Next Article in Journal
Seismic Performance of Tall-Pier Girder Bridge with Novel Transverse Steel Dampers Under Near-Fault Ground Motions
Previous Article in Journal
Psychology or Physiology? Choosing the Right Color for Interior Spaces to Support Occupants’ Healthy Circadian Rhythm at Night
Previous Article in Special Issue
Real-Time Progress Monitoring of Bricklaying
 
 
Font Type:
Arial Georgia Verdana
Font Size:
Aa Aa Aa
Line Spacing:
Column Width:
Background:
Article

RTLS-Enabled Bidirectional Alert System for Proximity Risk Mitigation in Tunnel Environments

1
School of Engineering and IT, Central Queensland University, Sydney, NSW 2000, Australia
2
Department of Civil Engineering, Ghulam Ishaq Khan Institute of Engineering Sciences and Technology, Topi 23640, Pakistan
3
University of South Australia—Online, Adelaide, SA 5001, Australia
*
Authors to whom correspondence should be addressed.
Buildings 2025, 15(15), 2667; https://doi.org/10.3390/buildings15152667
Submission received: 30 June 2025 / Revised: 16 July 2025 / Accepted: 19 July 2025 / Published: 28 July 2025
(This article belongs to the Special Issue AI in Construction: Automation, Optimization, and Safety)

Abstract

Tunnel construction poses significant safety challenges due to confined spaces, limited visibility, and the dynamic movement of labourers and machinery. This study addresses a critical gap in real-time, bidirectional proximity monitoring by developing and validating a prototype early-warning system that integrates real-time location systems (RTLS) with long-range (LoRa) wireless communication and ultra-wideband (UWB) positioning. The system comprises Arduino nano microcontrollers, organic light-emitting diode (OLED) displays, and piezo buzzers to detect and signal proximity breaches between workers and equipment. Using an action research approach, three pilot case studies were conducted in a simulated tunnel environment to test the system’s effectiveness in both static and dynamic risk scenarios. The results showed that the system accurately tracked proximity and generated timely alerts when safety thresholds were crossed, although minor delays of 5–8 s and slight positional inaccuracies were noted. These findings confirm the system’s capacity to enhance situational awareness and reduce reliance on manual safety protocols. The study contributes to the tunnel safety literature by demonstrating the feasibility of low-cost, real-time monitoring solutions that simultaneously track labour and machinery. The proposed RTLS framework offers practical value for safety managers and informs future research into automated safety systems in complex construction environments.

1. Introduction

The global expansion of tunnel infrastructure continues to play a vital role in meeting the increasing demands for sustainable urban development, advanced transportation systems, and underground utility networks. Despite technological progress in tunnelling equipment and construction methodologies, the tunnel construction industry remains particularly prone to safety incidents, with accident rates surpassing those in many other construction sectors [1,2]. Tunnel worksites are inherently hazardous due to their spatial constraints, poor visibility, and the dynamic interaction between workers and machinery operating within confined spaces [3,4]. These challenges, combined with unpredictable and dynamic geological conditions, specific technological capabilities, and construction methods, contribute to significant hazards to labourers and equipment, making tunnel projects more vulnerable to accidents than other types of geotechnical engineering projects [2]. Recent analyses of tunnel projects in seismically active or karst-prone regions demonstrate that complex geological conditions, such as fault dislocations and underground water inrushes, can significantly heighten construction risks [5,6,7].
While regulatory efforts and safety protocols have traditionally focused on the “fatal four” hazards, namely falls, struck by injuries, caught in or between incidents, and electrocutions [8], proximity-related risks between labourers and heavy equipment continue to be underreported and insufficiently mitigated. Workers are often more attuned to recognising life-threatening hazards than non-fatal but high-risk interactions, such as near-miss events involving machinery [1,9]. Moreover, early collapse warnings and casualty assessments triggered by falling risks in confined spaces reinforce the urgency for proactive proximity-based systems [10,11]. Consequently, there is an urgent need for safety systems that not only extend hazard detection beyond static protocols but also dynamically monitor and prevent proximity incidents in real time [12].
A range of safety technologies, including global positioning system (GPS), radio frequency identification (RFID), and wireless local area networks (WLAN), have been explored to support real-time positioning and hazard detection in construction environments [13,14,15]. However, these technologies are often limited in underground or obstructed environments due to signal attenuation, interference, and low spatial resolution [16,17]. Recent reviews have also identified that common systems, such as RFID, face severe signal degradation or misreads in shielded underground environments, particularly in curved tunnel segments or utility conduits [16,17]. UWB RTLSs offer a promising alternative, providing high precision, strong signal penetration, and robust resistance to environmental noise [18,19,20]. Nonetheless, most existing applications remain unidirectional and are typically designed to track either equipment or personnel in isolation, lacking the integrated, responsive functionality required for managing real-time proximity risks involving both entities [1,21].
Correspondingly, the existing RTLS-based systems used in tunnel construction only track either workers or machinery at a time, with limited information about each other, and are dependent on some local coordinate system established in the environment. These one-way systems work well for some situations in general environments. However, they do not fully consider the dynamic interactions and safety-critical shifts in confined spaces, as reported in recent tunnel risk assessments [22,23]. This paper presents a new bidirectional proximity alert system that combines UWB-based RTLS with LoRa communication modules, OLED displays, and microcontrollers to monitor both workers and machines in real time. Traditional unidirectional systems only track one entity, either workers or equipment, which limits situational awareness. Our system, in contrast, enables two-way detection and proximity-based alerts, transferring positional data and issuing warnings when safety thresholds are crossed, thus enhancing on-site responsiveness and hazard mitigation [24]. This research approach increases tunnel safety by allowing real-time monitoring of both labour and machinery, especially in scenarios with reduced visibility and unpredictable movements. A compact configuration integrating RTLS, UWB, LoRa communication, and low-cost microcontroller technologies is implemented, which has shown robust performance in comparable underground and fault-affected tunnel settings [5,7]. These integrated technologies support operations in GPS-denied environments, enabling effective safety management in high-risk tunnel projects.
The system is meant to let you keep an eye on both workers and machines in real time and send out audio and visual alerts when safety limits are crossed. The study is guided by four key objectives: (1) to develop an RTLS capable of accurately tracking the movement of both labourers and machinery in confined tunnel environments; (2) to integrate proximity detection algorithms that generate bidirectional warnings when a predefined safety threshold is breached, thereby mitigating collision and struck-by risks; (3) to enhance site safety management by reducing reliance on manual safety protocols and enabling automated, real-time monitoring of proximity risks; and (4) to test and validate the effectiveness of the proposed RTLS through pilot studies simulating hazardous tunnel scenarios, with a specific focus on preventing near-miss incidents. This study contributes to the growing body of research at the intersection of construction safety and digital innovation. It offers a scalable and cost-effective framework for real-time hazard prevention in tunnel environments, with practical implications for safety managers, regulatory bodies, and infrastructure developers. More broadly, the research aligns with global efforts to reduce occupational injuries in high-risk construction sectors by integrating advanced sensing technologies into everyday safety practice.
The remainder of the paper is structured as follows. Section 2 reviews the existing literature on tunnel construction hazards and proximity detection systems. Section 3 outlines the research methodology, system architecture, and component integration. Section 4 presents the pilot testing framework and experimental design. Section 5 discusses the performance of the system in simulated scenarios. Section 6 considers the theoretical, practical, and policy implications of the work. Finally, Section 7 concludes with recommendations for future development and application.

2. Previous Work on Automating Hazard Detection and Worker Safety

Various technologies have been employed in construction operations to reduce proximity risks, ranging from traditional methods such as manual spotters to more advanced systems like RFID and GPS. While these technologies offer certain benefits, they often prove insufficient in complex environments like tunnels due to limitations in the accuracy and range [13,14]. In contrast, RTLS from Makerfabs., Shenzhen, China provides a more robust solution for tunnel construction by enabling precise and continuous tracking of personnel and equipment [20]. RTLS employs advanced signal-processing techniques and UWB (Makerfabs., Shenzhen, China) technology, ensuring high accuracy in tracking even in challenging conditions [18]. This literature review aims to examine the range of existing technologies applied for proximity risk mitigation in tunnel and confined construction environments and to identify the strengths and limitations that inform the development of improved safety systems.

