A method for postural critical factors checking: The case of library chairs

Abstract

BACKGROUND:

According to ergonomic researches regarding a good sitting posture, it is essential to ensure a natural back-curve in order to prevent musculoskeletal disorders. A brief observation among the Scientific Technology Library inside the University of Salerno showed that students used to complain about neck and lumbar pain, especially after a study day.

OBJECTIVE:

On the light of this background, a sitting posture comfort analysis had been performed on chairs inside the library to check the critical factors that influence the postural comfort and, consequently, the learning.

METHODS:

A prolonged sitting posture, that is common during the daily study activity, had been simulated with fifteen volunteer students performing 1-hour tests (divided into four 15-minutes tasks). Subjective perceptions had been gathered through questionnaires rating on a 5-point Comfort scale, both the expected comfort at the beginning of the experiment and the Localized Postural Comfort at the end of each task have been investigated. Then, postural angles had been gathered through photographic acquisition and Kinovea®. CaMAN software had been used to calculate the objective (dis)comfort indexes. Finally, subjective and objective data had been statistically processed and compared.

RESULTS:

Lumbar area scored the lowest perceived comfort while the perceived comfort was independent of participants and tasks, but dependent on time.

CONCLUSIONS:

After this comfort-driven analysis, critical factors of the chair-design were checked, and a proposal for a future re-design was hypothesized.

Keywords

Comfort office seat students library postural comfort-driven redesign

1 Introduction

Students spend the majority of their time studying, usually sitting on a chair. The importance of the environment cannot be underestimated since negative feelings can affect the learning, especially for a long time sitting [1, 2]. Indeed, uncomfortable and awkward body postures can decrease a student’s in-terest in learning, even during the most stimulating and interesting lessons [3]. As far as the position of chair and desk, there have been several studies in the literature regarding the correct sitting posture and the awareness of a good sitting posture [4 –7]. Furthermore, it exists even an equation to quantify the comfort in the function of measurements and distan-ces between chairs, student, and desk [8].

The context influences any seat design. Some stu-dies, moreover, gave guidelines to design a comfor-table seat, taking into account the natural curve of backbones, the body sensitivity [9 –12], the perfor-med activities, and anthropometric measurements of the target group [13 –17]. Different target groups have different body sizes, and this implies differences in seat width, backrest length, seat pan length, armrest height, which should be designed to fit at least the 95% of the population [10]. However, the body’s optimal position in terms of comfort requires every joint and eye position to be close to the neutral position, where the perception of comfort is high [18 –21].

‘Postural comfort’ is commonly defined as the ab-sence of discomfort, or a state where the need to cha-nge position is not present [22, 23]. The comfort zone, defined as the area of the most comfortable motions/postures for a given task, does not predicate an absolute measure of well-being. Users within their comfort zone are unlikely to change into other postures.

The evaluation of postural comfort can be achieved through the acquisition of subjective or objective data. Subjective data are related with questionnaires, such as Localized Postural (Dis)comfort (LPD), Body Part (Dis)comfort (BPD), etc. [24 –26], while the objective one can be obtained with tools that analyze and process data coming from pressure mats, sensors and so forth [27, 28] or evaluate (dis)comfort through the acquisition of postural angles. One of these tools is the software CaMAN [21 , 29–31] realized by the Engineering Design Methods group at the University of Salerno: the software considers the human joints and the comfort curves over angles associated with them. Thus, for a given angle of a human joint, the software gives the associated comfort index (on an 11-point scale where 10 is the maximum comfort).

Despite this background, some applications in da-ily life do not follow the ergonomists’ tips, as shown in this study.

The Science and Technology Library (S&T Lib-rary), designed by the architect Nicola Pagliara [32], is located inside the campus of the University of Salerno (UNISA) and is actually used as a place to study [33].

With a brief analysis among students (an anonymous questionnaire spread out among students thr-ough social networks, that received more than 200 answers) inside the S&T Library, it came out there had been several complaints about neck and lumbar pain after a study day. Regarding this, one hypothesis was the students used to assume the wrong sitting posture on those chairs.

Since the students tended to change posture frequently, a sitting postural comfort analysis had been done [34 –36] applying a methodology that evaluates the critical factors for postural comfort in order to have a systematic basis for the redesign of the chair.

Thus, the research question is: considering the current chair, which are the critical factors that influence the postural comfort? Moreover, once defined them, how could these factors be involved in the redesign process?

2 Materials and method

2.1 Experiment setup

The experiment had been set up on the last floor of the S&T Library when there was less affluence of students by permission of the library staff and of the university manager.

On each floor, there are 36 desks with corresponding chairs, grouped six by six, where three are aligned, and the other three are opposed to them. For the experiment, three consecutive desks had been occupied with having a free space.

Three Nikon D3300 cameras had been used and fixed on tripods among the desks: one had been placed on the left and one on the right to obtain the lateral views; and the last one behind the chair, at an adequate distance, to obtain the rearview. Also, one phone-camera had been stably fixed on selfie-stick support to take photos from the top view (see Fig. 1).

Fig. 1

Example of photos acquisition from the four views. The chair was positioned as close as possible to the desk – in Chapter 2.1 and Chapter 2.3.

Two main tasks had been performed to simulate a study day: writing and reading. Thus, books, pens and papers had been provided. To consider the time effect, each experiment lasted 1 hour, where the two tasks had been performed for 15 minutes each one, switching them at the end of the 15 minutes timeslot. Between the tasks, a pause of 1 minute had been given in order to fill the questionnaire. Photos had been taken from all cameras simultaneously at the end of each task to capture body posture and obtain then the changing of postures over time. The task order was chosen randomly in order to avoid any task-order effect.

2.2 Experimental sample

Fifteen students of the University of Salerno, 8 males and 7 females aged between 23 and 31, took part in the experiment. Table 1 shows the anthropometrics data of participants. All the students enjoyed good health. These anthropometric data had been gathered, measuring the participants’ body parts directly and recording them in an Excel^® file.

