Industrial Internet of Things for Assembly Line Worker’s Work

Fatigue Recognition

Venkata Krishna Rao Pabolu

1 a

, Divya Shrivastava

and Makarand S. Kulkarni

Department of Mechanical Engineering, Shiv Nadar University, UP 201314, India

Department of Mechanical Engineering, Indian Institute of Technology Bombay, India

Keywords: Internet of Things, Machine Learning, Worker’s Work Fatigue, Assembly Line, Sensors.

Abstract: The fourth industrial revolution or Industry 4.0 is based on the Internet of Things (IoT) and other intelligent

technologies. IoT is mature enough to make seamless real-time communication between data-grasping sensors

and intelligent machines. Recognition and prevention of workers' work fatigue remain challenging for

manufacturing industries. The objective of this research is to develop an IoT-based worker’s work fatigue

recognition system to recognize the real-time fatigue status of assembly line workers. A learning-based

knowledge model is prepared from the historical worker’s work fatigue status to classify the worker’s work

fatigue status (as ‘Yes’ or ‘No’) using the real-time monitoring system. Where a sensor-connected IoT

framework is adopted for monitoring the real-time state of an assembly worker. Finally, an intelligent system

is proposed to recognize the real-time worker’s fatigue status from the IoT real-time monitored data using the

learning-based worker’s work fatigue recognition model. A use-case illustration is given to demonstrate the

research scope for a manual assembly line.

1 INTRODUCTION

Workers' work stress or work fatigue is one of the

leading contributors to assembly line work errors,

which could become a bottleneck to work

productivity or product quality. Assigning a task to an

operator for an extended period can lead to worker

work stress or work fatigue, resulting in an efficiency

loss or disinclination of efforts or Musculoskeletal

disorders (MSDs) (Grandjean, 1979). Work-related

musculoskeletal disorders (WMSD) are caused by the

poor work environment or Ergonomics of the

workplace (Govaerts et al., 2021). Neck, shoulder,

and back-related problems are widely seen MSDs in

manufacturing workers (Yang et al., 2023), causing

34% of annual work time loss and 7% loss of

manufacturing productivity (Nur et al., 2014).

Moreover, the objective of Industry 5.0 is to

transform manufacturing companies from a

technology-centred approach to human-centric,

sustainable, and resilient manufacturing (Abdous et

al., 2023). This research aims to develop an intelligent

https://orcid.org/0000-0002-1480-4822

https://orcid.org/0000-0002-6842-2803

https://orcid.org/0000-0002-1930-5555

worker’s fatigue status recognition system to identify

the assembly line fatigued worker, which supports the

assembly line manager or supervisor while finding

and facilitating the assembly line fatigued workers.

Industry 4.0, or the Fourth Industrial Revolution,

is providing a new direction to the manufacturing

industries by the application of digital technologies.

The Internet of Things (IoT) is a modern and fast-

growing technology that has become a part of the

realization of Industry 4.0. IoT is a real-time

connecting link between the physical and digital

entities (Manavalan & Jayakrishna, 2019). A typical

IoT framework comprises sensors, edge gateways,

and a middleware cloud platform deployed to enable

real-time data. The Industrial Internet of Things

(IIoT) refers to the use of IoT technologies in

manufacturing or factory floor applications. IoT is an

interconnection of many devices through an internet

system capable of monitoring, collecting,

exchanging, analyzing, and delivering data or

information (Al-Turjman & Alturjman, 2018). IoT

can be built with various “smart” sensors for

302

Pabolu, V., Shrivastava, D. and Kulkarni, M.

Industrial Internet of Things for Assembly Line Worker’s Work Fatigue Recognition.

DOI: 10.5220/0012726200003705

Paper published under CC license (CC BY-NC-ND 4.0)

In Proceedings of the 9th International Conference on Internet of Things, Big Data and Security (IoTBDS 2024), pages 302-309

ISBN: 978-989-758-699-6; ISSN: 2184-4976

intelligent monitoring and data extraction. The

monitored data can be processed or mined with data

analytics and machine learning techniques to make

knowledge objects or models, eventually used for

intelligent decision-making.