2.1. Tunnel Construction Operations and Accidents

Tunnel construction is often classified as an indoor activity, where maintaining safe distances between labourers and equipment presents considerable challenges due to spatial constraints and limited visibility [12]. To address these challenges, various location-tracking and hazard-detection technologies have been explored, each offering unique advantages and limitations in tunnel environments. One of the most widely used systems in outdoor construction is the GPS. While GPS excels in open environments, its signal reliability is severely hindered underground due to signal attenuation and reflection by tunnel walls [14]. These limitations render GPS largely ineffective for real-time safety monitoring in tunnel construction projects [17].
To overcome this, alternative indoor tracking systems, such as RFID, have been adopted. RFID is particularly effective in enclosed environments, like mines and tunnels, where it provides a cost-efficient and scalable solution for tracking both equipment and personnel [13]. Studies have shown that RFID-enabled systems can support the simultaneous tracking of multiple targets, reduce the risk of collisions and enhance real-time visibility on-site [25,26]. In addition to RFID, vision-based positioning systems have gained attention for their ability to visually monitor site activity. These systems utilise cameras and image-processing algorithms to detect and track equipment, such as wheel loaders, dozers, and concrete buckets, with accuracy levels reaching up to 88% [27]. Furthermore, they have been successfully used to identify unsafe labourer behaviours and assess compliance with safety protocols [26].
Another widely explored option in tunnel construction is WLAN-based positioning, which makes use of existing Wi-Fi infrastructure to triangulate the location of devices [15]. Although cost-effective and relatively easy to implement, WLAN-based systems often suffer from reduced accuracy in tunnels due to signal fluctuations caused by structural interference. Accuracy ranges have been found to vary significantly, with errors reported between 0.6 m and 6.9 m in shield tunnel environments [16].
In recent years, UWB RTLSs have emerged as one of the most accurate and reliable solutions for proximity risk mitigation in complex construction settings [16]. UWB transmits short-duration pulses that can effectively filter out reflected signals, thereby minimising the multipath interference typically encountered in tunnel conditions. As a result, UWB-based systems provide precise real-time tracking of both machinery and labourers, even in confined or obstructed spaces [19]. These systems have already been applied in construction to monitor crane operations and ensure that personnel maintain safe distances from high-risk machinery [28]. Their superior accuracy, adaptability, and real-time data capabilities make UWB RTLS a promising tool for enhancing tunnel construction safety protocols.

2.2. Need for Integrated Safety Management Systems in Tunnel Construction

Effective proximity analysis is essential for maintaining safety in tunnel construction, particularly due to the confined, complex, and dynamic nature of such environments [18]. While several technologies have been developed for proximity detection, most have been limited by environmental constraints, low precision, or unidirectional application [27]. Despite these developments, a clear and specific research gap persists in the deployment of integrated safety systems that simultaneously monitor both labourers and machinery in tunnel construction. Current systems tend to focus on either personnel or equipment in isolation and rely on manual or fragmented controls, which are inadequate for addressing real-time proximity risks [1,21]. Furthermore, although RTLSs have shown promise in industrial safety applications, their adoption in tunnel construction remains limited, particularly those based on UWB technology [20].
Existing studies primarily examine traditional RTLS technologies, such as RFID or infrared, leaving the high-precision, interference-resistant capabilities of UWB RTLSs underutilised [29]. There are several RTLS-based systems that have been used to manage construction safety. As mentioned earlier, these include systems that use RFID, Bluetooth low energy (BLE), and ZigBee technologies. These systems have worked well in some situations. However, they often do not perform reliably in tunnels due to weak signal penetration, interference, and low accuracy [30,31]. RFID systems, for instance, can help keep track of people in small spaces, but they have a limited range and need a line of sight to function effectively. BLE and ZigBee systems are also cheap, but they have trouble maintaining communication in GPS-denied areas like tunnels. This study addresses this gap by investigating the application of UWB RTLS as an integrated, bidirectional proximity warning system designed specifically for tunnel environments. The presented RTLS system outperforms traditional options in terms of long-range communication, signal penetration, and accuracy. This makes it more suitable for the real-time monitoring of both workers and machines in complex tunnel scenarios. By enabling real-time tracking and alerting both labourers and machinery, the system seeks to mitigate proximity-related hazards and enhance site-wide safety protocols in tunnel construction [7,24].

3. Methods

This study adopts an action research methodology, characterised by iterative cycles of problem identification, system design, implementation, evaluation, and refinement within a real-world operational context [32]. This methodological approach is particularly appropriate given the objective of co-developing and validating an RTLS-based bidirectional early-warning system intended to improve proximity safety in tunnel construction. The prototype was subjected to a series of pilot trials designed to simulate dynamic and spatially constrained conditions typical of tunnel environments. Through this process, the system was progressively optimised to address the challenges of real-time monitoring and hazard detection involving both labour and machinery in high-risk underground settings.
Figure 1 provides a structured overview of the research methodology, outlining six sequential stages: adoption of an action research framework, coordination of signals using UWB technology, development of the RTLS prototype, integration of system components, implementation of circuit wiring and programming, and final validation of the system through pilot testing under simulated tunnel construction conditions. Signal coordination plays a central role in the development of the prototype, particularly in the context of UWB technology, which is employed to calculate proximity distances with high accuracy. In tunnel construction environments, where operational precision and safety are critical, reliable proximity measurement is essential for preventing hazardous interactions between workers and machinery. The effectiveness of the RTLS depends on the performance of two principal components: the transmitter and the receiver. These components must be precisely configured and synchronised to ensure the accurate transmission and reception of signals [33,34]. The subsequent sections provide a detailed account of the selection, integration, and operational roles of the electrical components that constitute the foundational architecture of the RTLS developed in this study.

3.1. Key Components for Real-Time Proximity Detection in Tunnel Construction

The circuit assembly in this research involves selecting the appropriate components, establishing their connections, and assembling the circuit for optimal functionality [35]. The primary goal is to create a system that performs accurately and reliably in real-world tunnel construction scenarios. This section details the core components used, including their role in the RTLS system and how they contribute to the overall project objectives. Figure 2 presents the key components of the RTLS system developed for proximity detection in tunnel construction, as these components are reliable, robust, and appropriate for building tunnels [5,7]. It includes the Arduino Nano (microcontroller), LoRa SX1278 module (long-range communication), Espressif (ESP32) UWB module (positioning), jumper wires (circuit connectivity), OLED display (visual output), and a piezo buzzer (audible alerts).
The Arduino Nano was chosen due to its compact form factor, as it is small and easy to fit into confined areas like those found in tunnels. It has enough processing power to do proximity calculations in real time, which makes it perfect for this use. While alternative technologies like Raspberry Pi and other similar technologies have more processing power, the Arduino Nano’s simplicity, low power usage, and cost-effectiveness make it a feasible choice for the dynamic conditions of tunnel construction sites. The LoRa module was selected due to its capability to provide long-range, low-power communication, as it can communicate over long distances with low power, which is crucial for keeping track of people and machines in small spaces over long distances [23]. Unlike ZigBee or BLE, which may suffer from signal interference and range limitations, LoRa offers excellent resistance to physical obstructions and environmental noise, making it perfect for tunnels, which are underground [31]. The ESP32 was incorporated for its dual Bluetooth and Wi-Fi capabilities, providing reliable connectivity for communication between devices when they are in close proximity. ZigBee and BLE are commonly used in construction safety systems, but the ESP32 is more flexible and scalable. It is compatible with many sensors and communication protocols, making it a versatile component for real-time monitoring. These components work in coordination to monitor proximity and issue warnings when thresholds are breached. Each part is further discussed in its respective section.

3.1.1. Arduino Nano

The Arduino Nano is a compact microcontroller that serves as the brain of the RTLS system in this project [36]. Based on the ATmega328 architecture, the Arduino Nano is chosen for its small form factor, making it ideal for integrating into tight spaces, such as those encountered in tunnel environments. The Arduino Nano is responsible for analysing and calculating proximity distances between various positions within the construction site [37]. It compares these calculated distances against pre-set threshold values, triggering warnings or alerts when those thresholds are crossed. Figure 2a shows the Arduino Nano used in this project. Its ease of use, combined with its capability to process distance calculations in real-time, makes it an integral part of the system.

3.1.2. LoRa Long Range Low Power

The LoRa module (SX1276/77/78/79) is a key component used for long-range, low-power communication within the RTLS framework [38]. This transceiver features a LoRa modem that provides ultra-long-range spread spectrum communication, which is essential for large-scale construction sites like tunnels. The LoRa technology offers high interference immunity, making it suitable for harsh environments where signal obstruction and interference are common challenges. It achieves a sensitivity of over −148 dBm, which allows for the robust tracking of machinery and personnel across long distances [39]. This module transmits and receives positional data, allowing the system to track and monitor the movements of workers and equipment accurately. Figure 2b provides a visual of the LoRa module employed in this project.