Table 1
Anthropometrics data of participants –in Chapter 2.2

Male (n = 8) Female (n = 7)

Mean SD Range Mean SD Range

Height (cm) 178.5 6.2 168 –185 162.7 5.7 154 –169

Weight (kg) 72.5 10.2 57 –90 57.1 4.8 50 –63

	Male (n = 8)	Female (n = 7)
Height (cm)	178.5	6.2	168 –185	162.7	5.7	154 –169
Weight (kg)	72.5	10.2	57 –90	57.1	4.8	50 –63

2.3 The chair

The library is furnished with 600 identical chairs, designed by the same architect that designed the whole building. No other kind of chairs are available in the library. So, to obtain a complete overview, the dimensions of the chair had been compared with human body measurements. The dimensions of the chair are shown in Fig. 2.

Fig. 2

Pictures of the chair in three views. Measurements of the chair were measured and reported in figure –in Chapter 2.3 and Chapter 4.

The height of armrests is about 61 cm from the ground, while the lower part of the desk is 60 cm height from the floor: it means the chair cannot be positioned under the desk, as visible in Fig. 1. Moreover, there is a gap between the backrest and the seat-pan of about 14 cm (66,1 cm –52,4 cm); it means students have to move backward, by bending their spine, in order to lay on the backrest.

From DINED [37], choosing the international population, values regarding the sitting height, the hip breadth, popliteal height, buttock-knee depth, and elbow-grip length had been gathered, as shown in Fig. 3.

Fig. 3

Anthropometric measurements from DINED. The numbers refer on the picture placed on the right. Measurements refers on 5- and 50-percentile of the international female population, 50- and 95-percentile of the international male population –in Chapter 2.3 and Chapter 4.

Comparing the measurements, they had been figured out the following pitfalls:

Popliteal height is not suitable for P50 of the female population;

Buttock-knee depth is not suitable for P50 of the female population;

The hip breadth for both populations is smaller than the seat pan width.

Thus, the expected results of comfort analysis are not favorable.

2.4 Questionnaires

Localized Postural Comfort Questionnaires (see Fig. 4) have been used to collect subjective data regarding the postural comfort of participants. Before the experiment, participants were asked to rate on a 5-point Comfort scale (from 1 = No comfortable to 5 = Extremely comfortable) the expectation of perceived comfort once sitting on the chair, that is, how the chair seemed comfortable at first sight [38].

Fig. 4

Questionnaire used during the experiments. The first question (expected comfort) was asked only once, before the experiment. The second question (# task) was repeated after each performed task – in Chapter 2.4.

At the end of each task, participants were asked to rate on a 5-point Comfort scale [26, 39]:

The perceived comfort on the following body parts: neck, left shoulder, right shoulder, left arm, right arm, left forearm, right forearm, left wrist, right wrist, thoracic area, lumbar area;

The global comfort.

2.5 Experiment protocol

The experiment protocol meets the requirements of the Ethical Committee regulation for the University of Salerno. Prior to the experiment, participants have been asked to sign an informed consent and instructed about the experiment. They performed experiments one by time, scheduling their free time with the availability of the place. Then participants sat on the chair, positioning it closer to the desk and assuming a correct sitting posture, that is, forearms on the desk, rai-sed back, 90 ° legs, and feet leaning against the floor.

Tasks have been performed in sequence, alternating between writing and reading, both among the tasks and the sequential participants (Table 2). Each one lasted 15 minutes, a stopwatch told the time, and with a pause of about 1 min between tasks to fill the questionnaire; photos have been taken just before the end of the task to capture the mean posture var-iation over time considering the assumed posture after each task. While participants were filling the questionnaire, the examiners recorded their feedback, according to survey investigations methods [40], to achieve as much information as possible. Survey data have been analyzed calculating weighted averages and the comfort trend over time starting from the expectation.

Table 2
Protocol regarding time –in Chapter 2.5 and Chapter 4

Task 1 Task 2 Task 3 Task 4

Participant A Reading Writing Reading Writing

Participant B Writing Reading Writing Reading

15 min 15 min 15 min 15 min

1 min 1 min 1 min

	Task 1	Task 2	Task 3	Task 4
Participant A	Reading	Writing	Reading	Writing
Participant B	Writing	Reading	Writing	Reading
	15 min	15 min	15 min	15 min
		1 min	1 min	1 min

2.6 Postural angles and the simulation

A total of 240 photos (15 participants x 4 tasks x 4 views) have been analyzed with Kinovea® to gather postural angles, trying to be as accurate as possible, aware of any human errors, both in visual perception and in the program operation.

Analysis has been made of the following upper limb movements: head rotation, head bending, head flexion, shoulder rotation, shoulder bending, shoulder flexion, trunk rotation, trunk bending, and trunk flexion. Body rotation has been analyzed in the transverse plane, body flexion in the frontal plane, and body bending in the sagittal plane. Considering the aforementioned correct sitting posture as a reference posture, gathered angles had been defined as the deviations from the reference posture.

A virtual environment of S&T Library has been realized in Delmia® (Fig. 5), representing one floor with fifteen students. French mannequins, that represent the South-European standard, have been used to simulate participants’ movements through the gathered postural angles. Anthropometric data, movements, and tasks have been respected. Through the simulation, it was possible to see the temporal changes for each student, going from a correct posture to the last one gathered.

Fig. 5

Virtual representation of S&T Library on Delmia® – in Chapter 2.6 and Chapter 4.

2.7 CaMAN: Objective (dis)comfort indexes

The CaMAN software [21] has been used to obtain objective comfort indexes from the collected angles, and it is based on experimental studies conducted by Naddeo et al. to give a comfort index according to postural angles assumed by human joints. As far as upper limbs, it considers:

Neck: frontal flexion, rotation, and lateral flexion

Shoulder: frontal flexion, abduction/adduction

Elbow: flexion/extension, pronation/supination

Wrist: flexion/extension, radio/ulnar deviation

For each joint, curves of postural (dis)comfort over angles are used. (Dis)comfort indexes are rated on a 10-point scale where 1=“no comfortable” and 10=“extremely comfortable” (see Table 3). These comfort indexes consider the limbs moving freely in the space, without any support or load: as students used to lay their wrist, forearms, and elbow on the desk for the whole of time, only neck and shoulder comfort indexes have been evaluated. A conversion of the CaMAN scale from 11-point to 5-point has been done to compare the CaMAN results with the ones from Questionnaires, as shown in Table 3.

