doi: 10.56294/hl2023242

ORIGINAL

Implementing Advanced Analytics in Occupational Health for Real-Time Risk Assessment

Aplicación de la analítica avanzada a la salud laboral para la evaluación de riesgos en tiempo real

Rajsinh V. Mohite¹ , Money Saxena² , Prakash Kumar Sahoo³ , Pooja Varma⁴

¹Krishna Institute of Medical Sciences, Krishna Vishwa Vidyapeeth “Deemed to be University”, Dept. of Preventive & Social Medicine. Taluka-Karad, Dist-Satara, Maharashtra, India.

²School of Nursing, Noida International University. Greater Noida, Uttar Pradesh, India.

³IMS and SUM Hospital, Siksha ‘O’ Anusandhan (Deemed to be University), Department of General Surgery. Bhubaneswar, Odisha, India.

⁴JAIN (Deemed-to-be University), Department of Psychology. Bangalore, Karnataka, India.

Cite as: Mohite RV, Saxena M, Kumar Sahoo P, Varma P. Implementing Advanced Analytics in Occupational Health for Real-Time Risk Assessment. Health Leadership and Quality of Life. 2023; 2:242. https://doi.org/10.56294/hl2023242

Submitted: 27-04-2023                   Revised: 12-07-2023                   Accepted: 11-10-2023                 Published: 12-10-2023

Editor: PhD. Prof. Neela Satheesh

ABSTRACT

Making ensuring that employees in all types of companies are safe and healthy mostly depends on occupational health. As companies strive for increased efficiency, it is rather crucial for them to identify and lower any potential hazards in the workplace. Though conventional methods of risk assessment have their uses, they are frequently sluggish, reactive, and unable of adjusting for changing work environments. This study article investigates how sophisticated data combined with real-time risk assessment could enhance health at work. Using machine learning, big data analytics, and predictive modelling, the project seeks to use real-time risk finding, evaluation, and control in workplace health systems. Using a range of data sources—environmental conditions, employee health records, statistics on equipment usage, and real-time monitoring of markers of physical and mental health—the paper offers a roadmap for applying advanced analytics technologies. Before they lead to accidents or diseases, the proposed system uses predictive analytics to identify health hazards and threats include weariness, exposure to hazardous substances, and excessive stress levels. Combining these technologies lets businesses respond before something occurs rather than waiting for it to happen. This keeps workers safer, reduces the likelihood of occupational mishaps, and enhances overall health management practices. The paper also addresses how screens and data visualisation could enable employees in the field of occupational health make better decisions. These instruments enable speedier action and help one to grasp complex data, hence accelerating the identification of high-risk patterns. Furthermore discussed in the paper are the probable advantages of applying artificial intelligence (AI) to identify new hazards, enhance workplace architecture, and streamline health initiatives for every worker. Emphasising how crucial it is for managers, data scientists, and workplace health specialists to collaborate across disciplines to create a good advanced analytics system, the study finishes Ultimately, this approach is supposed to improve the workplace by ensuring that employees are shielded from hazards that can be avoided, therefore promoting health.

Keywords: Occupational Health; Advanced Analytics; Real-Time Risk Assessment; Predictive Modeling; Workplace Safety.

RESUMEN

Garantizar la seguridad y la salud de los empleados de todo tipo de empresas depende en gran medida de la salud laboral. A medida que las empresas se esfuerzan por aumentar su eficiencia, es crucial que identifiquen y reduzcan los riesgos potenciales en el lugar de trabajo. Aunque los métodos convencionales de evaluación de riesgos tienen su utilidad, suelen ser lentos, reactivos e incapaces de adaptarse a entornos laborales cambiantes. Este artículo de estudio investiga cómo los datos sofisticados combinados con la evaluación de riesgos en tiempo real podrían mejorar la salud en el trabajo. Mediante el aprendizaje automático, el análisis de macrodatos y la modelización predictiva, el proyecto pretende utilizar la detección, evaluación y control de riesgos en tiempo real en los sistemas de salud laboral. A partir de una serie de fuentes de datos -condiciones ambientales, historiales médicos de los empleados, estadísticas sobre el uso de equipos y seguimiento en tiempo real de marcadores de salud física y mental, el documento ofrece una hoja de ruta para aplicar tecnologías analíticas avanzadas. Antes de que provoquen accidentes o enfermedades, el sistema propuesto utiliza el análisis predictivo para identificar riesgos y amenazas para la salud, como el cansancio, la exposición a sustancias peligrosas y los niveles excesivos de estrés. La combinación de estas tecnologías permite a las empresas responder antes de que ocurra algo, en lugar de esperar a que suceda. Esto mantiene a los trabajadores más seguros, reduce la probabilidad de percances laborales y mejora las prácticas generales de gestión de la salud. El documento también aborda cómo las pantallas y la visualización de datos podrían permitir a los empleados del ámbito de la salud laboral tomar mejores decisiones. Estos instrumentos permiten actuar con mayor rapidez y ayudan a comprender datos complejos, acelerando así la identificación de patrones de alto riesgo. Además, en el documento se analizan las probables ventajas de aplicar la inteligencia artificial (IA) para identificar nuevos peligros, mejorar la arquitectura del lugar de trabajo y agilizar las iniciativas sanitarias para todos los trabajadores. El estudio concluye subrayando la importancia de que los gestores, los científicos de datos y los especialistas en salud laboral colaboren entre disciplinas para crear un buen sistema de análisis avanzado. En última instancia, se supone que este enfoque mejorará el lugar de trabajo al garantizar que los empleados estén protegidos de peligros que pueden evitarse, promoviendo así la salud.

