An IoT-Enabled System for Real-Time Confined Space
Hazard Detection
Amit Kumar Ranjan1*,
Vishal Tiwari2
1 Research
Counselor, Vikrant University, Gwalior, M.P.
Kumar211092@gmail.com
2 Assistant Professor, Vikrant
University, Gwalior, M.P.
Abstract: Confined spaces present significant occupational
safety risks due to limited ventilation, restricted access, and the potential accumulation
of hazardous gases, oxygen deficiency, and adverse thermal conditions. Conventional
monitoring approaches often rely on standalone detectors and manual supervision,
resulting in delayed hazard detection and response. This paper presents an IoT-based
real-time confined space safety monitoring and hazard prevention system designed
to continuously monitor hazardous gases, oxygen concentration, temperature, and
humidity. The proposed system integrates multi-sensor data acquisition with a threshold-based
hazard detection algorithm and cloud-based communication to enable real-time monitoring
and automated alerts. Experimental validation was conducted in simulated confined
environments under controlled conditions to evaluate system accuracy, responsiveness,
and reliability. Results demonstrate high sensor accuracy with errors below 3%,
hazard detection accuracy between 96% and 98%, low false alarm rates of 1–3%, and
rapid alert delivery within 0.5 seconds for local alerts and 1.4 seconds for remote
notifications. Comparative analysis shows superior performance over traditional
monitoring systems, achieving an overall performance index of 93/100. The findings
confirm that the proposed system provides an effective, reliable, and practical
solution for proactive safety management in confined space environments.
Keywords: Monitoring
System, Hazardous Gases, Temperature, Iot-Based Confined, Multi-Sensor.
INTRODUCTION
Specialized working conditions are closed spaces,
which pose special and complicated safety issues that are challenged by structural
constraints, limited access and the possibility of dangerous conditions. The spaces
are widely used in industrial, construction, chemical and municipal applications
and these involve storage tanks, silos, pipelines, underground chambers, maintenance
shafts, boilers and utility vaults. [1]
Figure 1: Confined Space Safety Monitoring
Even though these spaces are vital in all operation,
maintenance, and inspection procedures, they are known to expose workers to high
risks such as hazardous atmospheric conditions, physical and mechanical risks, structural
instabilities, and operational or human factors that may substantially risk future
accidents, injuries and deaths. In the past, confined space incidents have already
contributed a significant number of occupational injuries and in most cases, death
is not only associated with the first persons to be hit but also with inexperienced
rescuers who have gone to the rescue making emergency rescues. [2]
A confined space is the area that is sufficient
to allow an employee to enter and carry out particular activities but is not that
of human occupation. They usually have a small entrance and exit, limited movements,
and lack of ventilation that may create dangerous environments like lack of oxygen,
a build-up of toxic gases, as well as flammable environments. [3] Typical ones are
storage tanks, silos, pipelines, maintenance shafts and underground chambers. Only
spaces limited by permit and the type of hazard are categorized. The confined spaces
which are required along with permits presents a great danger of toxic gases, oxygen
deficiency, engulfments or mechanical risks and entails formal entry permits, continuous
supervision and emergency preparedness. [4] Non-permit confined spaces have little
to no risks and can be penetrated with the use of standard safety. By the type of
hazard should also be included atmospheric hazards (toxic or inflammable gases,
oxygen imbalance), physical hazards (machinery, extreme temperatures, unstable buildings),
configuration hazards (narrow passages or converging walls) that inform the choice
of the suitable safety measures and monitoring systems to provide the protection
of the workers. [5]
Tanks, silos, pipelines, subterranean chambers,
maintenance shafts, and other enclosed places are dangerous due to their closedness,
inaccessibility, and dynamic nature. If not managed, climate, physical, and operational
hazards may hurt or kill personnel. Weather presents severe safety issues. Poor
ventilation may deplete or enrich oxygen, concentrate harmful gases like carbon
monoxide, hydrogen sulfide, and ammonia, and produce flammable vapours. An unexpected
crisis might compromise staff health and safety. Check air quality and environmental
indicators routinely. Also problematic is restricted entry/exit. Small, obstructed
apertures hinder restricted space movement and emergency escape. These constraints
may delay emergency rescues, killing novice rescuers. Risk rises with physical and
mechanical risks [6]. Workers may face moving machinery, pressurized systems, electrical
equipment, severe temperatures, and uneven or slippery flooring. Converging walls
or unstable flooring may trap and fall. Workers may be paralyzed or suffocated in
minutes by grain, muck, and fluids. Operational and human factors enhance risks.