A learning mechanism makes a learning-based

model, which makes a relation between monitoring

factors and the effect. However, the learning can be

either from sensing, historical experience, or a well-

defined source of knowledge. A classification model

can be developed through a learning process with

historical experiences or data points. Clustering

learning and classification learning are sequential

steps to recognize the patterns. Where the historical

experiences or data points are grouped into

meaningful clusters or classes, the class labels can be

used to build the classification mechanism (Cai et al.,

2009). Historical fatigued worker status can be

monitored. The monitored fatigued status data can be

classified as ‘Yes’ and ‘No’ classes. Furthermore, a

fatigued status classification model can be developed

with the worker fatigue status classified data. The

application of sensors, IoT, and classification

learning to detect the real-time fatigue status of the

assembly workers is the main research objective of

this work.

2 LITERATURE REVIEW

Worker’s work stress and Musculoskeletal disorders

(MSDs) remain a considerable problem to the

manufacturing industry around the world. Workers’

work fatigue impairs work performance in alertness,

emotional stability, and mental and physical ability

(Sawatzky, 2017). (Sundstrup et al., 2020) have

discussed some of the typical MSDs observed in

industrial workers. Back or neck pains are the typical

MSDs observed in industrial workers. Health issues

such as memory or concentration loss, brain fog,

increased risk of stroke, cardiovascular disease,

hypertension, and decreased immunity are seen in

fatigued workers (Sawatzky, 2017). (Sedighi Maman

et al., 2020a) have attempted to recognize physical

fatigue using wearable sensors and machine learning

methods.

(Sadeghniiat-Haghighi & Yazdi, 2015) have

presented a detailed review on workers’ work stress

or fatigue. Furthermore, they elaborated fatigue

measurement techniques at the workplace regarding

physical, mental, and environmental aspects.

(Fardhosseini et al., 2020) used a three-axis

accelerometer to recognize the physical fatigue of

construction workers. (Ma et al., 2009) used

electromyography and heart rate methods to measure

muscular fatigue. (Baghdadi et al., 2019) used

electroencephalography (EEG) to detect the mental

fatigue of workers. (Charbonnier et al., 2016) used

electromyography (EMG) to recognize localized

muscle-related fatigue of workers. (Iskander et al.,

2018) used optical sensors to detect sleepiness.

(Halim et al., 2012) have done fatigue assessments for

prolonged standing workers using surface

electromyography (sEMG).

(Givi et al., 2015) and (Elmaraghy et al., 2008)

discussed worker’s work stress or fatigue-causing

parameters, such as assembly task repetition, task

complexity, task duration, task environment, worker

work skill, workplace ergonomic design, assembly

line speed, worker personality, working teams,

coordination, management strategy towards the work,

work distance, working space, worker motivation

level, length of task cycle, work interruptions, Etc.

(Pabolu & Shrivastava, 2021) given learning-

based fatigue classification function in terms of

worker and workload attributes. (Usuga Cadavid et

al., 2020) discussed learning methods widely using in

the Industry 4.0 era. Association rules, K-Nearest

Neighbors (k-NN), Bayesian networks, Naïve bayes

models, linear regression models, polynomial

regression, Gaussian process regression, Q-Learning,

R- learning, decision trees, bagging, gradient

boosting, random forest, ensemble learning, support

vector machine, etc., are some of the techniques

widely using in the industry 4.0 era. (Dogan & Birant,

2021) gave a detailed review of machine learning

methods useful to resolve manufacturing challenges

during the fourth industrial evolution. (Sedighi

Maman et al., 2020b) have discussed three types of

learning-based models to formulate the worker’s

stress behavior. Those are statistical models,

classifiers, and ensemble models. (Pabolu et al.,

2022) have proposed a prediction system using

machine learning applications to predict the

comfortable work-duration time of an assembly line

worker by considering the worker, work, and work

environment.