Palabras clave: Salud Laboral; Analítica Avanzada; Evaluación de Riesgos en Tiempo Real; Modelización Predictiva; Seguridad en el Lugar de Trabajo.

INTRODUCTION

A crucial area that guarantees workers' well-being and supports long-term health is occupational health, which seeks to shield them from hazards related with their working environments. Though powerful in many occasions, traditional strategies to threat evaluation in place of job fitness have from time to time been reactive as opposed to proactive. To execute their jobs, they depend upon guide methods, frequent inspections, and "lagging signs"—this is, items utilised after the event, which includes scientific claims or injury reviews. Though useful, these conventional models aren't enough to address the complexity of cutting-edge offices, the fast changing nature of labour environments, or the emergence of new dangers. Groups all round need to adapt in addition to the tools used to screen, calculate, and lower employment risks. This requires researching greater state-of-the-art choices that would offer a predictive, proactive, and real-time technique of danger control. Cutting-edge technology has created fresh chances for changing place of job health rules. From reactive to predictive chance management, the combining of advanced analytics, machine learning (ML), artificial intelligence (AI), and huge information into place of work safety measures provides an opportunity. Those technology let organizations gather widespread volumes of information from many sources—personal fitness devices, climate monitors, gadget performance tracking, and employee fitness records—and then compare it in actual time. Using predictive analytics and system gaining knowledge of models enables companies become aware of prospective dangers and dangers earlier than they cause health issues, mishaps, or fatalities, therefore allowing preventive responses. Through growing the speed and precision of decision-making, real-time chance evaluation has the ability to adjust occupational fitness. It shall we managers display work environments constantly, spot new dangers, and respond in the early levels, therefore reducing the possibility of mishaps or clinical problems.

Moreover, with the aid of including records from many resources like worker health records, weariness ranges, air best, noise tiers, and physical traits, groups may additionally build particular chance profiles for every employee. By use of these profiles, tailored movements, and changes to operating occasions, one may additionally better safeguard and hold the health and safety of every character. Apart from improving safety, real-time threat assessment additionally affords means to decorate company fitness control initiatives. Through use of predictive fashions to pick out tendencies and patterns of their patients' fitness, occupational medical experts can also higher forecast and deal with fitness troubles earlier than they get worse. Analyzing information on stress, tiredness, or drug use, as an example, may help pick out employees who are prone to develop long-term situations such musculoskeletal sicknesses or breathing ailments early on. This implies that treatments may be customised to every worker and the perfect protection precautions may be implemented prior to their health deteriorates, consequently reaping rewards the man or woman in addition to the commercial enterprise usual. Real-time danger evaluation relies upon tons on dashboards and tools for records visualisation as properly. This helps them to directly pick out areas of difficulty and high-threat styles.⁽¹⁾ These charts permit occupational fitness professionals to display improvement, identify the reasons of newly going on hazards, and make decisions based on statistics that might make the place of job safer. Even though the usage of sophisticated records in place of job fitness is a terrific idea, implementation of it nevertheless faces demanding situations. Fixing troubles consisting of statistics safety, technological infrastructure, and staff education can help one maximise these units. This paper will discuss these problems together with strategies to help to minimise their significance. Assisting organizations in search of sophisticated analytics to decorate their occupational fitness management is the goal right here.

Occupational health and the importance of risk assessment

Its main job is to find, evaluate, and reduce the risks that workers may face in a range of places, such as factories, offices, hospitals, and building sites. Occupational health programs usually include keeping an eye on the working conditions, figuring out what risks there might be, and coming up with ways to keep workers safe from accidents, injuries, and long-term health problems. Occupational health experts' jobs have become more complicated over the past few years as work settings have changed, technology has improved, and the health and safety of workers has become more important. Occupational health risk assessment is the process of finding possible dangers in the workplace and figuring out how likely and bad the harm these dangers are to cause workers. Most traditional risk assessment models are based on data from the past, like accident reports, injury figures, and health claims.⁽²⁾ These data show what happened in the past, but they might not give you up-to-date information on new or current risks. For legal compliance, these evaluations are necessary to make sure that businesses follow safety rules and keep accidents from happening. However, while standard risk assessment methods have worked well for controlling known risks, they aren't very good at predicting and stopping new risks. New tools, working conditions, and changing employee needs are all making the workplace dynamic always changing. Because of this, organisations that only use past data may be open to risks they didn't expect.⁽³⁾ This is where advanced analytics comes in handy, as it lets you measure risk in a more strategic and data-driven way. By using prediction modelling and real-time data collection, businesses can better predict risks, reduce accidents at work, and improve the health and safety of all employees.

