A novel bamboo sheet chair and its influence on sitting comfort

Participants

Twenty-five participants (14 males and 11 females) volunteered for this study. The overall mean ± SD of male participants for age, mass and stature were 21.71  ± 1.98 years, 66.49  ± 8.23 kg and 173.00  ± 4.74 cm, respectively. Female participants’ mean age, mass and stature were 21.36  ± 2.01 years, 53.83  ± 4.07 kg and 161.74  ± 3.73 cm, respectively. We divided the 14 male subjects into three groups and the 11 female subjects into three groups according to their body characteristics using IBM SPSS statistics software (version 20.0; IBM Corp., USA) to conduct cluster analysis.

All the participants were healthy with no musculoskeletal injuries in the past 12 months. The participants were asked to refrain from strenuous exercise, caffeine and alcohol in the 24 h period prior to each testing session. All participants provided consent in compliance with the Declaration of Helsinki (ver. 7.0; October 2013) (World Medical Association, 2013). All protocols were approved by the Anhui Agricultural University’s ethical committee.

Manufacturing

A novel sheet chair was constructed from natural bamboo. First, we cut the rectangular bamboo sheets with a thickness of 5 mm. These sheets underwent six cycles of anti-corrosion, anti-insect, and anti-moth processing, as well as drying, carbonization, and assembly surface decoration. According to the requirements, the seat pan and backrest of the chair were stacked with three layers of bamboo sheets, and the bamboo sheets were fixed using slots. The bamboo sheets in each layer could be moved up and down for vertical elastic support. The structural design is shown in Fig. 1.

Experimental design and procedure

The office chairs (Model:KL-702) used in the experiment were purchased from Anhui JinRuiXin Office Furniture Co. Ltd., China. We designed both the bamboo sheet chairs and wooden chairs, which were commissioned by the cooperative enterprise, Anhui Longhua Bamboo Product Co. Ltd., China.

The study evaluated three chairs: an office chair, a self-made bamboo sheet chair, and a wooden chair (Fig. 2). The dimensions of the three chairs are listed in Table 1, and the measurement indices are shown in Fig. 3A. We measured body characteristic parameters included stature, weight, shoulder breadth, hip breadth, waist width, popliteal height, buttocks-popliteal length, and buttock-abdomen depth. Detailed measurement methods are shown in Fig. 3B (Pheasant, 2005). We divided the male subjects into three groups and the female subjects into three groups according to their body characteristics using cluster analysis in the IBM SPSS statistics software (version 20.0; IBM Corp., USA).

Before reporting comfort scores on the “Subjective evaluation for sitting comfort of seven-degree scale” questionnaire, the subjects were not informed of the objective of the test. The subjective sitting comfort evaluation was conducted by combining the semantic differentiation method with the free modulus amplitude estimation method (Zhang, Jiang & Wang, 2004). Semantic differentiation is expressed in the form of a seven-level scale that mainly evaluates the comfort of various body parts (Guo, Sheng & Chen, 2011). The subjective method was based on the participants’ feelings of comfort or discomfort while sitting on chairs during a given period of time Subjects could freely adjust their sitting positions to determine whether they are comfortable or to identify discomfort in different seats for comfort comparison. The questionnaire consisted of seven rating categories that asked the participants to evaluate the intensity of pressure and level of comfort or discomfort while sitting. Corresponding numerical values were then circled by the participants. The presented categories are listed in Table 2. The chairs’ identities were deliberately hidden, so that the participants could not be influenced by any predisposed bias/opinion. Although the chairs were covered, the office chair’s arm rests were an identifiable feature; this is a shortcoming of the study design.

Each subject completed a body pressure test for each of the three different chairs. During the test, the participants were asked to naturally depend on the backrest, and keep their eyes straight ahead, thighs were roughly parallel, knees bent at approximately 90°, feet flat on the floor, shoulders relaxed, and hands gently resting on the thighs (Chen et al., 2009). The CONFOR Mat Pressure Measurement System (BPMS, 5400D, Tekscan Inc., USA) was used to evaluate the pressure distribution between the chair and the subject. During the test, the pressure sensing mat was placed between the subject, seat pan and backrest. After the pressure distribution became relatively stable, the data was recorded for 3 min with a sampling rate of 8 f s⁻¹. The subjects were required to sit in the test chair according to the specified sitting posture during the duration of the body pressure test. Participants were not allowed to talk during the body pressure test.

Data processing and reduction

Pressure data were converted into a pressure distribution matrix ASCII file. Considering the pressure matrix as the basis of calculations and statistical analysis, the pressure distribution index values for the seat pan and backrest were calculated using MATLAB 2015b, which mainly included maximum pressure (P_m), average pressure (P_a), pressure exponent (P_e) and contact area index (P_AI).

(1) Maximum pressure (P_m)

The maximum pressure was defined as the pressure at the 95th percentile of all test points, and is described in Eq. (1).

(1)${\displaystyle P_m={\overline{P}}_N+\left(SD\times K\right)}$

where N is the number of measurement points, ${\overline{P}}_N$ isthe average pressure of all the measurement points, SD is the standard deviation of all test points, and K = 1.64. Furthermore, ${\overline{P}}_N$ and SD are outlined Equations in (2) and (3), respectively.