The application of intelligent systems in the

predictive maintenance of industrial equipment by

fault diagnosis and prognosis can be seen during the

Industry 4.0 era (Li et al., 2017). (Vogl et al., 2019)

have given a detailed review of diagnostic and

prognostic capabilities and corresponding

possibilities to make best practices in manufacturing.

Furthermore, it is described that prognostics and

health management can be done with intelligent

decision-making using inference knowledge, which is

made with real-time and historical state information

Industrial Internet of Things for Assembly Line Worker’s Work Fatigue Recognition

303

Figure 1: IoT-based intelligent fatigue recognition system.

of subsystems and components. The application of

cyber tools to manage operator health using

diagnostic and prognostic capabilities is the ongoing

research trend during this decade. The potential

requirement or research gap noticed during the

literature study is to display or show the real-time

fatigue status of all final assembly line workers using

diagnostic and prognostic capabilities. The

development of a wearable sensor-connected IoT-

based intelligent work fatigue recognition system to

recognize the real-time fatigue status of the assembly

line workers is the contribution of this research to fill

the research gap partially.

3 IOT-BASED WORKER’S

FATIGUE STATUS

RECOGNITION SYSTEM

An IoT-based intelligent fatigue recognition system

is proposed to recognize workers' fatigue status.

Figure 1 shows the proposed framework to recognize

the assembly line fatigued worker. A set of sensors is

to be placed on the assembly line worker to monitor

the worker’s fatigue status. Furthermore, the IoT

framework transfers the fatigue factor’s sensor data to

the cloud. Then, an intelligent fatigue recognition

system reads the worker’s fatigue factor data and

recognizes the worker’s work fatigue status using a

learning-based fatigue status recognition function.

Details of worker fatigue status recognition function

(i.e., inference rule to invoke knowledge), IoT status

monitoring, and intelligent fatigue status recognition

are discussed in the following part of this section.

Worker Fatigue Status Recognition Function

The worker fatigue status recognition function is a

function to recognize the fatigued status of an

assembly worker from their current status. Equation

1 represents the worker fatigue status recognition

function. The worker fatigue status recognition

function is a function of the worker's age, sex, eye

blink rate, heart rate, and active hand moment. Where

eye blink rate is considered in blinks per minute, heart

rate is considered as heart rate reserve in beats per

minute, and active hand moment is in the form of

acceleration (m/s

Table 1: Description of indexes.

Index Descri

tion

Worker Fatigue Status

a Worker’s a

Worker’s sex

EBR E

e Blink Rate

HRR Heart Rate Reserve

DHR Direct Heart Pulse Rate

HR Resting Heart Pulse Rate

HM Hand Moment

Worker Fatigue Status (WFS) =



𝑎,𝑠,𝐸𝐵𝑅,𝐻𝑅𝑅,𝐻𝑀



(1)

𝐻𝑅𝑅 



𝐷𝐻𝑅  𝑅𝐻𝑅



𝑅𝐻𝑅

∗ 100

(2)

𝐻𝑀 







𝐴









𝐴











𝐴









(3)

Equation 2 calculates heart rate reserve (HRR),

which is a function of the direct heart pulse rate

(DHR) in beats/min., and resting heart pulse rate

IoTBDS 2024 - 9th International Conference on Internet of Things, Big Data and Security

304

Figure 2: IoT based worker fatigue factor monitoring.

(RHR) of the corresponding worker. Equation 3 is to

calculate the active hand moment, which is a vector

product of hand acceleration components (i.e., A

, A

) in m/s

Table 2: Historical Fatigue Response Data.

a s EBR HRR HM

Fatigue

Status

xx x xx xx% x.xx Yes

xx x xx xx% x.xx No

The worker fatigue status recognition function is

a learning model learned from the historical worker

fatigue status data. Table 2 shows the format of

historical worker fatigue status data, will be given to

the learning machine. A range of machine learning

classification models are available in the literature as

Logistic Regression, Polynomial SVM, Decision

Tree, Random Forest, xgBoost, Naïve Bayes, k-

Nearest Neighbour, Linear Discriminant Analysis,

Quadratic Discriminant Analysis, Etc, (Wang, 2019).