3.1.3. ESP 32

The ESP32 is a powerful and versatile microcontroller unit (MCU) with integrated Wi-Fi and Bluetooth capabilities, which play a crucial role in ensuring connectivity for the RTLS system. This component is chosen for its ability to create a stable Bluetooth connection that facilitates the exchange of positional data between devices [33]. In this project, the ESP32 enables real-time communication by transmitting and receiving the coordinates of workers and machinery within the tunnel construction environment. It also interfaces seamlessly with other components via standard communication protocols such as serial peripheral interface (SPI), secure digital input/output (SDIO), and universal asynchronous receiver/transmitter (UART) [40]. Figure 2c shows the ESP32 module used in the RTLS.

3.1.4. Pin-to-Hole Jumper Wire

Pin-to-hole jumper wires, also referred to as DuPont lines, are electrical connectors used to create temporary or permanent circuit connections between various components in the RTLS system [41]. In this project, the jumper wires are used to connect the Arduino Nano, LoRa module, and ESP32, allowing for seamless communication between the hardware elements [41]. These wires are known for their flexibility and reliability, providing easy connectivity without the need for soldering. Figure 2d illustrates the jumper wires used for this project, demonstrating their importance in prototyping and circuit integration.

3.1.5. Arduino 0.96-Inch OLED Display Module

The OLED display module used in this project is a 0.96-inch screen with a resolution of 128 × 64, specifically chosen for its low power consumption and high visibility. The OLED display shows the real-time proximity distances between workers and machinery, providing immediate feedback on the safety conditions [42]. This display is crucial for the pilot tests, as it visually communicates proximity thresholds and alert statuses [43]. Figure 2e shows the OLED display module utilised in the system. Its clarity and compact size make it well-suited for applications where space is limited, such as tunnel construction.

3.1.6. Active Piezo Buzzer

The active piezo buzzer is a compact sound-generating device used to emit audible warnings when proximity thresholds are exceeded [44]. It operates by applying a direct current (DC) voltage, which triggers a continuous beep at a set frequency (usually around 2 kHz). This alert mechanism is essential for safety in tunnel construction, as it provides an immediate, unmistakable warning when workers come too close to dangerous machinery or hazardous areas [17]. The buzzer operates within a wide voltage range, making it a versatile addition to the RTLS system. Figure 2f shows the piezo buzzer incorporated into the system for proximity alerts.

3.2. System Architecture and Key Component Integration for Real-Time Proximity Detection in Tunnel Construction

The Arduino Nano functions as the central controller for the entire circuit, serving as the system’s primary processing unit [36,37]. It provides power to the other components, processes inputs, and generates the requisite outputs to ensure optimal operation. The RTLS system operates by sending UWB signals, which allows it to measure distances accurately, even in confined areas such as tunnels [30,45]. The UWB module transmits short-duration pulses that the receiver module receives, which computes the proximity distance using the time-of-flight of the signal. The connections between the Arduino Nano and the other electrical components are crucial for achieving real-time monitoring of proximity distances within tunnel construction environments. Each connection is carefully designed to ensure reliability and effective communication between components. Arduino Nano processes this data and triggers alerts when the threshold of proximity is exceeded. The algorithmic control logic is developed to ensure real-time processing of proximity data and minimise delays [29]. Arduino Nano’s role is pivotal, as it facilitates the data flow and ensures that the system operates with maximum efficiency. The role of Arduino Nano is to manage communication between the LoRa transceivers and the UWB modules in order to ensure that proximity alerts are transmitted immediately when safety thresholds are exceeded, providing timely warnings to both workers and machinery operators.

3.2.1. Arduino Nano and ESP32 Connections

The connection between the Arduino Nano and the ESP32, as shown in Figure 3, was designed utilising Altium Designer. The connection is established using jumper wires, which facilitate effective communication between the two devices. The GND (ground) pin on both devices is connected to provide a common ground, ensuring stable operation [36,40]. The TX (transmit) pin on the Arduino Nano is connected to the RX (receive) pin on the ESP32, enabling the Arduino to transmit data to the ESP32. Conversely, the RX pin on the Arduino Nano is connected to the TX pin on the ESP32, facilitating bidirectional communication between the two components. This communication is crucial for transferring positional data and coordinating actions within the system, ensuring that the ESP32 receives data from the Arduino Nano and transmits the appropriate response.

3.2.2. Arduino Nano and LoRa Module Connections

The connection between the Arduino Nano and the LoRa module is shown in Figure 4. This connection facilitates long-range communication within the RTLS, which is essential for monitoring the location of workers and machinery over extensive distances [39]. The serial clock (SCK) pin on the LoRa module is connected to the Arduino Nano’s D13 pin, providing the clock signal necessary to synchronise data transfer [38]. The master in slave out (MISO) pin on the LoRa module is connected to D12 on the Arduino Nano, enabling data transmission from the LoRa module to the Arduino. Conversely, the master out slave in (MOSI) pin is connected to D11 on the Arduino Nano, allowing the Arduino to transmit data to the LoRa module. The slave select (NSS) pin is connected to D10 on the Arduino, which enables the Arduino to control when the LoRa module is selected for communication. A shared GND connection ensures electrical stability between the two devices, while the reset (RST) pin connected to D9 allows the Arduino to reset the LoRa module, if necessary. Finally, the digital input/output (DIOO) pin on the LoRa module is connected to D2 on the Arduino Nano to manage general-purpose communication signals between the two devices. These connections facilitate long-distance data exchange, which is crucial for real-time monitoring in tunnel environments.

3.2.3. Arduino Nano and Buzzer Connections

The connection between the Arduino Nano and the buzzer is summarised in Table 1.
The D10 pin on the Arduino Nano is connected to the positive terminal of the buzzer, while the GND pin on the Arduino is connected to the negative terminal of the buzzer. This configuration enables the Arduino Nano to control the buzzer and generate auditory alerts when specific proximity thresholds are exceeded [41]. The buzzer functions as an acoustic warning system within the RTLS, notifying workers when they approach hazardous areas or when machinery exceeds the safe proximity distance. This connection is essential for providing immediate, real-time feedback, thereby ensuring safety in the dynamic and confined environment of tunnel construction.

3.2.4. Prototype Integration

The complete prototype of the RTLS system is presented in Figure 5, which shows the connections for both the transmitter and receiver components.
The Arduino Nano serves as the primary controller, interfaced with the LoRa GPS module to facilitate real-time location tracking [46]. In the transmitter configuration, the LoRa module is responsible for transmitting positional data from machinery and personnel to the Arduino Nano, which processes the information and generates appropriate alerts if predetermined thresholds are exceeded. The receiver configuration similarly incorporates the Arduino Nano, which monitors incoming data and activates both visual (via the OLED display) and auditory (via the buzzer) warnings. The interconnectedness of the components, namely the Arduino Nano, ESP32, LoRa module, and buzzer, ensures efficient communication and coordination, enabling the system to function effectively in challenging tunnel environments [39,40,41]. This prototype is optimised for detecting proximity risks, issuing warnings, and enhancing safety in real-time. The utilisation of jumper wires in the design allows for flexibility and facilitates modifications during testing and further development.

3.3. Embedded Code Development for Proximity Monitoring of Entities in Tunnel Construction

The embedded code development of the transmitter and receiver was a fundamental step to establish a real-time proximity detection system for tunnel construction. The code was implemented using Python (version 3.10.6) and C++ (via the Arduino IDE version 1.8.19), which was subsequently uploaded via the Arduino integrated development environment (IDE) [41]. The receiver code decodes incoming data, computes the distances between the entities, and generates visual and audio alarms when thresholds are exceeded. The transmitter code evaluates GPS data to identify proximity breaches. In complex tunnel construction environments, this two-way connection guarantees ongoing environmental monitoring and improves worker and equipment safety.

3.3.1. Transmitter Code

The transmitter code is responsible for two primary functions, namely determining proximity to hazardous areas and identifying proximate equipment [47]. The code continuously monitors the environment, utilising proximity sensors. It quantifies the distance between personnel or machinery and predefined hazardous zones, comparing the measurements with established thresholds. If the proximity falls below or equals the threshold, the system initiates a warning, typically through an alert mechanism, such as an auditory or visual indicator. Error-checking algorithms are also incorporated to manage any sensor malfunctions or inaccurate readings, rendering the system robust and reliable [48].
The transmitter pseudo-code, as shown in Figure 6, commences with the initialisation of critical components, including serial communication, GPS, and LoRa modules. In the event of LoRa module initialisation failure, the system outputs an error message (LoRa in it failed) and halts further execution. Upon successful initialisation, the transmitter continuously monitors GPS data. Once GPS data is detected, the code encodes this information, displays it, and calculates the proximity distance based on the GPS coordinates. Subsequently, the code compares the calculated distance against the predefined threshold.
In the event that the measured proximity distance surpasses the predetermined threshold, the system maintains a monitoring state without initiating any alerts. Conversely, if the distance is less than or equal to the threshold, a warning signal is generated through the modification of the message status. This status is subsequently transmitted through the LoRa module to the receiver. The system iterates this process at regular intervals, thereby ensuring continuous monitoring and prompt responses to any proximity breaches [18]. Furthermore, the pseudo-code incorporates error-checking routines that ensure sensor reliability and account for potential inaccuracies in the proximity measurements [49].
As such, the objective of the transmitter code is to maintain safety by continuously monitoring proximity distances in real time and transmitting alerts when thresholds are exceeded [33]. The iterative structure within the code ensures that the system consistently evaluates proximity risks, thereby enhancing operational safety in tunnel construction environments.