Table 3
Conversion scale to compare questionnaire scale with the CaMAN one – in Chapter 2.7

Original CaMAN Scale 1 2 3 4 5 6 7 8 9 10

Translated CaMAN Scale 1 2 3 4 5

Questionnaire Scale 1 2 3 4 5

Original CaMAN Scale	1	2	3	4	5	6	7	8	9	10
Translated CaMAN Scale	1	2	3	4	5
Questionnaire Scale	1	2	3	4	5

2.8 Statistical methods

Wilcoxon Signed Rank test has been performed to calculate whether there were significant differences between data, particularly between tasks. Correlati-ons between subjective comfort indexes from questionnaires and objective comfort indexes from CaMAN have been calculated with IBM® SPSS® Sta-tistics version 24 to evaluate the strength of the relationship between these variables. Spearman correlations have been computed since data could be not normally distributed [41].

This analysis has been done taking into account even the dependence on the type of activity; it means to evaluate whether the comfort depends or not on the initial performed task (writing or reading). Thus, correlations have been calculated between the following comfort indexes:

Body parts from Questionnaires & body parts by CaMAN

Body parts from Questionnaires & Global Comfort from Questionnaires

Body parts by CaMAN & Global Comfort from Questionnaire

The critical factors are defined as those variables with lower perceived comfort and presented a high correlation with other variables, particularly with the Global Comfort. It means, improving the perceived comfort of those parts, the global perceived comfort could arise.

3 Results

With 15 participants and 4 tasks, 60 questionnaires (subjective data) and 240 photos, obtaining 60 postures to analyze in terms of postural angles (objective data), have been collected. Then, Spearman correlations have been calculated between these subjective and objective data to evaluate the strengths of the relationships. With regards to the trend of global comfort over time, results from the questionnaires are shown in Fig. 6. The values represent the average of expected comfort and global comfort for each task. In this case, “Tasks 1, 2, 3, 4” are the mean value of the global comfort rated by participants, without considering the type of the task, just the effect of the prolonged sitting on the chair. Thus, “Task 1” is the average global comfort rated after 15 minutes of sitting, “Task 2” after 30 minutes, “Task 3” after 45 minutes, and “Task 4” after 60 minutes. There was a decay over time, starting from a higher comfort expectation to the lower perceived comfort in “Task 4”.

Fig. 6

Evolution of the average global comfort over time, starting from the expected comfort and going to the perceived global comfort rated by participants. Values are the average mean on a scale from 1=”no comfortable” to 5=”Extremely comfortable”. –in Chapter 3 and Chapter 4.

Figure 7 shows the values of the average mean of postural comfort for each body part coming from the questionnaires. Comfort indexes in “Task 4” scored lower values than the ones in “Task 1”: this confirms the comfort decay over time. Furthermore, the lumbar area scored the lowest values of comfort, followed by the neck, torso, and shoulders, while the arms, forearms, and wrist scored the highest values.

Fig. 7

Mean values from questionnaires for each body part (1 = no comfortable; 5 = extremely comfortable) –in Chapter 3.

Wilcoxon test has been performed to compare each task, and results were significant at p < 0.05, especially between “Task 1” and “Task 4”. It means there are significant differences between the first task and the last task.

Furthermore, in Table 4 are reported mean, maximum, and minimum values of perceived comfort, considering the whole experiment (1-hour experiment, thus grouping all tasks) and the comfort decay over time. The decay was calculated considering only the first task and the last task, to understand the influence of the time on the perceived body part comfort.

Table 4

Mean, maximum and minimum values of comfort indexes and values of comfort decay over time. (n = 15) –in Chapter 3 and Chapter 4

	Comfort index (scale 1–5)
	Mean	Max	Min	Decay
Neck	2.12	2.60	1.80	28%
Right shoulder	2.52	3.13	1.93	38%
Left shoulder	2.47	2.87	2.00	30%
Right arm	2.87	3.33	2.33	30%
Left arm	2.98	3.27	2.60	20%
Right forearm	2.90	3.53	2.33	34%
Left forearm	2.95	3.47	2.53	27%
Right wrist	3.25	3.80	2.80	26%
Left wrist	3.28	3.67	2.93	20%
Thoracic area	2.25	3.07	1.73	43%
Lumbar area	1.72	2.13	1.40	25%
Global comfort	2.22	2.73	1.73	37%

The critical areas are clearly shown in Table 4, it is, the thoracic area and the lumbar area supposedly due to the absence of support for the back. Indeed, the thoracic area presented the highest decay value, while the lumbar area the lowest comfort value.

Considering the postural angles, Fig. 8 shows an example of posture changing over time where, starting from the correct one, participants used to bend over time their back more forward, ending with a slouched posture, trying to unload the upper body weight to the desk.

Fig. 8

Example of posture changing over time starting from a correct one to a slouched one (only one view has considered in this example) – in Chapter 3.

Figure 9 shows the values of the average mean of postural comfort for each body part coming from CaMAN. Comfort indexes in “Task 4” scored lower values than the ones in “Task 1”: this confirms in part the comfort decay over time, like the questionnaire-results above. The elbows and wrists were not considered because participants used to lay them on the desk during the experiment.

Fig. 9

Mean values from CaMAN for each body part (0 = no comfortable; 5 = extremely comfortable) –in Chapter 3 and Chapter 4.

3.1 Correlations

It has been found that height, weight, and gender did not affect neither any body part or the Global Comfort. Within each task, strong correlations between subjective data (Questionnaire) and objective data (CaMAN) were found (mean p∼0.8).

