr Int Conhec Inov. 2022;11:1-15. https://doi.org/10.48090/ciki.v1i1.1169
  » https://doi.org/10.48090/ciki.v1i1.1169
• ¹⁴ Silva FF, Vita JB. Wearables, telemedicina e futuro da saúde no brasil: novas tendências em tecnologias de saúde para enfrentamento de pandemias. Hum Inov. 2022 [citado 6 jun 2025];9(19):195-211. Disponível em: https://revista.unitins.br/index.php/humanidadeseinovacao/article/view/5608
  » https://revista.unitins.br/index.php/humanidadeseinovacao/article/view/5608
• ¹⁵ Tourinho FSV, Schuelter PI, Fermo VC, Caldas MM, Alves TF, Barbosa SS, organizadores. Desenvolvimento de tecnologias em pesquisa e saúde: da teoria à prática. São Paulo: Científica Digital; 2022.
• ¹⁶ Boyes H, Hallaq B, Cunningham J, Watson T. The industrial internet of things (IIoT): an analysis framework. Comput Ind. 2018;101:1-12. https://doi.org/10.1016/j.compind.2018.04.015
  » https://doi.org/10.1016/j.compind.2018.04.015
• ¹⁷ Boonruksa P, Nontapa S, Theppitak C. Occupational health and safety digitalization in thailand: a framework and case study. J UOEH. 2024;46(1):93-101. https://doi.org/10.7888/juoeh.46.93
  » https://doi.org/10.7888/juoeh.46.93
• ¹⁸ Khan WZ, Rehman MH, Zangoti HM, Afzal MK, Armi N, Salah K. Industrial internet of things: recent advances, enabling technologies and open challenges. Comput Electr Eng. 2020;81:106522. https://doi.org/10.1016/j.compeleceng.2019.106522
  » https://doi.org/10.1016/j.compeleceng.2019.106522
• ¹⁹ Gonçalves M, Marques T, Gaspar PD, Soares VNGJ, Caldeira JMLP. Protótipo de solução para detetar e sinalizar defeitos em pavimentos rodoviários baseado em técnicas de visão computacional. Rev Ibérica Sist Tecnol Inf. 2023;(52):25-44. https://doi.org/10.17013/risti.52.25-44
  » https://doi.org/10.17013/risti.52.25-44
• ²⁰ Yang J, Liang Z, Qin M, Tong X, Xiong F, An H. Lightweight object detection model for food freezer warehouses. Sci Rep. 2025;15(1):2350. https://doi.org/10.1038/s41598-025-86662-z
  » https://doi.org/10.1038/s41598-025-86662-z
• ²¹ Olmo JJL, Gómez ÁLP, López-de-Teruel PE, Ruiz A. A few-shot learning methodology for improving safety in industrial scenarios through universal self-supervised visual features and dense optical flow. Appl Soft Comput. 2024;167:112375. https://doi.org/10.1016/j.asoc.2024.112375
  » https://doi.org/10.1016/j.asoc.2024.112375
• ²² Peters MDJ, Godfrey C, McInerney P, Munn Z, Tricco AC, Khalil H. Scoping reviews. In: Aromataris E, Munn Z, editores. Joanna Briggs Institute reviewer's manual. Adelaide: Joanna Briggs Institute; 2020 [citado 6 jun 2025]. p. 408-452. Disponível em: https://doi.org/10.46658/JBIMES-20-12
  » https://doi.org/10.46658/JBIMES-20-12
• ²³ Tricco AC, Lillie E, Zarin W, O'Brien KK, Colquhoun H, Levac D et al. PRISMA extension for scoping reviews (PRISMA-ScR): Checklist and explanation. Ann Intern Med. 2018;169(7):467-73. https://doi.org/10.7326/M18-0850
  » https://doi.org/10.7326/M18-0850
• ²⁴ Vaca-Recalde ME, Marcano M, Matute J, Hidalgo C, Martinez-Rodriguez B, Bilbao-Arechabala S. Connected and intelligent framework for vehicle automation in smart-ports. IEEE Access. 2024;12:120347-61. https://doi.org/10.1109/ACCESS.2024.3434628
  » https://doi.org/10.1109/ACCESS.2024.3434628
• ²⁵ Kang Y, Huang W, Zhu X, Wu J, Gao X, Sheng X. Investigating the two-by-two relationships between situational awareness, eye movements, & performance in forklift drivers. SSRN. 2024:1-28. https://doi.org/10.2139/ssrn.4867031
  » https://doi.org/10.2139/ssrn.4867031
• ²⁶ Motroni A, Buffi A, Nepa P. Forklift tracking: industry 4.0 implementation in large-scale warehouses through UWB sensor fusion. Appl Sci. 2021;11(22):10607. https://doi.org/10.3390/app112210607
  » https://doi.org/10.3390/app112210607
• ²⁷ Bostelman R, Teizer J, Ray SJ, Agronin M, Albanese D. Methods for improving visibility measurement standards of powered industrial vehicles. Saf Sci. 2014;62:257-70. https://doi.org/10.1016/j.ssci.2013.08.020
  » https://doi.org/10.1016/j.ssci.2013.08.020
• ²⁸ Bragatto PA, Pirone A, Gnoni MG. Application of RFID technology for supporting effective risk management in chemical warehouses. In: Steenbergen RDJ, van Gelder PHAJM, Miraglia S, Vrouwenvelder ACWM, editores. Safety, reliability and risk analysis: beyond the horizon. London: CRC Press; 2014 [citado 6 jun 2025]. p. 1479-1486. Disponível em: https://doi.org/10.1201/b15938
  » https://doi.org/10.1201/b15938
• ²⁹ Jagadeesh RS, Murugesh MC, Viji U. Design and development of an automated forklift with iot weight system. Int J Multidiscip Res Anal. 2024;7(12):5450-6. https://doi.org/10.47191/ijmra/v7-i12-11
  » https://doi.org/10.47191/ijmra/v7-i12-11
• ³⁰ Kovavisaruch L, Maneerat K, Sanpechuda T, Chinda K, Wongsatho T, Wisadsud S et al. UWB tracking for Forklift in near-automate warehouse. In: Proceedings of the 2024 5th Technology Innovation Management and Engineering Science International Conference (TIMES-iCON). New York: IEEE; 2024 [citado 6 nov 2025]. p. 1-5. Disponível em: https://ieeexplore.ieee.org/document/10630749
  » https://ieeexplore.ieee.org/document/10630749
• ³¹ McNinch M, Parks D, Jacksha R, Miller A. Leveraging IIoT to improve machine safety in the mining industry. Min Metall Explor. 2019;36(4):675-81. https://doi.org/10.1007/s42461-019-0067-5
  » https://doi.org/10.1007/s42461-019-0067-5
• ³² Sabir A, Hussain R, Abbas MS, Zaidi SFA, Yang J, Park C. Computer vision-based safe distance-estimation method to prevent forklift tip-overs. Int J Occup Saf Ergon. 2025;32(2):420-39. https://doi.org/10.1080/10803548.2025.2543150
  » https://doi.org/10.1080/10803548.2025.2543150
• ³³ Sang-ngenchai A, Nakazawa M. Smart warehouse safety: computer vision for forklift driver monitoring in warehouse setting. Proceedings of 2025 IEEE International Conference on Consumer Electronics (ICCE). New York: IEEE; 2025 [citado 6 nov 2025]. p. 1-6. Disponível em: https://ieeexplore.ieee.org/document/10930064
  » https://ieeexplore.ieee.org/document/10930064
• ³⁴ Seixas RM, Maués LMF, Rosa CCN, Oliveira FA. Building information modeling (BIM) para gestão da segurança do trabalho em obras de habitações populares. Ambient Constr. 2022;22(3):235-54. https://doi.org/10.1590/s1678-86212022000300617
  » https://doi.org/10.1590/s1678-86212022000300617