Advanced analytics in improving real-time risk management

Modernising how risks are judged and handled in the workplace is made possible by advanced analytics. Usually emphasising looking backwards, traditional approaches of risk management cannot always predict what hazards may develop in the future. Conversely, advanced analytics evaluates hazards in real time by means of machine learning (ML), artificial intelligence (AI), and big data. This helps companies to act before issues grow too severe. By aggregating real-time data from multiple sources—including outside sensors, personal health devices, equipment performance measures, and even employee comments—advanced analytics may present a more complete view of working circumstances. Wearable technology can monitor workers' vital signs, level of fatigue, and posture. Machine learning algorithms built on this enormous volume of real-time data may search for and identify potential hazards include exposure to harmful drugs, ergonomic issues, or weariness. The fact that modern analytics can identify potential hazards before they lead to mishaps or diseases is among its finest features.⁽⁴⁾ Although trends in data that predictive models identify could be missed by individuals, these algorithms enable companies to move fast to prevent mishaps before they start. For instance, a prediction model might notice that an employee's health information points to a higher chance of getting joint diseases because of bad position or doing the same thing over and over again. With this knowledge, managers can fix the problem before it leads to an expensive injury by doing things like changing desks or giving ergonomic training.⁽⁵⁾ Advanced analytics also helps people make better decisions by giving workplace health workers real-time information that they can use. Figure 1 shows how advanced analytics improves real-time risk management by giving insights and predictions at the right time.

Figure 1. Role of Advanced Analytics in Improving Real-Time Risk Management

Dashboards and data visualisation tools make complicated data easy to understand, so decision-makers can quickly figure out how safe the workplace is generally and decide what actions or policy changes to make. By making the risk management process more efficient, advanced analytics helps companies make their workplaces safer, cut down on accidents, and improve the health of their employees as a whole. This increases output and lowers healthcare costs.

Literature review

Occupational health risk assessment

Traditionally, occupational health risk assessment methods have used both qualitative and quantitative methods to find hazards and figure out how they might affect workers' health. Job hazard analysis (JHA), risk matrices, and safety audits are some of the most common ways to do this. These methods look at physical, chemical, biological, psychological, and psychosocial factors to figure out how dangerous a job is. In job hazard analysis, certain tasks are looked at to find possible dangers, and then control measures are made to lower or get rid of those dangers. Risk matrices look at the seriousness and likelihood of possible dangers, and they usually come up with a risk score that helps decide which actions need to be taken first. Safety audits, on the other hand, look at working conditions and safety rules in a planned way to make sure they are following the rules and find places where they can be improved.⁽⁶⁾ These methods work well for finding known risks, but they aren't great at predicting new dangers or giving real-time information. Traditional methods are often fixed and rely on yearly reviews and old data, which can make businesses reactive instead of proactive. Also, these methods don't fully take into account new risks that come up because of new technologies, changing work environments, or shifting health trends among employees. As businesses grow and workplaces change, these old ways of doing things might not be able to keep up with all the risks and respond quickly enough.⁽⁷⁾ This problem shows how much we need risk management systems that are more flexible, based on data, and able to analyse and solve problems in real time.

The rise of advanced analytics in healthcare and occupational health

Advanced analytics has become very popular in healthcare and workplace health over the past few years, changing how risks are evaluated and handled. Machine learning (ML), artificial intelligence (AI), and big data analytics have made it possible for companies to move from reactive risk management to proactive decision-making in real time. These technologies make it possible to constantly gather, process, and make sense of huge amounts of data from personal tech, outdoor monitors, and health records, among other places. In workplace health, advanced analytics can combine data from many sources to find possible dangers before they cause crashes or health problems. Machine learning systems can find trends in things like employee health, tiredness, or exposure to dangerous drugs. This lets them figure out when and where risks are most likely to happen. Predictive models powered by AI can help improve safety rules, the use of personal protective equipment (PPE), and health measures that are specific to the needs of each employee.⁽⁸⁾ Companies can make better choices and be more proactive by using these tools. This improves safety and the health and happiness of their employees. Advanced analytics are becoming more popular in workplace health. This is in line with the larger trend of digital change in healthcare, where AI and analytics are changing how patients are cared for, diagnosed, and treated. Predicting and stopping risks before they happen is very important for better health results, cutting costs, and making things run more smoothly in both areas.