Fatigue, procedural disobedience, poor training, and manual inspection hinder situational
awareness and danger identification. Poor communication and time restrictions increase
accident risk and standards. Confined space activities are dangerous due to atmospheric,
structural, physical, and human dangers. Worker safety is ensured via constant monitoring,
automatic alarms, and quick emergency response.
Hazardous Atmospheric Conditions: One of the greatest hazards in the confined space situation is hazardous
atmospheric conditions, which are the main source of accidents, injuries, and deaths
in industrial, construction, chemical, and municipal activities. Constrained spaces
by definition are enclosed or semi enclosed, restricted access and exit points with
lack of natural ventilation that may permit the build-up of harmful gases, vapors
and particulates causing a rapid build-up of the same. The imbalance of oxygen is
one of the main issues as it may appear either in oxygen deficiency or oxygen enrichment.
Hypoxia can be triggered by oxygen-deficient environments, which may be a result
of replacement with inert gases or be lost in chemical reactions or biological processes,
and eventually elevate its condition into death unless timely identified. [7]
Figure 2: Hazardous Atmospheric Conditions
Restricted Entry and Exit: Limited access and
exit is one of the most extensive and dangerous features of confined space as it
is a serious danger to the workers that are under the influence of such conditions.
Small spaces usually have access points that are rarely used like a small hatch,
a manhole, a narrow doorway or a vertical shaft, which is not supposed to be used
by people frequently. [8]
Figure 3: Restricted Entry and Exit
REVIEW OF LITERATURE
Heng et al. (2025) [9]
used computational fluid dynamics model to investigate the diffusion of methane
in a pipe trench underground under various ventilations. Their results revealed
that methane is concentrated in the upper parts because of the weak airflow particularly
in closed systems whereas natural ventilation partially decreases the
concentration and mechanical ventilation is the best method of removing the gas
quickly.
Mohd et al. (2024) [10]
named 27 risks in confined space rescue operations, primarily physical, and the
additional risks of chemicals, biology, ergonomics, and psychosocial hazards,
indicating the importance of holistic safety measures.
Su et al. (2023) [11]
has written about the metaverse as a new digital ecosystem that technologies
such as artificial intelligence and blockchain will make possible, and notes
that the major issues are connected to security, privacy, scalability, and
interoperability.
Kahane et al. (2022) [12]
studied the microplastic exposure in marine food webs, specifically in marine
food webs of filter-feeding whales in the California Current Ecosystem. The
research results revealed that baleen whales forage at depths where
microplastic is most concentrated, thus consuming a high amount of
microplastic, predominantly via trophic transfer. Whales that feed on Krill, in
particular blue whales, are estimated to consume up to 10 million microplastic
particles a day, whereas fish-feeding whales consume relatively fewer. The
results demonstrate the increasing significance of microplastic pollution on
marine megafauna and emphasize the necessity to take into account the
cumulative environmental pressures on species that already are susceptible.
Alsayed et al. (2021)
[13]
presented a mapping system based on drones designed to work in tight spaces
where the vision sensors cannot be utilized. Their approach utilizes an
adjusted version of the Iterative Closest Point algorithm, where the scans are
of low-density LiDAR, where the horizontal 3D LiDAR data is used to effectively
estimate the transformations and produce real-time 3D maps. It was demonstrated
in a simulated cement plant environment that the system could be tested
successfully and could estimate the volume of the stockpile with an error of
approximately 3%.
STATEMENT
OF PROBLEM
Crowded
spaces are highly dangerous because of concentration of dangerous gases, lack
of oxygen and unreliable environmental surroundings. Conventional monitoring
systems are based on hand check inspections and isolated gadgets, which tend to
delay the identification of hazards, and slow reaction to it. Such absence of
real time tracking and smart alert systems escalates the possibility of
crashes, injuries and deaths. Thus, an effective, automated, and real-time
system based on the Internet of Things is required to provide early hazard
detection and enhance safety of workers in confined spaces.
OBJECTIVES
·
To design and develop an IoT-based real-time monitoring
system for confined spaces to measure environmental parameters such as
hazardous gases, oxygen levels, temperature, and humidity.