Table 5 shows the most important correlations, it is, between subjective (Questionnaire) and objective (CaMAN) data for each task. The Global comfort from Questionnaires is always correlated with the other variables; this means each body part influenced the perceived comfort. The Lumbar area presents a strong correlation with the Global Comfort, except for the third task, while, the Thoracic area shows a significant correlation with the Global Comfort only in the last two tasks. Thus, the Lumbar area has ma-jor influence rather than the Thoracic area on the Global Comfort, this means, improving the perceived comfort on the Lumbar area particularly, the Global Comfort will arise. Meanwhile, improving on the Thoracic area, the effects will be seen over time. There are strong correlations between the global comfort from Questionnaire and CaMAN only in the first and in the last task. This aspect highlights the importance of the first impact with the chair and the effect of perceived comfort over time.

Table 5
Calculated correlations between Questionnaires and CaMAN data for each task (only correlations within each task are considered, for example, in the section “Task 1”, are considered just correlations between data gathered from questionnaire and CaMAN in the first task). The Table highlight only correlations between subjective (Questionnaire) and objective (CaMAN) data – in Chapter 3.1

Global Comfort Neck Right Shoulder Left Shoulder Global Comfort

Questionnaire CaMAN CaMAN CaMAN CaMAN

Task 1 Neck Questionnaire 0,614^* 0,658^** 0,614^*

Right Shoulder Questionnaire 0,815^** 0,526^* 0,793^** 0,550^*

Left Shoulder Questionnaire 0,703^ 0,776^

Right Arm Questionnaire 0,580^*

Left Arm Questionnaire 0,492

Right Forearm Questionnaire 0,641^*

Left Forearm Questionnaire 0,562^*

Right Wrist Questionnaire 0,813^**

Left Wrist Questionnaire 0,700^**

Thoracic Area Questionnaire 0,534^*

Lumbar Area Questionnaire 0,699^** 0,593^*

Global Comfort Questionnaire 1 0,783^** 0,573^*

Task 2 Neck Questionnaire 0,592^*

Right Shoulder Questionnaire 0,711^ 0,751^

Left Shoulder Questionnaire 0,664^ 0,727^

Right Arm Questionnaire 0,629^*

Left Arm Questionnaire 0,628^*

Right Forearm Questionnaire

Left Forearm Questionnaire 0,657^**

Right Wrist Questionnaire

Left Wrist Questionnaire

Thoracic Area Questionnaire

Lumbar Area Questionnaire 0,739^**

Global Comfort Questionnaire 1,000 0,536^*

Task 3 Neck Questionnaire 0,736^**

Right Shoulder Questionnaire 0,638^* 0,544^*

Left Shoulder Questionnaire 0,601^*

Right Arm Questionnaire 0,694^**

Left Arm Questionnaire

Right Forearm Questionnaire 0,539^*

Left Forearm Questionnaire

Right Wrist Questionnaire

Left Wrist Questionnaire 0,557^*

Thoracic Area Questionnaire 0,521^*

Lumbar Area Questionnaire 0,532^*

Global Comfort Questionnaire 1,000

Task 4 Neck Questionnaire 0,652^**

Right Shoulder Questionnaire 0,760^**

Left Shoulder Questionnaire 0,702^**

Right Arm Questionnaire

Left Arm Questionnaire

Right Forearm Questionnaire 0,542^*

Left Forearm Questionnaire

Right Wrist Questionnaire

Left Wrist Questionnaire 0,528^*

Thoracic Area Questionnaire 0,620^*

Lumbar Area Questionnaire 0,792^**

Global Comfort Questionnaire 1,000 0,589^* 0,602^*

		Global Comfort	Neck	Right Shoulder	Left Shoulder	Global Comfort
Task 1	Neck Questionnaire	0,614^*		0,658^**	0,614^*
	Right Shoulder Questionnaire	0,815^**	0,526^*		0,793^**	0,550^*
	Left Shoulder Questionnaire	0,703^**			0,776^**
	Right Arm Questionnaire	0,580^*
	Left Arm Questionnaire	0,492
	Right Forearm Questionnaire	0,641^*
	Left Forearm Questionnaire	0,562^*
	Right Wrist Questionnaire	0,813^**
	Left Wrist Questionnaire	0,700^**
	Thoracic Area Questionnaire	0,534^*
	Lumbar Area Questionnaire	0,699^**		0,593^*
	Global Comfort Questionnaire	1			0,783^**	0,573^*
Task 2	Neck Questionnaire	0,592^*
	Right Shoulder Questionnaire	0,711^**	0,751^**
	Left Shoulder Questionnaire	0,664^**	0,727^**
	Right Arm Questionnaire	0,629^*
	Left Arm Questionnaire	0,628^*
	Right Forearm Questionnaire
	Left Forearm Questionnaire	0,657^**
	Right Wrist Questionnaire
	Left Wrist Questionnaire
	Thoracic Area Questionnaire
	Lumbar Area Questionnaire	0,739^**
	Global Comfort Questionnaire	1,000	0,536^*
Task 3	Neck Questionnaire	0,736^**
	Right Shoulder Questionnaire	0,638^*	0,544^*
	Left Shoulder Questionnaire	0,601^*
	Right Arm Questionnaire	0,694^**
	Left Arm Questionnaire
	Right Forearm Questionnaire	0,539^*
	Left Forearm Questionnaire
	Right Wrist Questionnaire
	Left Wrist Questionnaire	0,557^*
	Thoracic Area Questionnaire	0,521^*
	Lumbar Area Questionnaire	0,532^*
	Global Comfort Questionnaire	1,000
Task 4	Neck Questionnaire	0,652^**
	Right Shoulder Questionnaire	0,760^**
	Left Shoulder Questionnaire	0,702^**
	Right Arm Questionnaire
	Left Arm Questionnaire
	Right Forearm Questionnaire	0,542^*
	Left Forearm Questionnaire
	Right Wrist Questionnaire
	Left Wrist Questionnaire	0,528^*
	Thoracic Area Questionnaire	0,620^*
	Lumbar Area Questionnaire	0,792^**
	Global Comfort Questionnaire	1,000		0,589^*		0,602^*

^**. Correlation is significant at the 0.01 level (2-tailed). ^*. Correlation is significant at the 0.05 level (2-tailed).