A suitable machine learning classification model can

be used to make the worker fatigue status recognition

function for the historical worker fatigue status data

(i.e., Table 2).

The machine learning model (i.e., Worker fatigue

status recognition function) acts as an inference rule,

which will be stored in the knowledge base. Apart

from the inference rule, the knowledge base also

contains the facts about the assembly line workers,

including details of the factory's assembly line

workers list, corresponding age, sex, and

resting heart

pulse rate (i.e., RHR).

IoT Status Monitoring

During the IoT status monitoring, a set of sensors,

such as an eye blink sensor to measure the eye blink

rate of the worker, a pulse sensor to measure direct

heart pulse rate (i.e., DHR), and an IMU sensor to

measure hand acceleration must be placed on the

assembly line worker. Furthermore, the sensors must

be connected to the IoT controller. The IoT controller

sends the worker's current status or sensor’s

monitored data to the cloud or server. The IoT

monitoring framework is shown in Figure 2.

Intelligent Fatigue Recognition

The intelligent fatigue status recognition algorithm

does the worker fatigue status recognition. Figure 3

shows the framework of an intelligent worker’s

fatigue status recognition algorithm. The algorithm

takes the parameters of the worker fatigue status

recognition function (i.e., eq. 1) from the facts of the

knowledge base and IoT-monitored data. Where the

worker’s age, sex, and RHR are from the facts of the

knowledge base, and DHR, EBR, and HR values are

from the IoT-monitored data. However, HRR can be

calculated using Equation 2. An inference engine

Industrial Internet of Things for Assembly Line Worker’s Work Fatigue Recognition

305

takes all parameters of the worker fatigue status

recognition function (i.e., a, s, HRR, EBR, HR) and

finds the worker’s fatigue status using the worker

fatigue status recognition function (i.e., eq. 1).

Figure 3: Intelligent Fatigue Status Recognition Algorithm.

4 USE-CASE ILLUSTRATION

Let the IoT monitored status of an assembly worker

(i.e., worker9) is Figure 4. However, facts of the

respective worker are in Table 3. The objective is to

find the fatigue status of the corresponding worker

(i.e., worker9).

Figure 4: Worker9 IoT monitored status.

Table 3: Facts of worker9.

Worker Name a s RHR

Worker9 36 F or

(

)

Let the historical worker’s fatigue response data

is Table 4. Where ‘0’ is considered for female

workers and ‘1’ is considered for male workers.

Similarly, ‘Y’ represents the worker under fatigued

conditions, and ‘N’ represents not fatigued. The data

set (i.e., Table 4) is prepared data (i.e., not-real)

exclusively prepared for illustration.

Equation 4 shows the linear fit for the historical

worker’s fatigue response data (i.e., Table 4) at 80%

of learning and 20% of the test data set with 10-fold

crass validation. However, Figure 5 shows the

accuracy of all learning models where the linear

regression fit (i.e., eq. 4) is considered as the worker’s

fatigue status recognition function for the use-case

illustrative example.

Table 4: Historical worker’s fatigue response data.

a s EBR HRR HM

Fatigue Status

22 0 23 10 2 N

35 0 32 22 3 Y

44 0 32 43 2 Y

53 0 25 35 2 Y

34 0 25 28 2 N

24 1 26 24 1 N

39 1 26 23 4 N

49 1 24 24 2 N

50 1 31 28 4 Y

39 1 27 31 2 Y

25 1 32 52 1 Y

22 0 23 10 2 N

𝑊𝐹𝑆



 1.282 𝑎 23.659𝑠

 5.722 𝐸𝐵𝑅 1.278 𝐻𝑅𝑅

 1.117 𝐻𝑀233.670

(4)

Figure 6 shows the input to the learning-based

worker’s fatigue status recognition function (i.e., eq.

4) to recognize the worker9’s work-fatigue status,

where HRR is calculated using equation 2.