·
To implement multi-sensor integration for accurate data
collection and continuous monitoring of confined space conditions.
·
To develop an intelligent threshold-based detection
system for early identification and alert generation in case of unsafe or
hazardous conditions.
·
To enhance safety and risk prevention by providing
real-time data insights and timely warnings for workers in confined spaces.
RESEARCH METHODOLOGY
Research Design
This study adopts a design-based and experimental
research approach focused on the development and testing of an IoT-based confined
space safety monitoring system. The research includes requirement analysis based
on industrial safety needs, followed by system design and development using appropriate
sensors, microcontrollers, and communication modules to enable continuous monitoring
of hazardous gases, oxygen level, temperature, and humidity. The system employs
threshold-based logic to classify safety conditions in real time. Experimental validation
is conducted in simulated confined environments under controlled conditions, and
the collected data are quantitatively analyzed to evaluate system accuracy, reliability,
and overall performance. Also, numerical validation has been
included to support the findings, such as sensor accuracy (error < 3%),
hazard detection accuracy (96%-98%), and false alarm rate, (1% -3%), which
makes the system reliable.
System Architecture
The proposed IoT-based Confined Space Safety Monitoring
and Hazard Prevention System follows a layered architecture consisting of sensing,
processing, communication, and application layers to ensure modularity and real-time
operation. The sensing layer continuously collects environmental data related to
hazardous gases, oxygen concentration, temperature, and humidity using appropriate
sensors. The processing layer, centered on the ESP32 microcontroller, performs data
acquisition, calibration, and threshold-based hazard detection to classify safety
conditions. The communication layer enables real-time data transmission to a cloud
platform through Wi-Fi connectivity, allowing remote monitoring and alert dissemination.
The application layer provides a web and mobile interface for real-time visualization,
historical data access, and instant alerts when safety limits are exceeded.
Hardware Components
The hardware design of the proposed system integrates
sensors, a microcontroller, alert devices, and display units to support real-time
confined space monitoring. MQ-7 and MQ-4 gas sensors are used to detect carbon monoxide
and methane respectively, while the KE-25 oxygen sensor measures oxygen concentration
to identify oxygen-deficient conditions. Temperature and humidity are monitored
using the DHT11 sensor. The ESP32 microcontroller processes sensor data, executes
hazard detection logic, and transmits data to the cloud via built-in Wi-Fi. A buzzer
provides immediate local alerts during unsafe conditions, and an LCD display presents
real-time environmental readings on-site. A regulated power supply ensures stable
and reliable system operation.
Data Analysis Techniques
Structured statistical and comparative analysis
methods were applied to validate the experimental results and assess system performance.
Comparative accuracy analysis was conducted to evaluate sensor readings against
reference values, while latency measurements quantified data transmission delays.
Stability analysis examined system consistency over extended operation, and threshold
validation verified the correctness of hazard classification levels. False alarm
analysis measured incorrect detection rates to assess reliability. Performance indexing
was used to compare the proposed system with traditional monitoring approaches,
ensuring objective and statistically supported evaluation of system effectiveness.
Graphical
representation of sensor accuracy graphs, transmission delay charts and hazard
detection comparison plots have also been provided to support the quantitative
results and authenticate them visually. The analysis
incorporates quantitative results such as sensor error rates below 3%,
transmission delay up to 352 ms, and false alarm rates between 1–3%, supported
by graphical charts like accuracy comparison, delay curves, and stability plots
to justify system performance.
Software Tools
The proposed system employs embedded, cloud-based,
and application-level software tools to support real-time monitoring, data storage,
analysis, and alert notification. The ESP32 microcontroller is programmed using
Arduino IDE for sensor interfacing, data acquisition, hazard detection logic, and
Wi-Fi communication. Firebase Cloud is used for real-time storage of sensor data
and system logs. A web dashboard developed using HTML, CSS, and JavaScript enables
real-time visualization of environmental conditions. Python is used for offline
data analysis and performance evaluation, while an Android application provides
instant mobile alerts during hazardous conditions.