Table 6 shows the calculated correlations between the subjective and objective Global Comfort values for each task. The presence of significant correlations confirms the idea that each task could influence others.

Table 6

Calculated correlations between Questionnaires and CaMAN considering the global comfort of the first and the last tasks – in Chapter 3.1

		Global Comfort – Task 1		Global Comfort – Task 2		Global Comfort – Task 3		Global Comfort – Task 4
		Question.	CaMAN	Question.	CaMAN	Question.	CaMAN	Question.	CaMAN
Global Comfort –	Question.	1,000	0,745^**			0,794^**
Task 1	CaMAN	0,745^**	1,000	0,644^**		0,701^**			0,664^**
Global Comfort –	Question.		0,644^**	1,000		0,669^**		0,730^**
Task 2	CaMAN				1,000
Global Comfort –	Question.	0,794^**	0,701^**	0,669^**		1,000		0,632^*
Task 3	CaMAN						1,000	0,270
Global Comfort –	Question.			0,730^**		0,632^*		1,000	0,602^*
Task 4	CaMAN		0,664^**					0,602^*	1,000

^**. Correlation is significant at the 0.01 level (2-tailed). ^*.Correlation is significant at the 0.05 level (2-tailed).

Outcomes from the results:

Strong correlations between body part questionnaire & body part CaMAN (mean p∼0.8)

Strong correlations between body part questionnaire & Global Comfort Questionnaire (mean p∼0.7)

Strong correlations between body part CaMAN & Global Comfort Questionnaire (mean p∼0.6)

Doing the same analysis by grouping the participants that began with the same task, only a few correlations had been found out; therefore, the postural comfort depends only on the time evolution.

4 Discussion

Answering the research question, a postural comfort analysis had been done, following the existent methods in literature [34 , 43] in order to identify the critical factors that influenced the postural comfort sitting on the library chair.

A brief evaluation showed the chair was not suitable for students (Fig. 1 & Fig. 2); it means there were already prerogatives to force students moving on the chairs to find a comfortable posture. As a matter of fact, considering the correct sitting posture, it means sitting up straight, leaning arms on the desk, keeping feet on the floor, the chair seems too large to fit an international population (Fig. 3) with medium anthropometric measurements [37]. Indeed, even if the chair was completely close to the desk, due to the height of armrests, the backrest was too far away from the edge of the desk (Figs. 1 and 2). Thus, the students, according to their experience (through the survey spread before the experiments, with 200 answers), were used to change frequently the posture going from the one close to the desk to the one far from it and bending backwards the back to lean on the backrest searching a comfortable posture while studying. Even though this aspect could be considered as “active sitting”, it is, a way to stimulate the movements for having positive effects [16 , 44–46], the negative aspect was the students weren’t able to find a comfortable posture in any moment. Thus, those aspects are all prerogatives of bad design, and experiments have been performed to identify the comfort level for each body part after a study-day. Thus, the postural comfort trend over time, starting from the correct sitting posture close to the desk, had been simulated through the two main tasks of the study (writing and reading). Four 15-minutes tasks had been performed in succession without a long pause to keep the importance of time effect (the experiment lasted 1 hour for each participant, see Table 2). After each 15-minutes task, photos had been taken from 4 views around the chair to acquire postural angles and then participants were asked to fill the Body Perceived Comfort questionnaire rating on 5-point scales.

As far as feedback registered during the experiments, all participants accused pain in the lumbar region because they were unable to lean their back on the backrest to sit properly and to unload the weight of the head and the back.

Results from questionnaires showed a decay over time (Figs. 6–7); it means the chair was not comfortable as expected at the beginning, scoring the lowest value of global comfort in the last task. Thus, the time influenced the perceived comfort, confirmed even by the absence of correlations grouping the data according to the same initial task. Moreover, the thoracic and the lumbar areas were confirmed as critical (Table 4); thus, improvements in these parts are needed: the higher postural comfort in these areas, the more perceived comfort on other body parts because they are strictly correlated. There were no significant differences between expectations and the values of global comfort because several students had previous experience with the chair, and this could have influenced the answers about the expectation.

The software CaMAN had processed postural angles in order to obtain objective postural comfort indexes and, in a second moment, to compare the objective data with subjective data from questionnaires.

Even the results from CaMAN showed comfort decay over time (Fig. 7). The elbows and wrists were not considered according to the constraints of CaMAN because participants used to lay them on the desk during the experiment. Indeed, only body parts without any physical support can be considered.

Correlations between objective and subjective data had been calculated: the global comfort from Questionnaires is always correlated with the Body Part Comfort values; this means each body part influenced the perceived comfort. In particular, the lumbar area presented a strong correlation with the Global Comfort, except for the third task, while, the Thoracic Area presented a significant correlation with the Global Comfort only in the last two tasks; this confirms the influence of those critical parts on the perceived comfort, thus improving the perceived postural comfort in the lumbar area, the overall perceived comfort could arise. There were strong correlations between the global comfort values from Questionnaire and CaMAN only in the first and in the last task. This aspect highlights the importance of the first impact with the chair, in terms of expectations and first approaches [47], and the effect of perceived comfort over time [10 , 48–50], where it is essential to maintain a satisficing postural comfort level over time.

From the perspective of Human-Centered Design, this paper is the basis for the chair redesign in order to improve the postural comfort of students. With postural angles, a virtual environment inside Delmia® had been realized, and the virtual simulation enables to highlight how the postures assumed by students changed over time: starting from the correct sitting posture, they used to assume a slouched one at the end of the experiment. In particular, this is due to the absence of the support for the back.

The chair could be improved by increasing the area of backrest using, for example, a pillow in the lumbar or thoracic region [50, 51]. Otherwise, it could be better to amend the chairs by reducing the depth in order to reduce the gap between the seat pan and the backrest.