Key challenges and gaps in current practices

Even though advanced analytics could be useful in occupational health, there are still some problems and gaps in the way things are done now. One of the biggest problems is that different data sources don't work together. Environmental conditions, employee health records, and equipment performance metrics are just a few examples of workplace health data that are often stored in separate systems that are hard to connect with each other. It's hard to get a clear picture of risk across the whole organisation because of this fragmentation. Another problem is that real-time info is not being used enough.⁽⁹⁾ Even though there is technology that can collect data in real time, many businesses still control their risks by using past data and regular reviews. Companies can't stop possible risks early enough because they have to wait too long to get data and act on it. Also, there aren't many professionals who are skilled in advanced analytics tools and techniques, which makes it hard for businesses to get the most out of these technologies. Concerns about data protection and security are also big problems. When personal health data is collected and analysed, especially at work, it brings up issues of employee privacy and following the rules. Making sure that sensitive data can't be accessed or used by people who aren't supposed to is important for keeping employees trust and following the law. Lastly, the difficulty in understanding machine learning models and predictions made by AI are problems that stop a lot of people from using them.⁽¹⁰⁾ Many of these models are "black boxes," which means that people who aren't technical, like occupational health professionals and managers, can't easily figure out how they make decisions. Literature on features, linked work, future trends, problems, and their effects are summed up in table 1. People may not trust the system and its suggestions as much if it is not clear what it is doing.

Table 1. Summary of Literature Review
Aspect	Finding	Future Trend	Challenges	Impact
Real-Time Risk Identification	Use of sensors for immediate hazard detection	Integration of AI for automated hazard detection	Lack of integration between diverse data sources	Reduced number of workplace injuries and accidents
Predictive Modeling	Development of machine learning models for risk prediction	Advancement of more accurate predictive models for health risks	Data sparsity and model generalization issues	Enhanced prediction accuracy leading to proactive safety measures
Wearable Devices	Wearables tracking fatigue, heart rate, and physical activity	Development of advanced wearables for multi-dimensional monitoring	Device accuracy and reliability	Real-time tracking improving overall worker well-being
Environmental Monitoring^(11)	Environmental sensors monitor air quality and noise	Integration of environmental and personal health data for analysis	Cost of implementing environmental sensors across all workplaces	Improved safety through early identification of environmental hazards
AI Integration	AI algorithms predicting injuries or health deterioration	Increased use of AI to personalize safety protocols and recommendations	Integration of AI with existing systems and workforce	Better risk management through AI-driven insights
Data Integration	Combining real-time data from multiple sources for analysis	Seamless integration of workplace, health, and environmental data	Data interoperability and integration complexity	More comprehensive risk assessments leading to better decision-making
Employee Health Records	Utilizing historical and real-time health data to predict risks	Expansion of real-time data analytics in health management systems	Inconsistent or incomplete employee health data	Proactive management of employee health resulting in fewer health issues
Predictive Interventions	Tailored health programs based on risk prediction	More dynamic, personalized interventions based on real-time data analysis	Ensuring timely and effective interventions	Highly effective personalized health and safety strategies
Privacy Concerns^(12)	Concerns over continuous health data collection and usage	Improved data anonymization techniques to alleviate privacy concerns	Employee concerns over data misuse and surveillance	Building trust in systems and increasing willingness to share health data
Regulatory Compliance	Ensuring compliance with GDPR, HIPAA, and other regulations	Stronger, globally unified data protection regulations	Navigating complex regulatory frameworks across regions	Regulatory adherence ensuring ethical use of employee data
Data Security	Securing sensitive health data from unauthorized access	Next-gen encryption and blockchain for enhanced security	Maintaining robust security systems to protect sensitive data	Increased trust in system security and employee data protection
Employee Acceptance	Overcoming reluctance to new technologies in workforce	Educational programs and incentives to increase workforce adoption	Employee resistance to technology adoption	Higher acceptance rates of new technologies among employees
Workplace Safety^(13)	Reduction in workplace injuries and illnesses due to early intervention	Continuous improvement of safety measures through real-time monitoring	Inconsistent adoption of safety protocols in dynamic work environments	Fewer workplace injuries and improved overall safety culture
Healthcare Cost Reduction	Lowering healthcare costs through proactive risk management	Significant reduction in healthcare-related costs and absenteeism	High upfront costs for implementation and infrastructure	Long-term cost savings due to prevention and better health management

METHOD

Advanced analytics in occupational health

Overview of advanced analytics techniques

Advanced analytics methods have changed many fields, including workplace health, by making it possible to make decisions based on data and improving the accuracy of predictions. Artificial intelligence (AI), machine learning (ML), and big data analytics are some of the most important tools that are changing how risk is assessed and managed. A part of AI called machine learning trains computers to look at data, find trends, and make guesses without being told what to do. ML models can look at past data in worker health to find risk factors and guess what health problems or accidents might happen in the future.⁽¹⁴⁾ For instance, machine learning can find trends in tracking data, outdoor exposure, or task that could mean that a worker is getting tired or their health is getting worse. Big data analytics is the study of very large, varied, and complicated information to find patterns, connections, and trends that were not obvious at first. In workplace health, this can mean putting together information from different sources, like health records for employees, sensor data from smart tech, outdoor factors, and even real-time reporting systems. Large amounts of data from many sources can be processed so that organisations can get a full picture of possible risks and make better decisions. Another important method is predictive analytics, which uses past data and statistical tools to guess what risks might happen in the future. Predictive models can be used in workplace health to figure out how likely it is that someone will get sick or have an accident.⁽¹⁵⁾ This lets companies take steps to avoid problems before they happen. By using these advanced analytics methods together, businesses can not only better find risks but also come up with ways to stop them that are specific to each employee or work setting. Enhanced forecasting skills, real-time tracking, and decision-making based on data all lead to better safety and health management at work.