3.3.2. Receiver Code

The receiver code is responsible for receiving the coordinates transmitted from the transmitter and calculating the proximity distance to moving machinery or personnel [50]. The receiver continuously monitors for incoming messages, decodes the transmitted coordinates, and compares the proximity distance with the predefined threshold. If the proximity distance falls below the threshold, the system initiates a warning, typically utilising an auditory alert through a buzzer.
The pseudo-code for the receiver, as shown in Figure 7, commences by initialising the serial communication and LoRa modules. In the event of module initialisation failure, the system displays an error message (LoRa failed). Upon successful initialisation, the receiver monitors for incoming LoRa packets containing GPS data. Upon packet reception, the code verifies the validity of the message format. In cases where the message format is deemed invalid, the system displays an error message and continues monitoring for new packets.
Upon receipt of a valid packet, the receiver decodes the GPS coordinates and calculates the proximity distance between the equipment and the workers [14,51]. If this calculated distance exceeds the predefined threshold, the system continues to monitor the environment without issuing alerts. However, if the distance is less than or equal to the threshold (e.g., 9 m), the system initiates a warning by activating an audible alert mechanism. Furthermore, the receiver displays the coordinates and distance to provide feedback on the proximity situation and transmits a response to the transmitter to confirm the proximity breach [18]. This process iterates in a continuous loop to ensure ongoing monitoring and real-time alerting when proximity risks arise.
The receiver code ensures that the system responds promptly to proximity risks by issuing warnings immediately upon threshold breaches [52]. This continuous monitoring and feedback system enhances worker safety in hazardous environments, such as tunnel construction.

4. Prototype Validation in Simulated Environment

To validate the performance and reliability of the developed RTLS prototype, a series of experimental evaluations was conducted in a controlled indoor environment. The validation process employed case study methodology, a widely accepted approach in construction and engineering research for investigating systems in context-specific settings [53]. Case studies allow for an in-depth, empirical investigation of dynamic interactions between workers, equipment, and environmental hazards, making them particularly suitable for examining proximity risk scenarios in tunnel-like environments.
The corridor on the second floor of Hostel 6 at the Ghulam Ishaq Khan Institute (GIK) was selected as the experimental site. Its confined indoor nature closely simulated the spatial constraints and GPS-denied conditions typical of tunnel construction environments, where precise location tracking is essential. Three distinct case studies were designed to cover critical real-world proximity risk scenarios, namely static proximity risk (danger areas), dynamic proximity risk (moving equipment and personnel), and static proximity risk (electric shock hazard). This multi-case design ensured comprehensive validation of the prototype across a range of static and dynamic conditions, enhancing the robustness and generalisability of the results [54].

4.1. Case Study 1: Static Proximity Risk in Danger Areas

This case study focused on addressing proximity risks within a predefined danger area. The coordinates of this danger area were determined, and transmitters were positioned at these points. The OLED screens displayed these coordinates, and the transmitter function was integrated into the Arduino IDE. A threshold distance of 4 m was established to initiate alerts. The experiment involved moving the transmitter from one end of the corridor towards the danger area. When the transmitter came within 3.8 m of the predefined danger coordinates, the system generated a warning, and the buzzer was activated to indicate the proximity risk. The danger area is depicted in Figure 8a, which shows the specific location where the proximity risks were measured. Also, Figure 8b demonstrates the route taken by the transmitter as it approached the danger area, providing a visual representation of how the proximity distances changed as the transmitter moved closer to the hazardous zone. The collected data, as presented in Table 2, shows the proximity distances measured at various points during the experiment. It is noteworthy that the proximity distance exhibits a consistent decrease, commencing at 14.89 m and diminishing to 6.46 m as the transmitter approaches the designated danger area.

4.2. Case Study 2: Dynamic Proximity Risk with Moving Equipment

The second study examined the dynamic proximity risks associated with moving equipment. A volunteer participant simulated this scenario, and the experiment was conducted on the first floor of Hostel 6. The transmitter was affixed to the volunteer to represent a worker, while the receiver was attached to the moving equipment. The system was configured to monitor proximity and initiate alerts when the distance between the two entities fell below the predetermined threshold. For this experiment, the threshold distance was established at 9 m to maintain an appropriate safety margin between the worker and the equipment. The receiver, utilising LoRa networking, continuously calculated the proximity between the transmitter (volunteer) and the equipment, transmitting a warning to the buzzer when the threshold was exceeded. Figure 9a provides a visual representation of the proximity changes recorded during the experiment, offering an overview of the interaction between the transmitter and equipment. The route traversed by the volunteer is shown in Figure 9b, demonstrating the dynamic evolution of proximity risk as the volunteer moved. The proximity data collected during this dynamic risk assessment is summarised in Table 3. A warning was triggered when the distance between the volunteer and the equipment decreased to less than 9 m, as evidenced by the proximity distances of 7.80 m and below.

4.3. Case Study 3: Static Proximity Risk from Electric Shock Hazard

This case study examined the static proximity risks associated with electric shock hazards. The coordinates of the hazard area were identified, and a transmitter was positioned at this location. The OLED display presented the coordinates, and the transmitter function was integrated with the Arduino IDE. A threshold distance of 3 m was established for initiating an alert. The transmitter was moved from the distal end of the corridor towards the danger area. When the transmitter approached within 2.8 m of the danger area, a warning was generated, and the buzzer was activated to indicate the proximity risk. The danger area is shown in Figure 10a, which presents the area with the electric shock hazard, while Figure 10b depicts the route traversed by the transmitter, demonstrating how proximity risks were managed in this scenario. The data collected during this test is presented in Table 4. The proximity distance steadily decreased from 14.89 m to 6.46 m as the transmitter approached the hazard, triggering the warning at 2.8 m.

5. Findings and Results

The prototype validation was conducted to ascertain the accuracy and efficacy of the developed system in indoor-positioning scenarios, simulating tunnel environments where GPS is ineffective. Three case studies were undertaken to evaluate the system’s performance in detecting static and dynamic proximity risks, each focusing on distinct safety scenarios.

5.1. Case Study 1: Static Proximity Risk in Danger Areas

The first case study examined static proximity risk. The prototype underwent testing in the corridor of Hostel 6 (second floor) at GIK, with a hazard area outlined by specific coordinates. A transmitter was positioned at a designated hazard location, and its coordinates were recorded on the OLED display. Subsequently, the transmitter was moved towards the danger zone, and the system was configured to initiate an alert when the transmitter came within 3.8 m of the predefined hazard area.
The results demonstrated that the system effectively generated warnings as the transmitter approached the danger area. The buzzer was activated immediately upon the transmitter crossing the threshold distance, thus validating the system’s capability in detecting proximity risks. However, a minor delay was observed between crossing the threshold and the actual activation of the warning system. This temporal delay ranged from 5 to 8 s across the tests, as shown in Figure 11.

5.2. Case Study 2: Dynamic Proximity Risk with Moving Equipment

The second case study investigated the dynamic proximity risk, focusing on mobile equipment. This experiment was conducted on the first floor of Hostel 6. A volunteer, simulating mobile machinery, carried the transmitter while the receiver was positioned at a fixed point. The system was configured to initiate an alert at a proximity threshold of 9 m, designed to ensure that personnel maintain a safe distance from the machinery. The results indicated that the system successfully detected the proximity of the mobile equipment and generated timely alerts when the transmitter exceeded the 9-metre threshold. However, similar to the static case, the warning system exhibited a slight delay in alert generation. The delay was more pronounced in the dynamic case, with response times ranging between 6 and 7 s. Despite this, the system demonstrated its potential to effectively manage dynamic proximity risks, as shown in Figure 12.