Actually, modifications have been already implemented and analyzed with another student project, where a prototype (Fig. 10) has been realized based on the results achieved in this paper. The gap between the seat pan and the seatback (13, 7 cm) has been eliminated realizing a back support in contact with the seat pan. This support involves the thoracic, lumbar and buttock areas. Furthermore, a non-slip coating has been added on the seat pan to prevent slipping, that is, to increase shear forces [52]. Results showed an improvement of comfort perception, especially in the lumbar area, thanks to the presence of physical support.

Fig. 10

Comparison between the chair analyzed in this work (on the left, “before”) and prototype realized by a student project (on the right, “after”). The gap between the seat pan and the seatback (13,7 cm) has been eliminated realizing a back support in contact with the seat pan. Furthermore, a non-slip coating has been added on the seat pan –in Chapter 4.

4.1 Limitations

There are some limitations of CaMAN software to be acknowledged. Firstly, the software considered the participant itself positioned in the space without any support: comfort perception is different in the presence of support. Indeed, if someone bends the upper limbs in the space, without any support, the feeling of comfort is very low; instead, with a presence of support to unload the weight, the comfort perception is higher. Since during the tests, participants laid their forearms on the desk, the comfort perception on this posture was higher than the same posture without the desk; this has been even confirmed by questionnaire results (Fig. 7). Thus, objective comfort indexes of elbow and wrist had been excluded.

Moreover, the resting of lower arms on the desk also could influence the shoulder angles, this aspect has not been considered in this study. Thus, further analysis should be done.

Secondly, when the experiment had been performed, the CaMAN version did not consider the lower limbs. Thus, it was not possible to compare the subjective results of the lumbar area with the objective indexes of lower limbs from CaMAN. Thus, it is recommended to repeat the experiment implementing the evaluation of lower limbs.

5 Conclusion

After a brief investigation among students inside the S&T Library, it had been found out a general physical complaint. Thus, a postural analysis had been performed following a systematic method. Two tasks of typical study-day in library had been simulated: writing and reading. During the experiments, photos had been taken from four different views to detect postural angles by Kinovea®. Those postural angles had been used both to realize a simulation inside the virtual environment of Delmia® and to obtain objective postural indexes by CaMAN.

In summary, this paper argued that:

The comfort perception decreased over time;

The lumbar region scored the lowest value of comfort, thus, this region influences all postural performance, as confirmed by literature studies;

Software CaMAN had been positively used as a tool to obtain objective data of postural comfort because strong correlations between subjective and objective comfort indexes have been detected.

The main goal was to demonstrate through the pos-tural comfort analysis that the chair was few comfortable, so it is necessary to make some modifications, like an extension of the back-support area or a reduction of the seat-pan depth. These renovations can be simulated with Delmia® through a careful analysis, in order to detect the areas quickly to be improved, then to realize a prototype already optimized.

Furthermore, in this work, a method for the definition of comfort indexes has been shown. All the acquisition methods used are very cheap and easy to use. The precision of the acquisition method, as well as the fact of not using complicated, expensive acquisition methods, gave us the possibility to reach a very good numerical/experimental level obtaining relevant results revealed by this paper. The method can be easily reproduced for other applications.

Future improvements could be a developing of CaMAN software in order to include the lower limbs, thus to obtain objective data to compare with all subjective data from questionnaires. Then, other physical parameters can be detected, such as the pressure at the interface between the human and the chair; or a study of physical loads can be carried in order to understand the influence of posture on each human joint.

Conflict of interest

None to report.

Footnotes

Acknowledgments

The research work reported here was made possible by the students that kindly took part to experiment.

References

, Jee

, Yee Lim

. Development of Ergonomics Guidelines for Improved Sitting Postures in the Classroom among Malaysian University Students. American Journal of Applied Sciences. 2016;13:907–912.

Sanjog

, Patel

, Chowdhury

, Karmakar

. Musculoskeletal ailments in Indian injection-molded plastic furniture manufacturing shop-floor: Mediating role of work shift duration. Int J Ind Ergon [Internet]. 2015 Jul 1 [cited 2019 Jan 20];48:89–98. Available from: https://www-sciencedirect-com-443.web.bisu.edu.cn/science/article/abs/pii/S0169814115000554

HIRA

. An ergonomic appraisal of educational desks. Ergonomics [Internet]. 1980;23(3):213–21. Available from: https://doi.org/10.1080/00140138008924735

Asundi

, Odell

, Luce

, Dennerlein

. Changes in pos-ture through the use of simple inclines with notebook com-puters placed on a standard desk. Appl Ergon [Internet]. 2012 Mar 1 [cited 2019 Jan 20];43(2):400–7. Available from: https://www-sciencedirect-com-443.web.bisu.edu.cn/science/article/pii/S000368701100086X

Lis

, Black

, Korn

, Nordin

. Association between sitting and occupational LBP. Eur Spine J [Internet]. 2007;16(2):286–98. Available from: https://link-springer-com-443.web.bisu.edu.cn/article/10.1007%2Fs00586-006-0143-7#citeas

Muhammad Hussain

. Awareness of Good Posture and Computer Ergonomics among Medical Students of ISRA University. Vol. 2, International Journal of Physiotherapy. 2015.

Netten

, Van Der Doelen

LHM

, Goossens

RHM

. Chair Based Measurements of Sitting Behavior a Field of Sitting Postures and Sitting Time in Office Work. Int Conf Digit Hum Model Appl Heal Safety, Ergonimics Risk Manag. 2013;8026:261–8.

Castellucci

, Arezes

, Molenbroek

JFM

. Equations for defining the mismatch between students and school furniture: A systematic review. Int J Ind Ergon [Internet]. 2015 Jul 1 [cited 2019 Jan 20];48:117–26. Available from: https://www-sciencedirect-com-443.web.bisu.edu.cn/science/article/abs/pii/S0169814115000670

Vieira

, Buckeridge Serra

MVG

, Brentini de Almeida

, Vieira Villela

, Domingos Scalon

, Veiga Quemelo

. Symptoms and risks for musculoskeletal disorders among male and female footwear industry workers. Int J Ind Ergon [Internet]. 2015 Jul 1 [cited 2019 Jan 20];48:110–6. Available from: https://www-sciencedirect-com-443.web.bisu.edu.cn/science/article/abs/pii/S0169814115000669

10.