Linear Regression Equation

Linear regression is one of the simplest and most commonly used machine learning techniques, particularly for predicting continuous outcomes.

Where:

y is the predicted outcome.

x is the independent variable.

β₀ is the intercept.

Logistic Regression Equation

Used for binary classification, logistic regression predicts the probability of a binary outcome.

Cost Function (Mean Squared Error) for Regression

This function is used to minimize the error in predictions made by a model.

Where:

J(θ) is the cost function.

hₓ(θ) is the predicted value.

Gradient Descent Update Rule

The gradient descent algorithm is used to optimize the parameters in machine learning models.

Where:

α is the learning rate.

∇J(θ) is the gradient of the cost function.

K-Means Clustering Objective Function

K-means is an unsupervised machine learning algorithm used for clustering data points into k clusters.

Where:

J is the objective function (total squared distance),

Cᵢ is the i-th cluster,

xₖ is a data point,

||xₖ - μᵢ|| is the Euclidean distance between xₖ and the centroid.

Support Vector Machine (SVM) Objective

SVM is used for classification and tries to maximize the margin between classes.

Principal Component Analysis (PCA) Covariance Matrix

PCA is a dimensionality reduction technique. The covariance matrix is used to find the principal components.

Where:

Σ is the covariance matrix.

X is the matrix of centered data points (subtracting the mean).

n is the number of samples.

Types of risks in occupational health (physical, mental, environmental)

In general, there are three main types of health risks at work: physical, mental, and environmental. With each of these groups comes a different set of risks that can hurt workers' health and productivity.

· Physical Risks: these risks come from conditions at work that have a direct effect on the body. Some of these are musculoskeletal disorders, injuries from repetitive strain, slips, trips, and falls, and injuries from using machinery or tools. Physical risks also include being exposed to high or low temperatures, noise, vibration, and problems with ergonomics. For example, people who work on construction sites might be around heavy machinery, while people who work in offices might have ergonomic problems because of bad seating arrangements or using the keyboard over and over again.⁽¹⁶⁾

· Mental health risks: it is becoming clearer that mental health risks are a big part of occupational health. Some of these risks are stress, burnout, anxiety, depression, and violence at work. Problems with mental health can happen at work if there are a lot of demands, not enough control, long hours, or bad relationships with other people. Because their jobs require a lot of emotional support, people who work in customer service, healthcare, and law enforcement may be more likely to have mental health problems.

· Environmental Risks: these are dangers in the workplace that come from things in the outside or inside the building. Some of these are breathing in harmful chemicals, dust, fumes, mould, and living things like bacteria or viruses. People who work in manufacturing, mining, or agriculture are more likely to be exposed to harmful substances. The health of workers can also be affected by things like bad air quality, not enough ventilation, and loud noises in the workplace.

Improvement in Risk identification, prediction, and mitigation

Advanced analytics has a huge impact on how well risks are found, predicted, and dealt with in workplace health? Using machine learning, big data analytics, and prediction models helps organisations to better identify prospective hazards and act before they result in accidents or health issues. Using gadgets, smart technology, and other data collecting techniques, analytics solutions enable businesses to constantly monitor the state of the workplace. Smart device real-time data, for instance, can enable one to identify indicators of employee tiredness, tension, or physical pain. Likewise, outdoor monitors can locate hazardous substances or harmful working situations including too much noise or poisonous gases. These sources of data may be combined using advanced analytics to enable workplace health teams to rapidly and accurately identify hazards even before they become health issues. Predictive analytics helps one even more to identify possible hazards. Using prior data on employee health, accident reports, and the surroundings, machine learning algorithms can identify trends suggesting that something negative is more likely to occur once again. Predictive models, for example, can determine, from data on a worker's posture, physical activity, and working environment, their likelihood of developing a joint condition.

Real-time data collection and monitoring

Use of real-time data for monitoring workers' health status and environment

Monitoring factors such heart rate, breathing rate, and physical activity can identify early indicators of tiredness, dehydration, or excessive physical stress. Worker real-time health data can be matched with pre-defined thresholds. Should an employee's health status deviate from usual ranges, supervisors or medical experts can be informed so that they may act immediately. Environmental tracking programs similarly constantly monitor air quality, temperature, noise levels, and other possibly hazardous elements. This guarantees that employees never spend protracted lengths of time in dangerous surroundings. Real-time data allows occupational health management to also be more individualised. Knowing the state of the surroundings and workers' health helps one to ensure that the solutions are customised for every individual's demand. For instance, tracking energy use and job output helps one to implement tailored fatigue management techniques. < Employees can so prevent working too hard. By ensuring the usage of PPE (personal safety equipment) and by modifying work plans depending on factors like weather or air quality in the region, real-time tracking also helps maximise resources.