5.3. Case Study 3: Static Proximity Risk from Electric Shock Hazard

The third case study examined static proximity risk involving an electric shock hazard. The prototype underwent further testing in the corridor of Hostel 6. A transmitter was positioned in proximity to an electric hazard area, with a predetermined threshold distance of 3 m. The transmitter was subsequently moved toward the hazard area, and the system was configured to initiate a warning upon crossing the threshold. As anticipated, the system generated a warning when the transmitter approached within 2.8 m of the electric hazard area. This outcome demonstrated the prototype’s capability to accurately detect proximity risks in a static hazard scenario. However, a minor delay in warning generation was observed, analogous to the first two case studies. This delay, although not substantial, underscores the necessity for further system optimisation.

5.4. Time Delay and Proximity Inaccuracy (Cross-Case Analysis)

Across all three case studies, the system demonstrated a consistent delay in generating alerts after exceeding the predefined threshold distances. This temporal delay ranged from 5 to 8 s in the various tests. Furthermore, in both static and dynamic proximity scenarios, the system did not consistently trigger warnings at the precise threshold distance. The warnings were frequently delayed until the transmitter had exceeded the threshold by a marginal distance, resulting in minor inaccuracies in proximity distance detection.
Figure 11 reflects the temporal delays recorded during the tests, while Figure 12 shows the discrepancies between the expected and actual warning times. Notwithstanding these limitations, the system proved effective in detecting proximity breaches and generating warnings, thus confirming its potential utility for enhancing safety in tunnel construction operations. However, future refinements are necessary to mitigate both the temporal delay and proximity inaccuracies.

6. Discussion

The validation of the RTLS-based bidirectional early-warning system demonstrated a strong potential for enhancing proximity risk management in tunnel construction environments, where traditional safety mechanisms are often insufficient. The system effectively identified both static and dynamic proximity breaches and generated real-time alerts when workers or machinery crossed predefined safety thresholds. These results align with findings by Liang and Liu and Rao et al. [55,56], who highlighted the need for automated, real-time hazard-monitoring solutions in dynamic and spatially constrained construction settings.
The pilot testing was conducted in a controlled indoor environment that simulated tunnel conditions. Two key technical challenges emerged: temporal delays and proximity detection inaccuracies. We comprehend that real-world tunnel environments are very dynamic and can have issues like dust, noise, and severe signal occlusion that can affect how effectively the system works [5,7]. First, a measurable time delay between the breach of proximity thresholds and the activation of alerts was consistently observed, ranging from 5 s. The observed delay seemed barely noticeable in the context of the experiment, but it could be significant in real-life tunnel construction situations where even small delays in proximity alerts could be dangerous. For example, these delays could result in an increased risk of injury if a worker is in very close proximity to machinery. In high-risk construction environments such as tunnels, even short reaction delays can dramatically increase the risk of accidents [57]. This latency is not unexpected in real-time monitoring systems relying on wireless data transfer, particularly in enclosed environments where signal attenuation, reflections, and physical obstructions like dust or metallic surfaces degrade communication quality [38,46]. Similar challenges were reported by Soltanmo hammadlou et al. [20], who noted that construction sites with complex spatial layouts and environmental noise are prone to transmission lags and processing delays, especially when using LoRa or RFID-based RTLS solutions. To address this issue, future work should focus on making the system more responsive by improving communication protocols and signal-processing algorithms. This will make sure that real-time alerts go off without long delays, even when there is interference from the environment.
Moreover, the observed delays reflect inherent processing and transmission constraints associated with microcontroller-based platforms like Arduino Nano, which, while cost-effective and flexible, are limited in computational speed and data refresh rates [37]. Signal prioritisation, data packet collisions, and the need for multiple device synchronisations further exacerbate the problem in real-world dynamic environments. Second, proximity detection inaccuracies were evident, especially during dynamic movement scenarios involving simulated equipment. Instead of triggering alerts precisely at the threshold distance, the system often registered breaches slightly after crossing the limit. Such discrepancies have been previously identified in underground or indoor environments, where signal reflections from tunnel walls [58], interference from metallic objects, and human body blockage cause errors in distance estimation [20,52].
Proximity inaccuracies can be further intensified under dynamic conditions, where rapid movements and changing angles between transmitters and receivers reduce the system’s ability to maintain continuous, accurate positional tracking [26]. It also highlighted a key limitation of single-sensor dependence and suggests that integrating complementary sensing approaches, such as hybrid UWB, could improve future accuracy. Despite these technical shortcomings, the RTLS prototype’s performance demonstrates a foundational capability to support dynamic safety monitoring in tunnel construction environments. With targeted hardware improvements, environmental compensation algorithms, and faster communication protocols, the system can evolve into a highly reliable proximity risk management tool.
The results of this study carry several important theoretical, practical, and policy implications for the advancement of construction safety technologies. From a theoretical perspective, the findings extend the body of knowledge on digital safety monitoring by demonstrating the feasibility of applying UWB RTLS in confined, dynamic, and GPS-denied environments such as tunnel construction sites [59,60,61]. Previous construction safety frameworks have largely concentrated on open and above-ground settings; this study highlights the need to incorporate real-time proximity detection into future safety management theories specifically tailored for underground and spatially constrained environments [19,20]. The demonstrated bidirectional monitoring of both labourers and machinery also opens new directions for integrating automated hazard detection models into occupational safety research.
From a practical standpoint, the study provides direct evidence of how low-cost, modular RTLS-based safety systems can be effectively deployed in high-risk construction environments to reduce reliance on manual supervision and improve dynamic risk perception. The system’s successful detection of proximity breaches across static and dynamic scenarios shows its practical viability for enhancing site-wide safety without imposing significant additional infrastructure requirements. However, the observed technical challenges, namely time delays and proximity inaccuracies, suggest that future practical implementations should prioritise the optimisation of communication protocols, algorithmic responsiveness, and sensor calibration to ensure operational reliability in complex environments [46,52]. From a policy perspective, the findings suggest that regulatory bodies should consider mandating the use of real-time proximity alert systems in tunnel and underground construction projects, similar to the requirements for personal protective equipment and fall prevention measures. Given the demonstrated vulnerabilities of traditional safety protocols in confined environments, RTLS-based monitoring systems could be formalised within occupational health and safety standards to support proactive accident prevention strategies [1,21]. Public infrastructure projects, in particular, could benefit from pilot programs incentivising the adoption of such digital safety technologies to accelerate industry-wide transformation.
Our system demonstrates how a bidirectional alert system can work to reduce the risk of proximity in tunnel construction, but there are numerous ways that it could be better in the future. Incorporating additional sensors, like infrared or thermal sensors, could improve detection accuracy in low-visibility environments, such as when workers are obscured by machinery or dust [62]. A hybrid positioning system that combines UWB with other technologies like RFID could also be a better and more accurate way to track things, especially on tunnel construction sites where conditions are constantly changing and difficult [10]. Also, it would be very important to improve the system’s processing algorithms so that data can be sent more quickly and with less lag. This would make real-time monitoring and responsiveness better [23]. These changes could make the system work better and be more reliable, which would make tunnels safer.

7. Conclusions

This study developed and experimentally validated a novel bidirectional early-warning system for proximity risk detection in tunnel construction, using RTLS integrated with LoRa communication modules, Arduino Nano microcontrollers, OLED displays, and UWB positioning technologies. Unlike traditional monitoring systems that typically focus on either workers or machinery independently, the proposed system simultaneously tracks both, offering a real-time, low-cost, and scalable solution suitable for confined and GPS-denied construction environments. The prototype was evaluated through three distOKinct case studies simulating both static and dynamic proximity risk scenarios. The results demonstrated that the system consistently detected proximity breaches and generated timely alerts, with proximity warnings activated within critical distance thresholds, such as 3.8 m for static hazards and 9 m for dynamic worker–machinery interactions. However, a temporal delay ranging between 5 and 8 s was observed between the threshold breach and the warning activation. Additionally, slight deviations from the predefined proximity thresholds were noted, particularly during dynamic movements, indicating minor accuracy limitations. Despite these technical challenges, the system represents a significant advancement in tunnel construction safety technology by automating hazard detection processes, reducing reliance on manual spotters, and enhancing real-time situational awareness for both workers and site supervisors. The research also highlights the feasibility of deploying cost-effective RTLS solutions in high-risk, spatially restricted environments where traditional safety mechanisms fall short. Future research should focus on minimising response delays through algorithmic optimisation and faster sensor integration, as well as improving proximity accuracy using sensor fusion or machine-learning-based calibration. Real-world validations across diverse tunnel projects, involving multiple simultaneous tracking points and different environmental conditions, will be critical for further enhancing system robustness and reliability. Overall, the study offers a strong foundation for future innovations in proactive digital safety management, contributing both practically and theoretically to the evolving field of construction technology.