Table 5: Predicted work-fatigue response for worker9.

a s EBR HRR HM

Fatigue

Status

36 0 34 24 1 Y

Table 5 shows the predicted work-fatigue

response by the worker’s fatigue status recognition

function (i.e., eq. 4) for the worker9 based on the

monitored IoT current status (i.e., Figure 4). The

worker’s fatigue status recognition function is

predicted as worker9 under fatigued condition.

IoTBDS 2024 - 9th International Conference on Internet of Things, Big Data and Security

306

Figure 5: Accuracy of the learning models.

Figure 6: Input to the fatigue status recognition function.

5 DISCUSSION AND WORK

IMPLICATION

An IoT-based intelligent fatigue recognition system

is proposed for finding an assembly line fatigued

worker. The proposed framework is helpful to an

assembly line supervisor or manager while

recognizing an assembly line fatigued worker and

facilitation. Furthermore, the framework improves

the worker’s work life and work productivity. The

discussed use case in section 4 explains the

application of an IoT-based fatigue recognition

system for recognizing the assembly line worker’s

fatigue status.

The preparation of an IoT-based fatigue status

monitoring setup and the development of a worker

fatigue status recognition function are critical for

deploying the proposed methodology for a real

assembly line. However, selecting appropriate

sensors by considering their sensitivity, correctly

placing sensors on the assembly worker, and

maintaining and replacing sensors are critical for the

IoT-based fatigue status monitoring setup.

Preparation and selection of a learning-based fatigue

recognition model and corresponding prediction

accuracy are critical for worker fatigue status

recognition function. Six learning models are

prepared for the use-case illustration with the

historical worker’s fatigue response data (i.e., Table

4). These are the Linear Regression, Polynomial

SVM, Decision Tree, Random Forest, Naïve Bayes,

Figure 7: Worker Fatigue Status Dashboard.

Industrial Internet of Things for Assembly Line Worker’s Work Fatigue Recognition

307

and Linear Discriminant Analysis models. Figure 5

shows their corresponding prediction accuracy. Since

the learning data set is small and the corresponding

prediction accuracy is low.

Making a historical worker’s fatigue response

data preparation system to make worker’s historical

fatigue-status data points (example Table 4), making

an appropriate learning-based worker fatigue status

recognition function, making a dashboard (i.e., Figure

7) to see real-time fatigue status of all assembly line

workers, improving the fatigue status recognition

accuracy and verifying the proposed methodology for

a large assembly line are the prospects to this

research.

6 CONCLUSION

IoT-based intelligent work fatigue status recognition

system framework is presented. The framework

comprises a worker fatigue status recognition

function, IoT-based status monitoring, and intelligent

fatigue status recognition. Learning-based methods

are used to make worker fatigue status recognition

function. Sensor-connected IoT-based worker status

monitoring system to monitor real-time status of the

worker in terms of the worker fatigue status

recognition function’s factors. Finally, the intelligent

system classifies the monitored status as ‘Yes’ or

‘No’ using the developed learning-based worker

fatigue status recognition function. A use-case

illustration is presented to demonstrate the proposed

framework for a manual assembly line. Linear

Regression, Polynomial SVM, Decision Tree,

Random Forest, Naïve Bayes model, and Linear

Discriminant Analysis model are used to make the

worker fatigue status recognition function. The linear

regression model has given better prediction accuracy

compared to others. Making a historical worker’s

fatigue response data preparation system, making

worker fatigue status recognition function with

acceptable accuracy, and making a dashboard to see

the real-time fatigue status of all assembly workers

are the prospects for this work.

REFERENCES

Abdous, M.-A., Delorme, X., Battini, D., & Berger-Douce,

S. (2023). Multi-objective collaborative assembly line

design problem with the optimisation of ergonomics

and economics. International Journal of Production

Research, 61(22), 7830–7845. https://doi.org/10.108

0/00207543.2022.2153185

Al-Turjman, F., & Alturjman, S. (2018). Context-Sensitive

Access in Industrial Internet of Things (IIoT)

Healthcare Applications. IEEE Transactions on

Industrial Informatics, 14(6), 2736–2744.

https://doi.org/10.1109/TII.2018.2808190

Baghdadi, A., Cavuoto, L. A., Jones-Farmer, A., Rigdon, S.