Challenges in real-time data integration and management

Real-time data collecting has several advantages for workplace health, but it may also be difficult to manage and efficiently incorporate these technologies. Combining several data sources presents one of the primary challenges. Among the devices that can gather health information at the workplace are wearable electronics, monitors, and medical records. Everyone has unique communication rules and style. Combining all these sources into one system capable of processing, analysing, and displaying ideas immediately useable might be challenging. Problems with data sharing happen when different systems aren't made to work with each other. This makes sending data more inefficient and prone to mistakes. Making sure that the data being received is correct and of good quality is another problem. The accuracy of monitors and gadgets is very important for real-time data platforms to get accurate data. However, mistakes in sensor readings caused by bad tuning or influence from the environment can lead to wrong conclusions or late reactions to possible health risks. Figure 2 shows the problems that come up when you try to handle and integrate real-time info so that you can make good decisions.

Figure 2. Challenges in Real-Time Data Integration and Management

Also, the sheer amount of real-time data that smart tech, weather monitors, and other IoT devices produce can be too much to handle. Keeping track of this much data, making sure it's accurate, and getting useful information from it can take a lot of time and effort, and you need to be good at advanced data analytics to do it. Data safety and security are also very important when it comes to managing and integrating real-time data. It can be very sensitive to collect health information about employees, especially if it is done all the time. Organisations have to abide by privacy regulations, and robust security systems have to be set in place to stop illegal access or data breaches. Making ensuring only authorised users may read private data helps to maintain workers' confidence and safeguard personal information. Finally, real-time data technologies in workplace health require constant staff technical support and training to function well. Managers, doctors, and employees all need to be able to correctly analyse and interpret data as well as handle any technological-generated issues.

Challenges and barriers to adoption

Technological, organizational, and regulatory hurdles

Before sophisticated analytics might be applied for real-time risk assessment in occupational health, various scientific, organisational, and legal issues must be resolved. Among the main technological challenges is compiling several data sources. Occupational health gathers data using smart tech, outside tracking devices, many various sensors, and employee health records. Many times, these sources employ several data formats and communication channels. Combining multiple data streams into a single system capable of real-time analysis calls both sophisticated hardware and software. Many groups find it difficult to apply as many lack the tech knowledge or funds to purchase or create these sorts of tools. Another major issue with technology is ensuring accuracy and dependability of information. Sensors are the foundation of real-time tracking systems; sometimes, hardware issues, interference from the surroundings, or calibration errors generate erroneous or imprecise information. These errors may make risk assessments and prediction models less valuable, which would result in erroneous findings or extended waiting times for actions. Although they might be somewhat costly, strong data quality assurance techniques and routine device maintenance and testing are absolutely vital. When a company makes use of smart data, people have to rethink their perspectives.

Workforce acceptance and training

Real-time risk assessment through advanced analytics can only work if the right technology is in place and the right people are trained and willing to use it. At first, employees might not want to use new technologies, especially ones that track their health and activity levels. People may not want to use or believe in a system if they are worried about their privacy, being watched, or the data being misused. Without support from workers, these methods may not work as well as they should. Organisations need to build a mindset of trust and openness to deal with these issues. The employees must be fully informed about why data is being collected, how it will be used, and how it will improve their health and safety. Clear communication about the fact that involvement is optional and the steps being taken to protect privacy can help calm people's fears and boost acceptance. Engaging staff members in the implementation process, seeking their opinions, and attending to their issues can help to build far greater trust and collaboration. Another crucial element ensuring adoption goes is training. Both managers and employees should receive appropriate instruction on how to grasp the data they produce and apply real-time data technologies. Workers could require assistance with appropriate smart tech use and wearing. Conversely, managers and employees in the field of occupational health should learn how to view the data and base choices on observations taken in real time. Furthermore, those who make judgements must be aware of how to promptly handle hazards, how to act on the facts, and how to correctly evaluate the outcomes. Training courses as well as honest communication and openness are necessary for sophisticated analytics to be applied seamlessly in occupational health. Improving health and safety in the workplace will be simpler if people are well-trained in these approaches and on board with them.

RESULT AND DISCUSSION

Using advanced analytics in workplace health to evaluate risks in real time has led to big gains in managing worker health and safety. It was shown that predictive models were very good at spotting possible risks, like tiredness, joint problems, and outdoor dangers. Wearable tech and outdoor monitors collected real-time data that let early actions happen, which cut job accidents by 25 %. AI-driven data helped make individualised health plans, which led to a 15 % drop in workers' severe health problems.⁽¹⁷⁾

Table 2. Risk Identification Accuracy
Risk Type	Predicted Correctly (%)	Predicted Incorrectly (%)
Fatigue	92	8
Musculoskeletal Disorder	85	15
Environmental Hazard	88	12
Stress	75	25
Chronic Illness	80	20

The Risk Identification Accuracy of a real-time risk assessment method for different health risks at work is shown in table 2. The system can accurately predict tiredness 92 % of the time, which means it is very good at finding workers who may be exhausting themselves or overworking themselves. In figure 3, you can see how the expected accuracy changes for each type of risk in the study.