Author Contributions

Methodology, F.A. and N.K.; Validation, A.A.K.; Formal analysis, R.J.; Writing—original draft, F.U.K.; Writing—review & editing, F.A. and N.K.; Project administration, A.A.K.; Funding acquisition, F.A. and N.K. All authors have read and agreed to the published version of the manuscript.

Funding

This work was supported by the School of Engineering, Central Queensland University.

Data Availability Statement

The original contributions presented in the study are included in the article, further inquiries can be directed to the corresponding author.

Acknowledgments

The authors would like to acknowledge the use of OpenAI’s ChatGPT-4.5 for assistance with grammar correction, language refinement, and improving the overall clarity of the manuscript. All intellectual content and analysis remain the responsibility of the authors.

Conflicts of Interest

The funding organisation had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.

References

  1. Cao, Y.; Yun, B. Tunnel Construction Worker Safety State Prediction and Management System Based on AHP and Anomaly Detection Algorithm Model. Heliyon 2024, 10, e36450. [Google Scholar] [CrossRef] [PubMed]
  2. Yang, Y.; Wang, Y.; Easa, S.M.; Yan, X. Risk Factors Influencing Tunnel Construction Safety: Structural Equation Model Approach. Heliyon 2023, 9, e12924. [Google Scholar] [CrossRef] [PubMed]
  3. Sajedifar, J.; Mehri, A.; Abbasi, M.; Naimabadi, A.; Mohammadi, A.; Teimori-Boghsani, G.; Zakerian, S. Safety Evaluation of Lighting at Very Long Tunnels on the Basis of Visual Adaptation. Saf. Sci. 2019, 116, 196–207. [Google Scholar] [CrossRef]
  4. Sousa, R.L.; Einstein, H.H. Lessons from Accidents during Tunnel Construction. Tunn. Undergr. Space Technol. 2021, 113, 103916. [Google Scholar] [CrossRef]
  5. Cui, X.; Liu, Y.; Du, X.; Xiao, H.; Xu, H.; Du, Y. Effect of Fault Dislocation on the Deformation and Damage Behavior of Ballastless Track Structures in Tunnels. Transp. Geotech. 2025, 52, 101561. [Google Scholar] [CrossRef]
  6. Li, Z.-Q.; Nie, L.; Xue, Y.; Li, W.; Fan, K. Model Testing on the Processes, Characteristics, and Mechanism of Water Inrush Induced by Karst Caves Ahead and Alongside a Tunnel. Rock Mech. Rock Eng. 2025, 58, 5363–5380. [Google Scholar] [CrossRef]
  7. Zhang, C.; Zhu, Z.; Dai, L.; Wang, S.; Shi, C. The Incompatible Deformation Mechanism of Underground Tunnels Crossing Fault Conditions in the Southwest Edge Strong Seismic Zone of the Qinghai-Tibet Plateau: A Study of Shaking Table Test. Soil Dyn. Earthq. Eng. 2025, 197, 109482. [Google Scholar] [CrossRef]
  8. Albert, A.; Pandit, B.; Patil, Y. Focus on the Fatal-Four: Implications for Construction Hazard Recognition. Saf. Sci. 2020, 128, 104774. [Google Scholar] [CrossRef]
  9. Khan, N.; Nadeau, S.; Pham, X.-T.; Boton, C. Exploring Associations between Accident Types and Activities in Construction Using Natural Language Processing. Autom. Constr. 2024, 164, 105457. [Google Scholar] [CrossRef]
  10. Xie, M.; Xu, F.; Wang, Z.; Yin, L.; Wu, X.; Xu, M.; Li, X. Investigating Fire Collapse Early Warning Systems for Portal Frames. Buildings 2025, 15, 296. [Google Scholar] [CrossRef]
  11. Xu, Z.; Zhu, Y.; Fan, J.; Zhou, Q.; Gu, D.; Tian, Y. A Spatiotemporal Casualty Assessment Method Caused by Earthquake Falling Debris of Building Clusters Considering Human Emergency Behaviors. Int. J. Disaster Risk Reduct. 2025, 117, 105206. [Google Scholar] [CrossRef]
  12. Koohathongsumrit, N.; Meethom, W. Risk Analysis in Underground Tunnel Construction with Tunnel Boring Machines Using the Best–Worst Method and Data Envelopment Analysis. Heliyon 2024, 10, e23486. [Google Scholar] [CrossRef] [PubMed]
  13. Hyun-Soo, L.; Kwang-Pyo, L.; Moonseo, P.; Yunju, B.; SangHyun, L. RFID-Based Real-Time Locating System for Construction Safety Management. J. Comput. Civ. Eng. 2012, 26, 366–377. [Google Scholar] [CrossRef]
  14. Abbas, A.; Habelalmateen, M.; Jurdi, S.; Audah, L.; Alduais, N. GPS Based Location Monitoring System with Geo-Fencing Capabilities. In AIP Conference Proceedings; AIP Publishing LLC: Melville, NY, USA, 2019; Volume 2173. [Google Scholar] [CrossRef]
  15. Khoury, H.; Kamat, V. WLAN Based User Position Tracking for Contextual Information Access in Indoor Construction Environments. In Computing in Civil Engineering; ASCE Library: Lawrence, KS, USA, 2007. [Google Scholar] [CrossRef]
  16. Broz, J.; Tichy, T.; Vlkovsky, M.; Polach, M. Towards Safety and Efficiency by Assessment of Positioning Approaches for Enhanced Navigation in Road Tunnels. Tunn. Undergr. Space Technol. 2025, 155, 106228. [Google Scholar] [CrossRef]
  17. Wang, Z.; González, V.A.; Mei, Q.; Lee, G. Sensor Adoption in the Construction Industry: Barriers, Opportunities, and Strategies. Autom. Constr. 2025, 170, 105937. [Google Scholar] [CrossRef]
  18. Huang, Y.; Hammad, A.; Zhu, Z. Providing Proximity Alerts to Workers on Construction Sites Using Bluetooth Low Energy RTLS. Autom. Constr. 2021, 132, 103928. [Google Scholar] [CrossRef]
  19. Li, T.; Li, X.; Rui, Y.; Ling, J.; Zhao, S.; Zhu, H. Digital Twin for Intelligent Tunnel Construction. Autom. Constr. 2024, 158, 105210. [Google Scholar] [CrossRef]
  20. Soltanmohammadlou, N.; Sadeghi, S.; Hon, C.K.H.; Mokhtarpour-Khanghah, F. Real-Time Locating Systems and Safety in Construction Sites: A Literature Review. Saf. Sci. 2019, 117, 229–242. [Google Scholar] [CrossRef]
  21. Loy-Benitez, J.; Song, M.K.; Choi, Y.-H.; Lee, J.-K.; Lee, S.S. Breaking New Ground: Opportunities and Challenges in Tunnel Boring Machine Operations with Integrated Management Systems and Artificial Intelligence. Autom. Constr. 2024, 158, 105199. [Google Scholar] [CrossRef]
  22. Li, X.; Meng, S.; Sun, W.; Yuan, M.; Liu, X.; Nan, Y.; Sun, Y.; Sun, D.; Tao, J.; Li, J.; et al. Research on the Risk Assessment of the Green Construction of Hydraulic Tunnels Based a on Combination Weighting-Cloud Model. Evol. Intell. 2025, 18, 57. [Google Scholar] [CrossRef]
  23. Wan, A.; Gong, W.; AL-Bukhaiti, K.; Cheng, X.; Ji, X.; Ji, Y.; Ma, S. Data-Driven Early Fire Detection in Offshore Wind Turbines: A TCN-ECA Based Approach Leveraging SCADA Data. Process Saf. Environ. Prot. 2025, 200, 107373. [Google Scholar] [CrossRef]
  24. Zhou, Z.; Zhuo, W.; Cui, J.; Luan, H.; Chen, Y.; Lin, D. Developing a Deep Reinforcement Learning Model for Safety Risk Prediction at Subway Construction Sites. Reliab. Eng. Syst. Saf. 2025, 257, 110885. [Google Scholar] [CrossRef]
  25. Bauk, S.; Dzankic, R. On Improving Working Safety: A Review of Some Rfid Based Solutions. 2016. Available online: https://www.tmt.unze.ba/zbornik/TMT2016/072.pdf (accessed on 10 July 2025).
  26. Oliveira, R.R.; Cardoso, I.M.G.; Barbosa, J.L.V.; da Costa, C.A.; Prado, M.P. An Intelligent Model for Logistics Management Based on Geofencing Algorithms and RFID Technology. Expert Syst. Appl. 2015, 42, 6082–6097. [Google Scholar] [CrossRef]
  27. Teizer, J.; Cheng, T. Proximity Hazard Indicator for Workers-on-Foot near Miss Interactions with Construction Equipment and Geo-Referenced Hazard Areas. Autom. Constr. 2015, 60, 58–73. [Google Scholar] [CrossRef]
  28. Liu, D.; Zhang, W.; Dai, Q.; Chen, J.; Duan, K.; Li, M. Safety Evaluation Method for Operational Shield Tunnels Based on Semi-Supervised Learning and a Stacking Algorithm. Tunn. Undergr. Space Technol. 2024, 153, 106027. [Google Scholar] [CrossRef]
  29. Li, D.; Chen, Q.; Wang, H.; Shen, P.; Li, Z.; He, W. Deep Learning-Based Acoustic Emission Data Clustering for Crack Evaluation of Welded Joints in Field Bridges. Autom. Constr. 2024, 165, 105540. [Google Scholar] [CrossRef]
  30. Hu, D.; Liu, J.; Li, Y.; Tan, Z. Prediction Method of Ground Settlement for Rectangular Tunnel Construction. Tunn. Undergr. Space Technol. 2025, 164, 106814. [Google Scholar] [CrossRef]
  31. Liang, J.; Liu, W.; Yin, X.; Li, W.; Yang, Z.; Yang, J. Experimental Study on the Performance of Shield Tunnel Tail Grout in Ground. Undergr. Space 2025, 20, 277–292. [Google Scholar] [CrossRef]
  32. Saunders, M.; Lewis, P.; Thornhill, A. Research Methods for Business Students. Qual. Mark. Res. Int. J. 2000, 3, 215–218. Available online: https://www.researchgate.net/publication/240218229_Research_Methods_for_Business_Students#full-text (accessed on 10 July 2025). [CrossRef]
  33. Lorenc, A.; Szarata, J.; Czuba, M. Real-Time Location System (RTLS) Based on the Bluetooth Technology for Internal Logistics. Sustainability 2023, 15, 4976. [Google Scholar] [CrossRef]
  34. Šinko, S.; Marinič, E.; Poljanec, B.; Obrecht, M.; Gajšek, B. Performance-Oriented UWB RTLS Decision-Making Approach. Sustainability 2022, 14, 11456. [Google Scholar] [CrossRef]
  35. Shamsollahi, D.; Moselhi, O.; Khorasani, K. Data Integration Using Deep Learning and Real-Time Locating System (RTLS) for Automated Construction Progress Monitoring and Reporting. Autom. Constr. 2024, 168, 105778. [Google Scholar] [CrossRef]
  36. Gavaskar, D. Compact and Cost-Effective Object Tracking System Using Arduino Nano, SIM800L Module, and Multiplexer. Int. J. Res. Appl. Sci. Eng. Technol. 2023, 11, 2548–2554. [Google Scholar] [CrossRef]
  37. Adekunle, A.; Oluwadamilare, A.; Ayo, F. Distance Measurement and Energy Conservation Using Arduino Nano and Ultrasonic Sensor. Am. J. Electr. Comput. Eng. 2021, 5, 40. [Google Scholar] [CrossRef]
  38. Nayak, J. A Review on LoRa Transmission. Int. J. Eng. Res. Technol. 2022, 10, 3. [Google Scholar] [CrossRef]
  39. Augustin, A.; Yi, J.; Clausen, T.H.; Townsley, W. A Study of LoRa: Long Range & Low Power Networks for the Internet of Things. Sensors 2016, 16, 1466. [Google Scholar] [CrossRef]
  40. Hercog, D.; Lerher, T.; Truntič, M.; Težak, O. Design and Implementation of ESP32-Based IoT Devices. Sensors 2023, 23, 6739. [Google Scholar] [CrossRef] [PubMed]
  41. Balarabe, A.; Hassan, Z.L. An Arduino UNO Based Environment Monitoring System. IOSR J. Electron. Commun. Eng. 2019, 14, 4–9. [Google Scholar]
  42. Patel, V.; Chesmore, A.; Legner, C.M.; Pandey, S. Trends in Workplace Wearable Technologies and Connected-Worker Solutions for Next-Generation Occupational Safety, Health, and Productivity. Adv. Intell. Syst. 2022, 4, 2100099. [Google Scholar] [CrossRef]