E., Esfahani, E. T., & Megahed, F. M. (2019).

Monitoring worker fatigue using wearable devices: A

case study to detect changes in gait parameters. Journal

of Quality Technology, 0(0), 1–25.

https://doi.org/10.1080/00224065.2019.1640097

Cai, W., Chen, S., & Zhang, D. (2009). A simultaneous

learning framework for clustering and classification.

Pattern Recognition, 42(7), 1248–1259.

https://doi.org/10.1016/j.patcog.2008.11.029

Charbonnier, S., Roy, R. N., Bonnet, S., & Campagne, A.

(2016). EEG index for control operators’ mental fatigue

monitoring using interactions between brain regions.

Expert Systems with Applications, 52, 91–98.

https://doi.org/10.1016/j.eswa.2016.01.013

Dogan, A., & Birant, D. (2021). Machine learning and data

mining in manufacturing. Expert Systems with

Applications, 166, 114060. https://doi.org/10.1016/

j.eswa.2020.114060

Elmaraghy, W. H., Nada, O. A., & Elmaraghy, H. A.

(2008). Quality prediction for reconfigurable

manufacturing systems via human error modelling.

International Journal of Computer Integrated

Manufacturing, 21(5), 584–598. https://doi.org/10.10

80/09511920701233464

Fardhosseini, M. S., Habibnezhad, M., Jebelli, H.,

Migliaccio, G., Lee, H. W., & Puckett, J. (2020).

Recognition of Construction Workers’ Physical Fatigue

Based on Gait Patterns Driven from Three-Axis

Accelerometer Embedded in a Smartphone. 453–462.

https://doi.org/10.1061/9780784482872.049

Givi, Z. S., Jaber, M. Y., & Neumann, W. P. (2015).

Modelling worker reliability with learning and fatigue.

Applied Mathematical Modelling, 39(17), 5186–5199.

https://doi.org/10.1016/j.apm.2015.03.038

Govaerts, R., Tassignon, B., Ghillebert, J., Serrien, B., De

Bock, S., Ampe, T., El Makrini, I., Vanderborght, B.,

Meeusen, R., & De Pauw, K. (2021). Prevalence and

incidence of work-related musculoskeletal disorders in

secondary industries of 21st century Europe: A

systematic review and meta-analysis. BMC

Musculoskeletal Disorders, 22(1), 751.

https://doi.org/10.1186/s12891-021-04615-9

Grandjean, E. (1979). Fatigue in industry. Occupational

and Environmental Medicine

, 36(3), 175–186.

https://doi.org/10.1136/oem.36.3.175

Halim, I., Omar, A. R., Saman, A. M., & Othman, I. (2012).

Assessment of Muscle Fatigue Associated with

Prolonged Standing in the Workplace. Safety and

Health at Work, 3(1), 31–42. https://doi.org/

10.5491/SHAW.2012.3.1.31

Iskander, J., Hossny, M., & Nahavandi, S. (2018). A

Review on Ocular Biomechanic Models for Assessing

Visual Fatigue in Virtual Reality. IEEE Access, 6,

IoTBDS 2024 - 9th International Conference on Internet of Things, Big Data and Security

308

19345–19361.

https://doi.org/10.1109/ACCESS.2018.2815663

Li, Z., Wang, Y., & Wang, K.-S. (2017). Intelligent

predictive maintenance for fault diagnosis and

prognosis in machine centers: Industry 4.0 scenario.