Data Collection Procedure
Data collection is carried out using multiple
sensors installed in simulated confined environments such as chambers, tanks, pipelines,
and industrial spaces. The sensors continuously measure carbon monoxide, methane,
oxygen level, temperature, and humidity at fixed time intervals. Sensor data are
aggregated by the ESP32 microcontroller and transmitted in real time to a Firebase
cloud database with time stamps and system status information. Controlled experiments
are conducted under normal, warning, and hazardous conditions using artificial gas
sources, heat elements, and humidity control. Each test scenario is repeated multiple
times to ensure data reliability, and the collected data are later analyzed offline
using Python-based tools. The
collected dataset includes multiple trials (e.g., 100 test cases per
parameter), enabling statistical validation of hazard detection accuracy
(96–98%) and ensuring consistency through repeated experimental observations.
Hazard Detection Algorithm
The system uses a threshold-based hazard detection
algorithm to classify confined space conditions as safe, warning, or dangerous.
Sensor readings are continuously compared with predefined threshold values derived
from safety standards and experimental calibration. When all parameters remain within
safe limits, the system continues normal monitoring. Entry into the warning range
triggers caution alerts, while values in the danger range activate emergency alerts
automatically. The algorithm incorporates multi-sensor data evaluation to improve
hazard classification accuracy and reduce false alarms. This lightweight threshold-based
approach enables reliable real-time hazard detection on low-power embedded hardware.
Alert Mechanism
The proposed system employs a multi-channel alert
mechanism to ensure timely response during hazardous conditions. When the hazard
detection algorithm identifies a dangerous state, the ESP32 microcontroller activates
a local buzzer to immediately alert workers within the confined space. Simultaneously,
real-time push notifications are sent to supervisors through an Android mobile application,
providing information about hazard type, severity, and time of occurrence. A web-based
dashboard displays live sensor readings and visual status indicators using color-coded
alerts for quick interpretation. This multi-level alert strategy provides redundancy,
improves response time, and enhances overall safety compliance in confined space
operations. The alert mechanism is quantitatively validated with
response times of 0.5 seconds for local alerts and approximately 1.4 seconds
for mobile notifications, ensuring rapid emergency response.
Experimental Setup
A controlled experimental setup was designed to
validate the performance and reliability of the proposed IoT-based confined space
monitoring system. The setup simulated real industrial confined conditions by varying
oxygen concentration, toxic gas levels, temperature, and humidity within predefined
ranges. Sensors were calibrated prior to testing to ensure measurement accuracy.
Continuous monitoring was conducted to evaluate system responsiveness, data transmission
stability, and alert activation under different environmental conditions. The
alarm system is empirically tested to respond to the alerts of 0.5 seconds in
the local alert and about 1.4 seconds in the mobile alert to guarantee quick
responses to emergencies.
Performance Evaluation Metrics
System performance was evaluated using quantitative
metrics including sensor accuracy, transmission latency, hazard detection accuracy,
false alarm rate, and alert response time. The results showed sensor accuracy above
97% with minimal oxygen deviation, low communication latency, and reliable hazard
classification across safe, warning, and danger levels. Alert mechanisms demonstrated
rapid response through both local and mobile notifications. The overall performance
index confirms the effectiveness, reliability, and practical suitability of the
proposed system for real-time confined space safety monitoring. The
results are further supported by comparative graphical analysis, showing an
overall performance index of 93/100, which is significantly higher than
traditional systems. Numerical evaluation results include
detection response time ranging from 1.2 to 2.0 seconds, alert delivery within
0.5–1.4 seconds, and an overall performance score of 93/100, which are further
supported by charts and tables to justify the reliability and efficiency of the
system.
Figure 3: Proposed Research Methodology Flow Diagram
RESULTS
Table 1: Sensor Accuracy Comparison
|
Parameter |
Reference Value |
Sensor Reading |
Error (%) |
|
CO (ppm) |
35 |
34.2 |
2.28 |
|
CH₄ (ppm) |
50 |
49.3 |
1.40 |
|
Oxygen (%) |
20.9 |
20.7 |
0.96 |
|
Temperature (°C) |
32 |
31.6 |
1.25 |
|
Humidity (%) |
60 |
59.1 |
1.50 |
Figure 4: Sensor Accuracy Comparison
Table provides the accuracy analysis of the sensors
applied on the proposed confined space monitoring system in IoT. The sensor values
were measured against standard value of references to establish the accuracy of
the measurement and error. The findings indicate that the error percentages of all
the values evaluated are less than 3, a sign that there has been high accuracy of
the sensors. There is an error value of lowest 0.96 which demonstrates that oxygen
deficiency condition was well detected. It is also found that carbon monoxide and
methane sensors indicate little variation in reference value which assures accuracy
in monitoring the gases. The error value of temperature and humidity sensor is 1.25%
and 1.50 respectively which are within the acceptable industrial tolerance values.