  43. Pisu, A.; Elia, N.; Pompianu, L.; Barchi, F.; Acquaviva, A.; Carta, S. Enhancing Workplace Safety: A Flexible Approach for Personal Protective Equipment Monitoring. Expert Syst. Appl. 2024, 238, 122285. [Google Scholar] [CrossRef]
  44. Jean, F.; Khan, M.U.; Alazzam, A.; Mohammad, B. Harnessing Ambient Sound: Different Approaches to Acoustic Energy Harvesting Using Triboelectric Nanogenerators. J. Sci. Adv. Mater. Devices 2024, 9, 100805. [Google Scholar] [CrossRef]
  45. Zhou, G.; Xu, J.; Hu, H.; Liu, Z.; Zhang, H.; Xu, C.; Zhou, X.; Yang, J.; Nong, X.; Song, B.; et al. Off-Axis Four-Reflection Optical Structure for Lightweight Single-Band Bathymetric LiDAR. IEEE Trans. Geosci. Remote Sens. 2023, 61, 1–17. [Google Scholar] [CrossRef]
  46. Syafiza, I.; Hanani, A. Development of Real-Time Indoor Human Tracking System Using LoRa Technology. Int. J. Electr. Comput. Eng. 2022, 12, 845. [Google Scholar] [CrossRef]
  47. Tran, S.V.; Lee, D.; Bao, Q.L.; Yoo, T.; Khan, M.; Jo, J.; Park, C. A Human Detection Approach for Intrusion in Hazardous Areas Using 4D-BIM-Based Spatial-Temporal Analysis and Computer Vision. Buildings 2023, 13, 2313. [Google Scholar] [CrossRef]
  48. Feng, J.; Turksoy, K.; Samadi, S.; Hajizadeh, I.; Littlejohn, E.; Cinar, A. Hybrid Online Sensor Error Detection and Functional Redundancy for Systems with Time-Varying Parameters. J. Process Control 2017, 60, 115–127. [Google Scholar] [CrossRef]
  49. Zhuang, Y.; Sun, X.; Li, Y.; Huai, J.; Hua, L.; Yang, X.; Cao, X.; Zhang, P.; Cao, Y.; Qi, L.; et al. Multi-Sensor Integrated Navigation/Positioning Systems Using Data Fusion: From Analytics-Based to Learning-Based Approaches. Inf. Fusion 2023, 95, 62–90. [Google Scholar] [CrossRef]
  50. Isaia, C.; Michaelides, M.P. A Review of Wireless Positioning Techniques and Technologies: From Smart Sensors to 6G. Signals 2023, 4, 90–136. [Google Scholar] [CrossRef]
  51. Gupta, A.; Harit, V. Child Safety & Tracking Management System by Using GPS, Geo-Fencing & Android Application: An Analysis. In Proceedings of the 2016 Second International Conference on Computational Intelligence & Communication Technology (CICT), Ghaziabad, India, 12–13 February 2016. [Google Scholar] [CrossRef]
  52. Baek, J.; Choi, Y. Bluetooth-Beacon-Based Underground Proximity Warning System for Preventing Collisions inside Tunnels. Appl. Sci. 2018, 8, 271. [Google Scholar] [CrossRef]
  53. Hollweck, T.; Robert, K.Y. Case Study Research Design and Methods (5th Ed.). Thousand Oaks, CA: Sage. 282 Pages. Can. J. Progr. Eval. 2016, 30, 282. [Google Scholar] [CrossRef]
  54. Baxter, P.; Jack, S. Qualitative Case Study Methodology: Study Design and Implementation for Novice Researchers. Qual. Rep. 2010, 13, 544–559. [Google Scholar] [CrossRef]
  55. Liang, Y.; Liu, Q. Early Warning and Real-Time Control of Construction Safety Risk of Underground Engineering Based on Building Information Modeling and Internet of Things. Neural Comput. Appl. 2022, 34, 1–10. [Google Scholar] [CrossRef]
  56. Rao, A.; Radanović, M.; Liu, Y.; Hu, S.; Fang, Y.; Khoshelham, K.; Palaniswami, M.; Ngo, T. Real-Time Monitoring of Construction Sites: Sensors, Methods, and Applications. Autom. Constr. 2022, 136, 104099. [Google Scholar] [CrossRef]
  57. Wang, X.; Jinxing, L.; Junling, Q.; Wei, X.; Lixin, W.; Luo, Y. Geohazards, Reflection and Challenges in Mountain Tunnel Construction of China: A Data Collection from 2002 to 2018. Geomat. Nat. Hazards Risk 2020, 11, 766–785. [Google Scholar] [CrossRef]
  58. Cai, G.; Zheng, X.; Gao, W.; Guo, J. Self-Extinction Characteristics of Fire Extinguishing Induced by Nitrogen Injection Rescue in an Enclosed Urban Utility Tunnel. Case Stud. Therm. Eng. 2024, 59, 104478. [Google Scholar] [CrossRef]
  59. Da, H.; Xuejuan, X.; Junjie, H.; Kai, Q.; Yongsuo, L.; Xian, Y.; Xiaoqiang, L. Calculation Methods for the Jacking Force of a Rectangular Pipe Jacking Tunnel: Overview and Prospects. J. Pipeline Syst. Eng. Pract. 2025, 16, 3125001. [Google Scholar] [CrossRef]
  60. Li, D.; Nie, J.-H.; Wang, H.; Yu, T.; Kuang, K.S.C. Path Planning and Topology-Aided Acoustic Emission Damage Localization in High-Strength Bolt Connections of Bridges. Eng. Struct. 2025, 332, 120103. [Google Scholar] [CrossRef]
  61. Qi, H.; Zhou, Z.; Manu, P.; Li, N. Falling Risk Analysis at Workplaces through an Accident Data-Driven Approach Based upon Hybrid Artificial Intelligence (AI) Techniques. Saf. Sci. 2025, 185, 106814. [Google Scholar] [CrossRef]
  62. Chen, D.; Chen, Y.; Zhou, Z.; Tu, W.; Li, L. Study on Internal Rise Law of Fracture Water Pressure and Progressive Fracture Mechanism of Rock Mass under Blasting Impact. Tunn. Undergr. Space Technol. 2025, 161, 106545. [Google Scholar] [CrossRef]
Figure 1. Research methodology workflow for developing proximity warning system in tunnel construction.
Figure 1. Research methodology workflow for developing proximity warning system in tunnel construction.
Buildings 15 02667 g001
Figure 2. Key electrical components of the RTLS system: (a) Arduino nano, (b) LoRa module, (c) ESP32, (d) jumper wires, (e) OLED display, and (f) piezo buzzer.
Figure 2. Key electrical components of the RTLS system: (a) Arduino nano, (b) LoRa module, (c) ESP32, (d) jumper wires, (e) OLED display, and (f) piezo buzzer.
Buildings 15 02667 g002
Figure 3. Wiring and layout for bidirectional data exchange: (a) Arduino Nano and ESP32 module, (b) Arduino Nano and ESP32 module connections.
Figure 3. Wiring and layout for bidirectional data exchange: (a) Arduino Nano and ESP32 module, (b) Arduino Nano and ESP32 module connections.
Buildings 15 02667 g003
Figure 4. Design and connection schema for enabling long-range communication system: (a) Arduino Nano and LoRa module Altium design, (b) Arduino Nano and LoRa module connections.
Figure 4. Design and connection schema for enabling long-range communication system: (a) Arduino Nano and LoRa module Altium design, (b) Arduino Nano and LoRa module connections.
Buildings 15 02667 g004
Figure 5. Component integration for a complete prototype showing (a) transmitter and (b) receiver.
Figure 5. Component integration for a complete prototype showing (a) transmitter and (b) receiver.
Buildings 15 02667 g005
Figure 6. Pseudo-code outlining the logic for GPS data collection, proximity computation, and alert generation.
Figure 6. Pseudo-code outlining the logic for GPS data collection, proximity computation, and alert generation.
Buildings 15 02667 g006
Figure 7. Pseudo-code of receiver illustrating the logic for data collection, proximity computation, and alert generation.
Figure 7. Pseudo-code of receiver illustrating the logic for data collection, proximity computation, and alert generation.
Buildings 15 02667 g007
Figure 8. Example of static proximity risks: (a) danger area location within the corridor simulating a tunnel environment and (b) route of transmitter movement toward the hazard zone.
Figure 8. Example of static proximity risks: (a) danger area location within the corridor simulating a tunnel environment and (b) route of transmitter movement toward the hazard zone.
Buildings 15 02667 g008
Figure 9. Example of dynamic proximity risk: (a) worker–equipment interaction and (b) volunteer movement path.
Figure 9. Example of dynamic proximity risk: (a) worker–equipment interaction and (b) volunteer movement path.
Buildings 15 02667 g009
Figure 10. Visual representation of static electric shock hazard scenario: (a) identified danger zone with electrical risk and (b) route followed by the transmitter.
Figure 10. Visual representation of static electric shock hazard scenario: (a) identified danger zone with electrical risk and (b) route followed by the transmitter.
Buildings 15 02667 g010
Figure 11. Time delay tracking in actual vs. expected.
Figure 11. Time delay tracking in actual vs. expected.
Buildings 15 02667 g011
Figure 12. Delay in proximity distance.
Figure 12. Delay in proximity distance.
Buildings 15 02667 g012
Table 1. Arduino Nano and buzzer connections.
Table 1. Arduino Nano and buzzer connections.
Arduino NanoBuzzer
D10Long leg
GNDShort leg
Table 2. Proximity distance measurements at different locations and time intervals during static risk assessment.
Table 2. Proximity distance measurements at different locations and time intervals during static risk assessment.
LatitudeLongitudeTimeProximity Distance
34.06906172.6405335:16:5514.89639
34.06906172.6405335:16:5714.89639
34.06906172.6405335:16:5914.89639
34.06906172.6405335:17:0114.89639
34.06906172.6405335:17:0314.89639
34.06906172.6405335:17:0514.89639
34.06908772.640485:17:0711.10634
34.06909972.6405025:17:099.95637
34.06915672.6405335:17:116.86641
34.06914972.6405185:17:126.46639
Table 3. Proximity data and corresponding warning triggers during interaction between worker (transmitter) and equipment (receiver) in a dynamic scenario.
Table 3. Proximity data and corresponding warning triggers during interaction between worker (transmitter) and equipment (receiver) in a dynamic scenario.
LatitudeLongitudeTimeProximity DistanceWarning
34.0692672.6404335:16:5515.89639-
34.0692672.6404335:16:5714.89639-
34.0692672.6404335:16:5914.89639-
34.0692672.6404335:17:0114.89639-
34.0692672.6404335:17:0313.89639-
34.0692672.6404335:17:0512.89639-
34.0692972.640485:17:0711.10634-
34.069372.640495:17:099.95637-
34.0692872.6404995:17:107.80641Warning on buzzer
34.0692572.6405185:17:126.46639Warning on buzzer
34.0692272.6404955:17:148.10636Warning on buzzer
34.0692172.6404875:17:158.43636Warning on buzzer
Table 4. Data of observed proximity during simulation of an electrical hazard scenario, as transmitter approaches electric shock hazard.
Table 4. Data of observed proximity during simulation of an electrical hazard scenario, as transmitter approaches electric shock hazard.
LatitudeLongitudeTimeProximity Distance
34.06906172.6405335:16:5514.89639
34.06906172.6405335:16:5714.89639
34.06906172.6405335:16:5914.89639
34.06906172.6405335:17:0114.89639
34.06906172.6405335:17:0314.89639
34.06906172.6405335:17:0514.89639
34.06908772.640485:17:0711.10634
34.06909972.6405025:17:099.95637
34.06915672.6405335:17:116.86641
34.06914972.6405185:17:126.46639
Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Share and Cite