• Data Availability:
  The entire dataset supporting the results of this study is available at: https://osf.io/8k39x/overview
• Statement on the use of Artificial Intelligence:
  During the preparation of this article, the ChatGPT Free GPT-5.5 artificial intelligence tool was used for language review. The authors declare that they reviewed and validated all content.
• Presentation at a scientific event:
  The authors report that the study has not been presented at a scientific event.
• Funding:
  SESI National Department through the 2024 Innovation Platform: Studies and Research on Connected Occupational Health; Coordination for the Improvement of Higher Education Personnel – Brazil (CAPES) – Funding Code 001; FAPESC Public Call for Proposals No. cp 31/2021, Inova Talentos Program – FAPESC, FAPESC Case No. 1081/2024.

Edited by

• Responsible editors:
  Francisco de Paula Antunes Lima https://orcid.org/0000-0003-4373-6424 Universidade Federal de Minas Gerais – UFMG
  Raoni Rocha Simões https://orcid.org/0000-0003-1181-0132 Universidade Federal de Ouro Preto – UFOP

Data availability

The entire dataset supporting the results of this study is available at: https://osf.io/8k39x/overview

Publication Dates

• Publication in this collection
  31 Aug 2026
• Date of issue
  2026

History

• Received
  09 June 2025
• Reviewed
  11 Feb 2026
• Accepted
  20 Apr 2026