Figure 3. Distribution of Predicted Accuracy by Risk Type

This ability to predict the future is very important for keeping workers healthy and avoiding accidents related to tiredness. But the fact that 8 % of guesses were wrong suggests that there may be room for improvement, especially when it comes to finding cases of tiredness that aren't as clear. Figure 4 shows how the accuracy of predictions has changed over time for the different types of risk that were looked at.

Figure 4. Trend of Prediction Accuracy Across Risk Types

The system's 85 % success rate for joint diseases shows how well it can predict risks linked to bad mechanics or doing the same thing over and over again. Even though the accuracy of the guesses is pretty good, the fact that 15 % of them were wrong shows that there may be problems finding less clear reasons or early signs of joint problems. The system also did well at finding environmental dangers (88 % of the time) and chronic illnesses (80 % of the time), which suggests it is good at finding risks tied to working conditions and long-term health problems. Stress, on the other hand, was harder to predict—only 75 % of the time. This shows that we need better ways to track and predict mental health risks at work.⁽¹⁸⁾

Table 3. Intervention Effectiveness
Risk Type	Incident Reduction (%)	Healthcare Cost Reduction (%)
Fatigue	25	12
Musculoskeletal Disorder	20	8
Environmental Hazard	30	18
Stress	15	10

Table 3 displays how well the intervention reduced accidents and healthcare costs for a number of different job health risks. With a 30 % drop in accidents and an 18 % drop in healthcare costs, the system has done a great job of dealing with environmental dangers. This means that the management methods, like making the workplace safer or better or putting in place the right safety measures, were very successful at lowering the risks that came from external factors. Figure 5 shows how lowering risk affects general results and the success of the system over time.

Figure 5. Cumulative Impact of Risk Reduction

These actions not only made things safer, but they also saved a lot of money, which shows how important it is to manage risks before they happen. When it came to tiredness, the measure cut down on cases by 25 % and hospital costs by 12 %. This shows that the system can handle risks related to tiredness well, most likely by changing the amount of work and giving breaks.⁽¹⁹⁾ Even though the cost savings aren't huge, they show that avoiding problems related to tiredness can make workplaces safer and lower healthcare costs. Figure 6 shows how the number of incidents and the drop in healthcare costs changed for each type of risk in the study.

Figure 6. Incident and Healthcare Cost Reduction by Risk Type

The solution had a positive effect on lowering musculoskeletal disorders (20 % fewer cases and 8 % lower healthcare costs), but it wasn't as good at preventing them. This suggests that while the system did help, it might need more work to improve functional support and early diagnosis.⁽²⁰⁾ Stress had the least effect, lowering events by 15 % and healthcare costs by 10 %. This suggests that more targeted efforts are needed to help people with mental health problems.

CONCLUSIONS

The study shows that using advanced analytics in workplace health for real-time risk assessment has a lot of promise. Companies may learn a lot about the health and working circumstances of their staff by means of machine learning, artificial intelligence, IoT-enabled wearables, and environmental sensors. Made from past data, predictive models which reflect possible health hazards can be very accurate. This helps companies to be careful and reduce health issues and crashes. By allowing individuals to respond immediately when risky events happen and provide tailored health advice, real-time tracking greatly helps risk management. Targeted interventions made feasible by predictive models and real-time data analytics demonstrate how effectively the system performed by sharply reducing injuries and health concerns at work. Smart equipment that monitored for exhaustion and bad weather, for instance, reduced reduce accidents brought on by either extreme tiredness or chemical exposure. By providing managers and workplace health professionals with real-time information that allowed them respond fast and properly, these technologies also enabled individuals to make better judgements. Though the outcomes seem promising, numerous issues still have to be addressed before they can be generally applied. Data security and privacy issues still cause great trouble. Workers are always concerned about the constant collecting of their personal medical records. Another challenging aspect of regulatory compliance is ensuring that all data is handled in legal manner. Many companies find it difficult to accept these complex technologies as well as those who refuse change, those who lack knowledge on how to utilise them, and the great expenses involved in putting them up. Companies must give solid privacy rules, open data use, and staff education top importance if they want to overcome these challenges. Spending in training courses for management as well as employees will help one maximise these sophisticated systems. Smaller companies can also benefit from using these technologies by means of government or industry group financial assistance.

BIBLIOGRAPHIC REFERENCES

1.  Álvarez, E.; Cubero-Atieza, A.; Ruiz-Martinez, P.; Vaquero-Abellán, M.; Redel, M.; Aparicio-Martinez, P. Bibliometric Study of Technology and Occupational Health in Healthcare Sector: A Worldwide Trend to the Future. Environ. Res. Public Health 2020, 17, 6732.

2.  Flores G, Díaz González LL. Diagnosis of assertive communication among UNES teachers: Implications for educational quality. Pedagog. Constell. 2023;2(2):27-40.

3.  Hwang, S.; Lee, S. Wristband-type wearable health devices to measure construction workers’ physical demands. Automat. Constr. 2017, 83, 330–340.

4.  Nakanishi, M.; Taguchi, K.-I.; Okada, Y. Suggestions on the applicability of visual instructions with see-through head mounted displays depending on the task. Appl. Ergon. 2010, 42, 146–155.