Vink

. Vehicle Seat comfort and design. 1st ed. 2016.

11.

Helander

, Little

, Drury

. Adaptation and Sensitivity to Postural Change in Sitting. Hum Factors [Internet]. 2000;42(4):617–29. Available from: https://doi.org/10.1518/001872000779698051

12.

Goossens

RHM

, Teeuw

, Snijders

. Sensitivity for pressure difference on the ischial tuberosity. Ergonomics [Internet]. 2005;48(7):895–902. Available from: https://doi.org/10.1080/00140130500123647

13.

Groenesteijn

, Hiemstra-van Mastrigt

, Gallais

, Blok

, Kuijt-Evers

, Vink

. Activities, postures and comfort perception of train passengers as input for train seat design. Ergonomics. 2014;57:1–12.

14.

Hiemstra-van Mastrigt

, Kamp

, van Veen

SAT

, Vink

, Bosch

. The influence of active seating on car passengers’ perceived comfort and activity levels. Appl Ergon. 2015;47:211–9.

15.

Kamp

, Kilincsoy

, Vink

. Chosen postures during specific sitting activities. Ergonomics [Internet]. 2011;54(11):1029–42. Available from: https://doi.org/10.1080/00140139.2011.618230

16.

Van Veen

, Orlinskiy

, Franz

, Vink

. Investigating Car Passenger Well-Being Related to a Seat Imposing Continuous Posture Variation [Internet]. Journal of Ergonomics. OMICS International. 2015;5:1–8. Available from: https://www.omicsonline.org/open-access/investigating-car-passenger-wellbeing-related-to-a-seat-imposingcontinuous-posture-variation-2165-7556-1000140.php?aid=62373

17.

Vink

, Brauer

. Aircraft interior comfort and design. 2011.

18.

Delleman

. Working Postures: prediction and evaluation [Internet]. TU Delft; 1999. Available from: https://repository.tudelft.nl/view/tno/uuid%3A17474ef9-1efd-4d9e-8bfd-c82ad50d6635

19.

Fagarasanu

, Kumar

, Narayan

. Measurement of angular wrist neutral zone and forearm muscle activity. Clin Biomech [Internet]. 2004 Aug 1 [cited 2019 Jan 31];19(7):671–7. Available from: https://www-sciencedirect-com-443.web.bisu.edu.cn/science/article/pii/S0268003304000919?via%3Dihub

20.

Cappetti

, Naddeo

, Soldovieri

, Vitillo

. A study on the correlation between the perceived comfort and the muscular activity, using virtual simulation techniques. In: Proceedings of the 1st International Comfort Congress (ICCSalerno [Internet]. 2017;1:223–31. Available from: http://hdl.handle.net/11386/4706654

21.

Naddeo

, Cappetti

, D’Oria

. Proposal of a new quantitative method for postural comfort evaluation. Int J Ind Ergon [Internet]. 2015;48:25–35. Available from: https://www-sciencedirect-com.web.bisu.edu.cn/science/article/pii/S0169814115000505

22.

Pearson

. Comfort and its measurement –A literature review. Disabil Rehabil Assist Technol. 2009;4:301–10.

23.

Kölsch

, Beall

, Turk

. An Objective Measure for Postural Comfort. Proc Hum Factors Ergon Soc Annu Meet [Internet]. 2003;47(4):725–8. Available from: https://doi.org/10.1177/154193120304700413

24.

Joshi

, Kale

, Chandel

, Pal

. Likert Scale: Explored and Explained. Br J Appl Sci Technol. 2015;7:396–403.

25.

Zhang

, Helander

, Drury

. Identifying Factors of Comfort and Discomfort in Sitting. Hum Factors [Internet]. 1996;38(3):377–89. Available from: https://doi.org/10.1518/001872096778701962

26.

Helander

, Zhang

. Field studies of comfort and discomfort in sitting. Ergonomics. 1997;40:895–915.

27.

Smulders

, Berghman

, Koenraads

, Kane

, Krishna

, Carter

, et al.Comfort and pressure distribution in a human contour shaped aircraft seat (developed with 3D scans of the human body). Work. 2016;54:1–16.

28.

Califano

, Naddeo

, Vink

. The effect of human-mattress interface’s temperature on perceived thermal comfort. Appl Ergon. 2017;58.

29.

Naddeo

, Cappetti

, Vallone

, Califano

. New trend line of research about comfort evaluation: proposal of a framework for weighing and evaluating contributes coming from cognitive, postural and physiologic comfort perceptions. In:Vink P, editor. Advances in Social and Organizational Factors. Cracovia; 2014. pp. 503-15.

30.

Trapanese

, Naddeo

, Cappetti

. A Preventive Evaluation of Perceived Postural Comfort in Car-Cockpit Design: Differences between the Postural Approach and the Accurate Muscular Simulation under Different Load Conditions in the Case of Steering-Wheel Usage. SAE Tech Pap. 2016;2016-April(April).

31.

Apostolico

, Cappetti

, D’Oria

, Naddeo

, Sestri

. Postural comfort evaluation: experimental identification of Range of Rest Posture for human articular joints. Int J Interact Des Manuf [Internet]. 2014;8(2):109–20. Available from: https://doi.org/10.1007/s12008-013-0186-z

32.

Pagliara

. Serie Fisciano [Internet]. Available from: https://nicolapagliara.wordpress.com/design/serie-fisciano/

33.

The Science and Technology Library of UNISA [Internet]. Available from: https://www.biblioteche.unisa.it/biblioteca_scientifica/biblioteca scientifica

34.

Commentale

, Naddeo

, Contrada

, Forlone

, Saturno

. Comfort and ergonomics evaluation of a checkout workstation. ARPN J Eng Appl Sci. 2018;13(13):4117–25.

35.

, Yu

, Yang

, Pei

, Zhao

. Effects of long-duration sitting with limited space on discomfort, body flexibility, and surface pressure. Int J Ind Ergon [Internet]. 2017 Mar 1 [cited 2019 Jan 20];58:12–24. Available from: https://www-sciencedirect-com-443.web.bisu.edu.cn/science/article/abs/pii/S0169814117300288

36.