Advances in Manufacturing, 5(4), 377–387.

https://doi.org/10.1007/s40436-017-0203-8

Ma, L., Chablat, D., Bennis, F., & Zhang, W. (2009). A new

simple dynamic muscle fatigue model and its

validation. International Journal of Industrial

Ergonomics, 39(1), 211–220. https://doi.org/10.1016/

j.ergon.2008.04.004

Manavalan, E., & Jayakrishna, K. (2019). A review of

Internet of Things (IoT) embedded sustainable supply

chain for industry 4.0 requirements. Computers &

Industrial Engineering, 127, 925–953.

https://doi.org/10.1016/j.cie.2018.11.030

Nur, N. M., Dawal, S. Z., & Dahari, M. (2014, January 7).

The Prevalence of Work Related Musculoskeletal

Disorders Among Workers Performing Industrial

Repetitive Tasks in the Automotive Manufacturing

Companies. International Conference on Industrial

Engineering and Operations Management, Bali,

Indonesia.

Pabolu, V. K. R., & Shrivastava, D. (2021). A dynamic job

rotation scheduling conceptual framework by a human

representing digital twin. Procedia CIRP, 104, 1367–

1372. https://doi.org/10.1016/j.procir.2021.11.230

Pabolu, V. K. R., Shrivastava, D., & Kulkarni, M. S. (2022).

A Dynamic System to Predict an Assembly Line

Worker’s Comfortable Work-Duration Time by Using

the Machine Learning Technique. Procedia CIRP, 106,

270–275. https://doi.org/10.1016/j.procir.2022.02.190

Sadeghniiat-Haghighi, K., & Yazdi, Z. (2015). Fatigue

management in the workplace. Industrial Psychiatry

Journal, 24(1), 12–17. https://doi.org/10.4103/0972-

6748.160915

Sawatzky, S. (2017). Worker Fatigue: Understanding the

Risks in the Workplace. Professional Safety, 62(11),

45–51.

Sedighi Maman, Z., Chen, Y.-J., Baghdadi, A., Lombardo,

S., Cavuoto, L. A., & Megahed, F. M. (2020a). A data

analytic framework for physical fatigue management

using wearable sensors. Expert Systems with

Applications, 155, 113405. https://doi.org/10.1016/j.es

wa.2020.113405

Sedighi Maman, Z., Chen, Y.-J., Baghdadi, A., Lombardo,

S., Cavuoto, L. A., & Megahed, F. M. (2020b). A data

analytic framework for physical fatigue management

using wearable sensors. Expert Systems with

Applications, 155, 113405. https://doi.org/10.1016/j.es

wa.2020.113405

Sundstrup, E., Seeberg, K. G. V., Bengtsen, E., &

Andersen, L. L. (2020). A Systematic Review of

Workplace Interventions to Rehabilitate

Musculoskeletal Disorders Among Employees with

Physical Demanding Work. Journal of Occupational

Rehabilitation, 30(4), 588–612. https://doi.org/10.100

7/s10926-020-09879-x

Usuga Cadavid, J. P., Lamouri, S., Grabot, B., Pellerin, R.,

& Fortin, A. (2020). Machine learning applied in

production planning and control: A state-of-the-art in

the era of industry 4.0. Journal of Intelligent

Manufacturing, 31(6), 1531–1558. https://doi.org/

10.1007/s10845-019-01531-7

Vogl, G. W., Weiss, B. A., & Helu, M. (2019). A review of

diagnostic and prognostic capabilities and best

practices for manufacturing. Journal of Intelligent

Manufacturing, 30(1), 79–95. https://doi.org/10.1007/

s10845-016-1228-8

Wang, L. (2019). From Intelligence Science to Intelligent

Manufacturing. Engineering, 5(4), 615–618.

https://doi.org/10.1016/j.eng.2019.04.011

Yang, F., Di, N., Guo, W., Ding, W., Jia, N., Zhang, H., Li,

D., Wang, D., Wang, R., Zhang, D., Liu, Y., Shen, B.,

Wang, Z., & Yin, Y. (2023). The prevalence and risk

factors of work related musculoskeletal disorders

among electronics manufacturing workers: A cross-

sectional analytical study in China. BMC Public Health,

23(1), 10. https://doi.org/10.1186/s12889-022-14952-6

Industrial Internet of Things for Assembly Line Worker’s Work Fatigue Recognition

309