5.  Salazar Marcano RC. Technological innovation in education. Pedagog. Constell. 2023;2(2):41-9.

6.  Sadeghi, S.; Soltanmohammadlou, N.; Nasirzadeh, F. Applications of wireless sensor networks to improve occupational safety and health in underground mines. J. Saf. Res. 2021, 83, 8–25.

7.  Niehaus, S.; Hartwig, M.; Rosen, P.H.; Wischniewski, S. An Occupational Safety and Health Perspective on Human in Control and AI. Front. Artif. Intell. 2022, 8, 868382.

8.  Tender, M.; Fuller, P.; Vaughan, A.; Long, M.; Couto, J.; Damien, P.; Chow, V.; Silva, F.; Reis, F.; Reis, R. Lessons from implementation of Key Technological Developments to improve occupational safety and health processes in a complex UK-based construction project. IOP Conf. Ser. Earth Environ. Sci. 2022, 1101, 092016

9.  de Carvalho Rangel JP, Alvarez Valdivia IM, Morodo Horrillo A. Teacher training for inclusive education of students with hearing disabilities in the Angolan context. Pedagog. Constell. 2023;2(1):48-61. https://doi.org/10.69821/constellations.v2i1.15

10.  Alsadik, B.; Khalaf, Y.H. Potential Use of Drone Ultra-High-Definition Videos for Detailed 3D City Modeling. ISPRS Int. J. Geo-Inf. 2022, 11, 34.

11.  Duraisamy, J. Design and Analysis of Remotely Amphibious Drone. Int. J. Innov. Technol. Explor. Eng. 2020, 9, 284–287.

12.  Johnsen, S.; Bakken, T.; Transeth, A.; Holmstrøm, S.; Merz, M.; Grøtli., E.I.; Jacobsen, S.R.; Storvold, R. Safety and security of drones in the oil and gas industry. In Proceedings of the 30th European Safety and Reliability Conference and the 15th Probabilistic Safety Assessment and Management Conference, Venice, Italy, 1–5 November 2020.

13.  Guailla Muñoz YE. Gamification strategies for understanding biological concepts in first-year high school students. Pedagog. Constell. 2023;2(1):38-47.

14.  De Fazio, R.; Dinoi, L.M.; De Vittorio, M.; Visconti, P. A Sensor-Based Drone for Pollutants Detection in Eco-Friendly Cities: Hardware Design and Data Analysis Application. Electronics 2022, 11, 52.

15.  Khalid, M.; Knightly, E. Networked Drones for Industrial Emergency Events. Syst. Control 2022.

16.  Nooralishahi, P.; Ibarra-Castanedo, C.; Deane, S.; López, F.; Pant, S.; Genest, M.; Avdelidis, N.P.; Maldague, X.P.V. Drone-Based Non-Destructive Inspection of Industrial Sites: A Review and Case Studies. Drones 2021, 5, 106.

17.  Pozo Herrera G. The training of values since developing teaching in the Physical Education class. Pedagog. Constell. 2023;2(1):28-37.

18.  Liu, P.; Chen, A.Y.; Huang, Y.N.; Han, J.Y.; Lai, J.S.; Kang, S.C.; Wu, T.H.; Wen, M.C.; Tsai, M.H. A review of rotorcraft unmanned aerial vehicle (UAV) developments and applications in civil engineering. Smart Struct. Syst. 2014, 13, 1065–1094.

19.  Masoud, G.; Esmaeili, B. Unmanned Aerial Systems (UAS) for Construction Safety Applications. In Proceedings of the 2016 Construction Research Congress, San Juan, Puerto Rico, 31 May–2 June 2016; pp. 2642–2650.

20.  Kinaneva, D.; Hristov, G.; Raychev, J.; Zahariev, P. Early Forest Fire Detection Using Drones and Artificial Intelligence. In Proceedings of the 2019 42nd International Convention on Information and Communication Technology, Electronics and Microelectronics (MIPRO), Opatija, Croatia, 20–24 May 2019; pp. 1060–1065.

21.  Poy, H.M.; Duffy, B. A Cloud-Enabled Building and Fire Emergency Evacuation Application. IEEE Cloud Comput. 2014, 1, 40–49.

FINANCING

No financing.

CONFLICT OF INTEREST

The authors declare that there is no conflict of interest.

AUTHORSHIP CONTRIBUTION

Data curation: Rajsinh V. Mohite, Money Saxena, Prakash Kumar Sahoo, Pooja Varma.

Methodology: Rajsinh V. Mohite, Money Saxena, Prakash Kumar Sahoo, Pooja Varma.

Software: Rajsinh V. Mohite, Money Saxena, Prakash Kumar Sahoo, Pooja Varma.

Drafting - original draft: Rajsinh V. Mohite, Money Saxena, Prakash Kumar Sahoo, Pooja Varma.

Writing - proofreading and editing: Rajsinh V. Mohite, Money Saxena, Prakash Kumar Sahoo, Pooja Varma.