Piro

, Fiorillo

, Anjani

, Smulders

, Naddeo

, Vink

. Towards Comfortable Communication in Future Vehicles. Appl Ergon. 2018;

37.

DINED. Available from: https://dined.io.tudelft.nl/en

38.

Naddeo

, Cappetti

, Califano

, Vallone

. The Role of Expectation in Comfort Perception: The Mattresses’ Evaluation Experience. Procedia Manuf [Internet]. 2015 Jan 1 [cited 2019 Jan 21];3:4784–91. Available from: https://www-sciencedirect-com-443.web.bisu.edu.cn/science/article/pii/S2351978915005831

39.

Corlett

, Bishop

. A technique for assessing postural discomfort. Ergonomics. 1976;19(2):175–82.

40.

Groves

, Fowler

, Couper

, Lepkowski

, Singer

, Tourangeau

. Survey Methodology. 2004;561.

41.

Groenesteijn

. Seat design in the context of knowledge work. TU Delft; 2015.

42.

Piro

, Fiorillo

, Anjani

, Smulders

, Naddeo

, Vink

. Towards comfortable communication in future vehicles. Appl Ergon [Internet]. 2019;78:210–6. Available from: https://www-sciencedirect-com.web.bisu.edu.cn/science/article/pii/S0003687019300602

43.

Naddeo

, Memoli

. Postural comfort inside a car: Development of an innovative model to evaluate the discomfort level. Vol. 2, SAE International Journal of Passenger Cars -Mechanical Systems. 2009.

44.

Vink

, Vink

, Lelie

C Van Der

. The sweetness of discomfort Designing the journey. In 2014.

45.

Fasulo

, Naddeo

, Cappetti

. A study of classroom seat (dis)comfort: Relationships between body movements, center of pressure on the seat, and lower limbs’ sensations. Appl Ergon. 2019;74:233–40.

46.

Naddeo

, Califano

, Vink

. The effect of posture, pressure and load distribution on (dis)comfort perceived by students seated on school chairs. Int J Interact Des Manuf. 2018;

47.

Naddeo

, Cappetti

, Califano

, Vallone

. The Role of Expectation in Comfort Perception: The Mattresses’ Evaluation Experience. Procedia Manuf. 2015;3:4784–91.

48.

Naddeo

, Califano

, Vallone

, Cicalese

, Coccaro

, Marcone

, et al. The effect of spine discomfort on the overall postural (dis)comfort. Appl Ergon. 2019;74.

49.

Vergara

, Page

. Relationship between comfort and back posture and mobility in sitting-posture. Appl Ergon [Internet]. 2002 Jan 1 [cited 2018 Sep 8];33(1):1–8. Available from: https://www-sciencedirect-com-443.web.bisu.edu.cn/science/article/pii/S0003687001000564?via%3Dihub

50.

Naddeo

, Di Brigida

, Fontana

, Montese

, Quartuccia

, Nasti

, et al. A body-shaped lumbar-sacral support for improving car-seat comfort. Work. 2019;

51.

Carvalho

DE De

, Callaghan

. Spine Posture and Discomfort During Prolonged Simulated Driving With Self-Selected Lumbar Support Prominence. Hum Factors [Internet]. 2015;57(6):976–87. Available from: https://doi.org/10.1177/0018720815584866

52.

van Geffen

, Reenalda

, Veltink

, Koopman

HFJM

. Decoupled pelvis adjustment to induce lumbar motion: A technique that controls low back load in sitting. Int J Ind Ergon. 2010;40(1):47–54.

	Male (n = 8)			Female (n = 7)
	Mean	SD	Range	Mean	SD	Range
Height (cm)	178.5	6.2	168 –185	162.7	5.7	154 –169
Weight (kg)	72.5	10.2	57 –90	57.1	4.8	50 –63

A method for postural critical factors checking: The case of library chairs

Abstract

BACKGROUND:

OBJECTIVE:

METHODS:

RESULTS:

CONCLUSIONS:

Keywords

1 Introduction

2 Materials and method

2.1 Experiment setup

Table 1 Anthropometrics data of participants –in Chapter 2.2 Male (n = 8) Female (n = 7) Mean SD Range Mean SD Range Height (cm) 178.5 6.2 168 –185 162.7 5.7 154 –169 Weight (kg) 72.5 10.2 57 –90 57.1 4.8 50 –63

Table 2 Protocol regarding time –in Chapter 2.5 and Chapter 4 Task 1 Task 2 Task 3 Task 4 Participant A Reading Writing Reading Writing Participant B Writing Reading Writing Reading 15 min 15 min 15 min 15 min 1 min 1 min 1 min

Table 3 Conversion scale to compare questionnaire scale with the CaMAN one – in Chapter 2.7 Original CaMAN Scale 1 2 3 4 5 6 7 8 9 10 Translated CaMAN Scale 1 2 3 4 5 Questionnaire Scale 1 2 3 4 5

3 Results

5 Conclusion

Conflict of interest

Footnotes

Acknowledgments

References

Table 1
Anthropometrics data of participants –in Chapter 2.2

Male (n = 8) Female (n = 7)

Mean SD Range Mean SD Range

Height (cm) 178.5 6.2 168 –185 162.7 5.7 154 –169

Weight (kg) 72.5 10.2 57 –90 57.1 4.8 50 –63

Table 2
Protocol regarding time –in Chapter 2.5 and Chapter 4

Task 1 Task 2 Task 3 Task 4

Participant A Reading Writing Reading Writing

Participant B Writing Reading Writing Reading

15 min 15 min 15 min 15 min

1 min 1 min 1 min

Table 3
Conversion scale to compare questionnaire scale with the CaMAN one – in Chapter 2.7

Original CaMAN Scale 1 2 3 4 5 6 7 8 9 10

Translated CaMAN Scale 1 2 3 4 5

Questionnaire Scale 1 2 3 4 5