A longitudinal single-subject experimental design following the ABAB sequence [19] will be developed in a manufacturing industry real context. In the ABAB design, first, a baseline is determined (A), then the intervention is introduced, with the use of the smart wear and real-time feedback to the worker for the first time (B). Next, the intervention is withdrawn, returning to the baseline condition (A). Finally, after baseline stability is reestablished, the intervention is presented (B) for the second time [1,2]. All 4 periods will end with an assessment of the selected outcomes. Figure 1 depicts the study flow. Generally, in this study design, the research is done very thoroughly on 3 to 5 subjects. The interest is always in the meticulous examination of each participant individually, not in the group's average performance. Multiple single-subject experimental studies also present the opportunity to consider not only the individual variables but also analyze the trial at the environmental level [19,20].

All participants will give informed consent to participate. All procedures are handled according to the Declaration of Helsinki and regulatory legislation, which includes approval by an ethics committee. The protocol is under review by the Ethics Committee of the University of Aveiro (17-CED/2023). Data will be processed complying with the European General Data Protection Regulation.

This study will include 5 active workers from an industrial manufacturing setting of household appliances in the Region of Aveiro, Portugal. To be included, workers must be aged between 35 and 45 years and have a work experience with the tested workstation for at least 6 months. Exclusion criteria are (1) history of vestibular disorders, (2) neurological, respiratory disease, or spinal surgery, (3) structural spinal problems, (4) medication or a health condition affecting balance, and (5) regular treatments for a musculoskeletal condition.

The workers will be recruited by the occupational health department, which will provide the identification numbers of each selected worker as well as shoe and clothing sizes.

A specific task representing a repetitive movement was chosen. The task consists of the vertical screw tightening of 5 screws in a standing position into a piece placed horizontally. A first assessment period will be executed to analyze baseline repetitive work (A) outcomes. Subsequently, the workers will use the smart wearables and be provided with real-time feedback for 4 weeks (B). A second assessment (assessment 2) will be performed to explore the efficacy of the system. Afterward, the workers will have a washout period of 4 weeks to reestablish baseline repetitive work condition (A), followed by assessment 3. Finally, a second 4-week period of intervention (B) and consequent assessment 4, will take place (Figure 1). Each assessment moment will be accomplished in 5 consecutive days, in 4 periods of the day, which includes measurements for each period of 5 cycles of the performed task in the first 10 minutes after starting the shift (to avoid the warm-up period), 10 minutes before the break time, 10 minutes after the break, and 10 minutes before the end of the day's shift. All tests will be performed guaranteeing that in the anterior 48-hour period each individual did not participate in strenuous physical activities, which may affect the selected outcomes.

The smart wearable systems consist of a sensorized jacket, smart footwear, a smart band, and a feedback system provided to the worker. The sensorized jacket aims to monitor the upper limb (shoulder flexion/extension and abduction/adduction and elbow flexion/extension) and trunk (lumbar flexion/extension) with strategically placed inertial measurement unit (IMU) systems and body temperature with a temperature sensor. Smart footwear consists of a shoe equipped with a set of 6 sensors Force Sensitive Resistors (Sensitronics), to map plantar pressures. The sensors have a 30-mm diameter and work in a 0-300–N force range. They were assembled in a custom substrate printed by RokuPrint (RP 2.2, RokuPrint GmbH). A smart band will monitor heart rate. The main system architecture is presented in Figure 2.

The primary outcomes are fatigue and musculoskeletal symptoms. Fatigue will be assessed by electromyography (EMG) and musculoskeletal symptoms by the Nordic Musculoskeletal Questionnaire [21].

Secondary outcomes include perceived effort, assessed by the Borg Rating of Perceived Exertion (RPE), range of motion of the main joints in the upper body and trunk, speed, acceleration, and deceleration of movement, through motion analysis, risk stratification of range of motion of the upper limb and trunk based on limits defined in the Rapid Upper Limb Assessment (RULA) [22], and cycle duration in minutes.

For patient characterization, baseline questionnaires including sociodemographic data, physical activity level, quality of life, anxiety and depression, insomnia and sleep quality, fatigue, and pain and disability will be collected. Table 1 displays the assessment flow.

The collection and data storage will be made in digital format, including those from the questionnaires. The entered data will be checked for completeness, validity, and plausibility by a member of the research team.

Structured visual analysis techniques will be conducted to observe the effects of the intervention using a Single-Case Visual Analysis package for R. Graphical (RStudio) representations of data for each participant will be plotted considering primary and secondary outcomes on the y-axis and time on the x-axis. The results for each variable of interest will be compared in-between the different time points of the work shift and longitudinally considering each assessment day as a time point.

For each outcome (primary or secondary) and each moment mean, median, range, and stability envelope will be calculated. The level of change and difference between the first and last values of each period will be compared. The next step includes the calculation trend using the split middle method of trend estimation. After this procedure, percent of data points within the stability envelope (80% of data points between ±25% of median) will be considered. The final step is using the “freehand method” to evaluate data path [38].

Given the high prevalence of WMSDs and the associated economic and personal burden, preventive measures such as smart wear for monitoring and providing feedback are crucial. The smart wearables analyzed under this study aim to create awareness of WMSDs risk and empower the worker to reduce the associated risk and have a healthier work environment and higher quality of life. Investing in preventive measures is especially rewarding, since workers in countries and sectors where more preventive measures are in place are less likely to report MSD complaints [10]. Consequently, preventive measures result in an increase in economic benefits for the industries with greater production and worker adherence to these jobs.

The proposed study is, to our best knowledge, one of the first to use devices that provides real-time feedback to industrial workers about posture. This real-time feedback will include the body location at risk, which can lead to the correction of body biomechanics. Posture and movement monitoring technology can help detect and subsequently reduce the compensatory movement patterns, promoting a decrease in fatigue and musculoskeletal symptoms.

This study is a single-case experimental design. This method design was primarily developed to assess individual behavior. Given the main goal is to develop and test smart wear for reducing fatigue and poor biomechanics, a single-subject approach seems adequate. This design enables the possibility of rapidly determining if the wearables are effective. Immediate changes can be observed firsthand, quickly, and in individual participants. Consequently, adjustments can be rapidly made, if necessary. Contrary to the single-subject approach, a large sample statistical approach will require weeks or months of participant testing, mean calculations, and statistical analyses before the effectiveness may be determined. Conceptually, the treatment condition must influence individual behavior and assure both research intra- and interparticipant replication. Therefore, an ABAB design will be executed, assuring both intra- and interparticipant intervention and control [38]. Graphical analysis of the variability of each outcome, in each of the repeated measures in time, will be performed to explore the possible indicators of fatigue when performing the selected activity.

This proposed study will explore a strategy for reducing fatigue and musculoskeletal symptoms in industrial workers, possibly preventing WMSDs, using smart wearables that provide real-time feedback about biomechanics. Results would showcase a novel approach for improving postural awareness and self-awareness of risk for WMSD providing an evidence-based support for the use of such devices. Further research could explore other health indicators and individualized approaches to prevent similar health risks.