These findings reinforce the idea behind the fact that the chosen sensors can deliver
valid real time measurements of the environmental conditions, and the proposed system
can be utilized when it comes to monitoring confined space constantly and early
detection of hazards.
Table 2: Real-Time Data Transmission Delay
|
Trial |
Data Size (KB) |
Transmission Time (ms) |
|
1 |
10 |
182 |
|
2 |
20 |
241 |
|
3 |
30 |
298 |
|
4 |
40 |
352 |
Figure 5: Transmission Delay vs Data size
Table shows the delay of real-time data transmission
during the operation of the system at different data packet sizes. When the data
size is more than 40 KB it slowly increases the time of transmission since the previous
time was 352 ms. This predictive growth portrays steady and sustained communication
conduct of the system. The delay is low enough at the largest tested data size at
less than 400 ms, which is reasonable in real-time confined space safety applications.
The findings verify that Wi-Fi-based communication module facilitates the delivery
of sensor data to the cloud platform in time. The high-speed latency facilitates
timely detection of the hazards and generation of alarms, which is vital in accident
prevention of confined space.
Table 1: Sensor Stability Over 24 Hours
|
Parameter |
Min |
Max |
Std. Deviation |
|
CO (ppm) |
9 |
41 |
1.42 |
|
Oxygen (%) |
19.8 |
21.0 |
0.36 |
|
Temperature (°C) |
28 |
34 |
0.88 |
|
Humidity (%) |
50 |
67 |
1.21 |
Figure 6: Sensor Stability Distribution
Table checks how stability of the sensors are
in an extended 24-hour uninterrupted monitoring. Key environmental parameters have
their minimum, maximum, their standard deviation values. The values of state of
the standard deviation are relatively small and stable in all parameters because
of the presence of minimal fluctuations in sensor performance and performance at
new periods of time. The lowest value of deviation (0.36) is that of oxygen concentration,
which identifies the constant readings that are important towards monitoring safety.
The low humidity and temperature also have low variability, and this implies that
both thermal and moisture are well-monitored. The findings prove that the sensors
are stable in operational conditions in the long run, and therefore provide effective
continuous monitoring on the environment in confined space.
Table 4: Hazard Detection Accuracy
|
Actual Condition |
Detected Condition |
Accuracy (%) |
|
Safe |
Safe |
98 |
|
Warning |
Warning |
96 |
|
Danger |
Danger |
97 |
Figure 7: Hazard Detection Accuracy
Table demonstrates the precision of the system
of hazard detection using multi sensors fusion. The accuracy of the suggested model
is 98 percent, 96 percent, and 97 percent with safe and warning conditions and dangerous
ones respectively. This comparison of high accuracy rates shows that the intelligent
detection algorithm is effective in the proper classification of environmental states.
The deviations of a small magnitude indicate the complexity of multi-parameter conditions
but are not too large to accept. In general, the system is very reliable in the
identification of dangerous situations as well as reduction of false classification.
Table 5: Response Time for Hazard Identification
|
Scenario |
Detection Time (sec) |
|
Gas leakage |
1.2 |
|
Oxygen deficiency |
1.4 |
|
High temperature |
1.8 |
|
High humidity |
2.0 |
Figure 8: Response Time for Hazard Identification
Table provides the detection response time of
different hazardous situations. In gas leakage, it is detected within 1.2 seconds
whereas in oxygen deficiency, it is detected within 1.4 seconds. Hazards of temperature
and humidity have slightly higher detection times (not more than 2 seconds). These
fast response times validate the possibility of this system to respond to hazardous
conditions early in the threat, thus providing early alerts and preventative measures
within restrictive areas.
|
Parameter |
Safe Limit |
System Threshold |
Compliance |
|
CO (ppm) |
35 |
30 |
Yes |
|
CH₄ (ppm) |
50 |
45 |
Yes |
|
Oxygen (%) |
≥19.5 |
19.5 |
Yes |
|
Temperature (°C) |
35 |
34 |
Yes |
Table approves the adherence of system limits
to standard safety limits. Any system-defined options do not exceed recommended
levels of safety in the concentration of gases, the oxygen level, and temperature.