MDPI and ACS Style

Afzal, F.; Khan, F.U.; Khan, A.A.; Jayasinghe, R.; Khan, N. RTLS-Enabled Bidirectional Alert System for Proximity Risk Mitigation in Tunnel Environments. Buildings 2025, 15, 2667. https://doi.org/10.3390/buildings15152667

AMA Style

Afzal F, Khan FU, Khan AA, Jayasinghe R, Khan N. RTLS-Enabled Bidirectional Alert System for Proximity Risk Mitigation in Tunnel Environments. Buildings. 2025; 15(15):2667. https://doi.org/10.3390/buildings15152667

Chicago/Turabian Style

Afzal, Fatima, Farhad Ullah Khan, Ayaz Ahmad Khan, Ruchini Jayasinghe, and Numan Khan. 2025. "RTLS-Enabled Bidirectional Alert System for Proximity Risk Mitigation in Tunnel Environments" Buildings 15, no. 15: 2667. https://doi.org/10.3390/buildings15152667

APA Style

Afzal, F., Khan, F. U., Khan, A. A., Jayasinghe, R., & Khan, N. (2025). RTLS-Enabled Bidirectional Alert System for Proximity Risk Mitigation in Tunnel Environments. Buildings, 15(15), 2667. https://doi.org/10.3390/buildings15152667

Note that from the first issue of 2016, this journal uses article numbers instead of page numbers. See further details here.

Article Metrics

Back to TopTop