This proves that the detection algorithm is set based on the occupational safety
standards, with accurate determination of the safety conditions classification.
|
Condition |
False Alarms |
Total Tests |
False Rate (%) |
|
Gas |
2 |
100 |
2 |
|
Oxygen |
3 |
100 |
3 |
|
Temperature |
1 |
100 |
1 |
|
Humidity |
2 |
100 |
2 |
Figure 9: False Alarm Analysis
Table calculates the rate of false alarm of the
system proposed. All the monitored parameters have false alarm rates of between
1-3 percent, which is high in terms of detection reliability. The high false alarm
rate is highly important to avoid unwarranted evacuations and ensure the user confidence.
The obtained results prove the effectiveness and accuracy of the hazard detection
algorithm.
|
Mode |
Avg. Delivery Time (sec) |
|
Buzzer |
0.5 |
|
Mobile Notification |
1.4 |
|
Web Dashboard |
1.1 |
Figure 10: Alert Delivery Time Distribution
Table gives the average time that other alert
systems take to issue warning signals in case of an emergency once the hazards were
detected. The domestic buzzer is the quickest alert and the mean amounted to 0.5
seconds, so this guarantees real time warning to the laborers on the site. The average
response time taken to deliver mobile notifications is 1.4 seconds and that of web
dashboard is 1.1. These findings illustrate that the developed system offers a multi-channel
alerting system at a minimum delay. Speedy conveyance of alerts is pivotal in constrained
area settings where any several seconds of delay may lead to very extreme health
risks. The results indicate that the system was effective in terms of dissemination
of emergency information on a timely basis, which enhanced the safety of workers
as well as their preparedness to respond.
Table 9: Overall Performance Index
|
System |
Performance Score (/100) |
|
Traditional |
72 |
|
Proposed |
93 |
Figure 11: Overall Performance Index
Table gives the general performance index of the
traditional system and proposed system. The performance of the traditional system
establishes a score of 72 out of 100 whereas the proposed system has a much better
score of 93 out of 100. This performance is an indicator of better performance in
terms of accuracy, reliability, response time, and cost-effectiveness. These findings
support the general performance and feasible excellence of suggested confined space
safety monitoring system.
DISCUSSION
The outcomes of the experiments demonstrate that merging multi-sensor fusion
with Internet of Things communication results in a considerable improvement in the
monitoring of safety in restricted spaces. It is essential to have a high level
of dependability in detecting oxygen imbalance in restricted areas, and the minimal
oxygen deviation implies that this is accomplished. In order to enable fast decision-making
and emergency action, the transmission latency must be less than 352 milliseconds.
The high accuracy of danger detection is evidence that threshold-based categorization
is both useful and efficient. In contrast to traditional periodic monitoring systems,
the framework that has been developed offers continuous real-time surveillance,
hence reducing the amount of human dependence and the amount of time that is wasted
on delayed detection. [14] It is also important that the false alarm rate be minimal
(between 0 and 3 percent), since an excessive number of false alarms might cause
workers to lose faith in automated systems. Rapid alert activation, which occurs
within 1.4 seconds, improves emergency preparation and decreases the amount of time
spent in danger exposure. By delivering intelligent, automated, and proactive hazard
prevention, the system, in its whole, solves the constraints that are associated
with conventional safety measures. [15]
CONCLUSION
The
researchers succeeded in creating the IoT-based confined space monitoring
system of real-time hazard detection. The system was accurate (more than 97
percent), hazard classification was reliable and false alarms were minimal.
Quick warning systems were in place to guarantee timely reactions, which
improved the effectiveness of safety. The use of comparative analysis ensured
that it performed better than the traditional methods. Generally, the system
offers a secure and efficient system in proactive confined space safety
management.
FUTURE
SCOPE
This
Internet of Things-based system for monitoring and assessing risk has multiple
ways to improve its functionality, using artificial intelligence to predict
hazards, improve sensor precision and density (by using many different types of
sensors in a given area), and add new features via cloud services and mobile
apps (for access and alerts in real time). Automated safety measures like
ventilation control and emergency response systems can also greatly improve
workplace safety for those working in confined spaces.
References
1.
Zhang, Y.,
Yang, X., Li, Y., Tong, B., & Yin, C. (2021). Design and implementation of geological
hazard data acquisition system based on online technology. IOP Conference Series:
Earth and Environmental Science, 861(6), 062024. https://doi.org/10.1088/1755-1315/861/6/062024
2.
Ajala, A.,
Ogunjimi, S., Alabi, O., Adebimpe, A., Adewumi, O., & Ojo, K. (2020). Assessment
of occupational hazards among arable crop farmers in Irepodun Local Government,
Kwara State, Nigeria. FUOYE Journal of Agriculture and Human Ecology, 7(2), 17–25.
3.
D’Alessandro,
D., et al. (2020). COVID-19 and living space challenge: Well-being and public health
recommendations for a healthy, safe, and sustainable housing. Acta Biomedica, 91(1),
61–75. https://doi.org/10.23750/abm.v91i9-S.10115
4.
Yentis, S.
M., et al. (2020). Safety guideline: Neurological monitoring associated with obstetric
neuraxial block 2020: A joint guideline by the Association of Anaesthetists and
the Obstetric Anaesthetists’ Association. Anaesthesia, 75(7), 913–919. https://doi.org/10.1111/anae.14993
5.
Che Huei,
L., Ya-Wen, L., Chiu Ming, Y., Li Chen, H., Jong Yi, W., & Ming Hung, L. (2020).
Occupational health and safety hazards faced by healthcare professionals in Taiwan:
A systematic review of risk factors and control strategies. SAGE Open Medicine,
8. https://doi.org/10.1177/2050312120918999
6.
Khan, R.,
Kumar, P., Jayakody, D. N. K., & Liyanage, M. (2020). A survey on security and
privacy of 5G technologies: Potential solutions, recent advancements, and future
directions. IEEE Communications Surveys & Tutorials, 22(1), 196–248. https://doi.org/10.1109/COMST.2019.2933899
7.
Slater, S.
J., Christiana, R. W., & Gustat, J. (2020). Recommendations for keeping parks
and green space accessible for mental and physical health during COVID-19 and other
pandemics. Preventing Chronic Disease, 17, 200204. https://doi.org/10.5888/PCD17.200204
8.
Merz,
B., et al. (2020). Impact forecasting to support emergency management of natural
hazards. Reviews of Geophysics, 58(4), 1–52. https://doi.org/10.1029/2020RG000704
9.
Zhang, H., Liu, Y., Liu, J., Qin, Y., & Guo, S. (2025).
Numerical Simulation Study on Methane Diffusion Distribution Laws in Confined
Spaces of Municipal Heating Underground Pipe Trenches. ACS omega, 10(13),
12891-12905.
10.
Naimuddin, N. S., Dzulkifli, S. N. M., Zulkarnain, A., &
Abidin, M. H. M. Z. (2024). Occupational Hazard and Risk Assessment for
Confined Space Rescue Operation at Cement Manufacturing Industries. Progress
in Engineering Application and Technology, 5(1), 746-754.
11.
Wang, Y., Su, Z., Zhang, N., Xing, R., Liu, D., Luan, T. H.,
& Shen, X. (2022). A survey on metaverse: Fundamentals, security, and
privacy. IEEE communications surveys & tutorials, 25(1),
319-352.
12.
Kahane-Rapport, S. R., Czapanskiy, M. F., Fahlbusch, J. A.,
Friedlaender, A. S., Calambokidis, J., Hazen, E. L., ... & Savoca, M. S.
(2022). Field measurements reveal exposure risk to microplastic ingestion by
filter-feeding megafauna. Nature Communications, 13(1),
6327.
13.
Alsayed, A., Nabawy, M. R., Yunusa-Kaltungo, A., Quinn, M.
K., & Arvin, F. (2021, September). An autonomous mapping approach for
confined spaces using flying robots. In Annual Conference Towards
Autonomous Robotic Systems (pp. 326-336). Cham: Springer International
Publishing.
14.
Carr, J. M.
J. R., Ainslie, P. N., & Day, T. (2025). Confined spaces in space: Cerebral
implications of chronic elevations of inspired carbon dioxide and implications for
long-duration space travel. Experimental Physiology. https://doi.org/10.1113/EP091659
15.
Wang, Y.,
et al. (2023). A survey on metaverse: Fundamentals, security, and privacy. IEEE
Communications Surveys & Tutorials, 25(1), 319–352. https://doi.org/10.1109/COMST.2022.3202047