(2)${\displaystyle {\overline{P}}_N=\frac{1}{N}\sum _{i=1}^N{P}_i}$ (3)${\displaystyle SD=\sqrt{\frac{1}{N-1}\left(\sum _{i=1}^N{P}_i^2-N{\overline{P}}_N^2\right)}}$

where N is the number of test points, and P_i is the pressure of each test point.

(2) Average pressure (P_a)

Average pressure was defined as the 50th percentile pressure value of all test points, which is the arithmetic mean of all test points, and is given by Eq. (4). (4)${\displaystyle P_a=\frac{1}{N}\sum _{i=1}^N{P}_i}$

where N is the number of test points, and P_i is the pressure of each test point.

(3) Pressure exponent (P_e)

Pressure exponent was calculated using Eq. (5). (5)${\displaystyle P_e=\sqrt{{\left(P_{M10}-10\right)}^2+S{D}_{10}^2}}$

where P_M10 is the average value of the contact surface pressure greater than 1.33 kPa, and SD₁₀ is the standard deviation of the contact surface pressure greater than 1.33 kPa (Shelton, Barnett & Meyer, 1999). Additionally, P_M10 and SD₁₀ are described by Eqs. (6) and (7), respectively.

(6)${\displaystyle P_{M10}=\frac{1}{N}\sum _{i=1}^{N_{10}}{P}_i}$ (7)${\displaystyle S{D}_{10}=\sqrt{\frac{1}{N_{10}-1}\left(\sum _{i=1}^{N_{10}}{P}_i^2-N_{10}{P}_{M10}^2\right)}}$

where P_i ≥1.33 kPa, and N₁₀ is the number of pressure test points greater than 1.33 kPa.

(4) Contact area index (P_AI)

Contact area index was defined as the “pressure contact area within a certain threshold pressure limit range”. Under normal circumstances, the capillary pressure of the arterioles is 3.3–4.6 kPa, the pressure in the venules is approximately1.6 kPa, the critical pressure is considered to be 4 kPa (Forciea, Humphrey & Qaseem, 2015; Keller, 2006). Therefore, three pressure threshold limits of 0.67 kPa, 4.00 kPa and 9.33 kPa and four contact surface indexes were used in the experiment to reflect the contact area between human and chair. We determined that a human-chair interface pressure between 0.67–4 kPa was a low-pressure threshold value, 4.00–9.33 kPa was a middle-to-high pressure threshold value, and 9.33 kPa and above was a high-pressure threshold value.

Statistic analysis

Statistical analysis was performed by using IBM SPSS statistics software (version 20.0; IBM Corp., USA). Variance test analysis was applied to the body characteristics we measured, the objective pressure indexes, and the subjective comfort evaluations. The variance of the pairwise comparisons between the various indicators is also listed in Tables 3 and 4.

Pressure indexes and subjective comfort evaluation of the three chairs

Figure 4 shows a representative body pressure distribution of the human-chair interface. Table 3 lists the three chairs’ seat pan and backrest data, including their contact areas; the contact areas within the low, middle-to-high, and high-pressure threshold ranges; pressure exponents; maximum pressure; and average pressure. The total backrest contact areas of the office chair (314.96 ± 118.62 cm²) and bamboo sheet chair (325.55 ± 73.99 cm²) were not significantly different, while the backrest contact area of the office chair in the low threshold pressure range (192.81 ± 89.49 cm²) was lower than that of the bamboo sheet chair (234.22 ± 55.80 cm², p = 0.047). In addition, the contact areas in the backrest (52.96 ± 32.94 cm²) and seat pan (307.75 ± 90.31 cm²) in the middle-to-high threshold pressure range, and the contact areas of the backrest (4.34 ± 5.95 cm²) in the high threshold pressure range of bamboo sheet chair were smaller than the corresponding indexes of the office chair (81.430 ± 45.04 cm², p = 0.00; 394.39 ± 98.99 cm², p = 0.02; 13.54 ± 12.00 cm², p = 0.00, respectively).

The pressure exponent, maximum pressure, and average pressure (2.68 ± 0.88 kPa, 6.66 ± 2.05 kPa, and 2.42 ± 0.59 kPa, respectively) of the bamboo sheet chair’s backrest were significantly lower than the corresponding indexes of the office chair (4.32 ± 1.62 kPa, p = 0.00; 10.50 ± 3.88 kPa, p = 0.00; 3.43 ± 0.97 kPa, p = 0.00, respectively). The maximum pressure and average pressure (26.26 ± 4.2 kPa, and 7.77 ± 1.30 kPa, respectively) of the seat pan were significantly lower than the corresponding indexes of the wooden chair (28.91 ± 4.90kPa, p = 0.00; 8.33 ± 1.41 kPa, p = 0.00, respectively).

Comfort evaluations for the three different chairs are listed in Table 4. Except for stability and lateral thigh pressure, most of the indices showed significant differences. The office and bamboo chair evaluation results were similar in most indexes, but there was a significant difference between their results and those of the wooden chair. The elastic feeling of the bamboo chair (2.28 ± 0.81) was lower than the wooden chair (4.76 ± 0.43, p = 0.00), and even lower than the office chair (2.76 ± 0.66, p = 0.01). The overall comfort index of the bamboo sheet chair (2.20 ± 0.49) was also lower than both the office (2.76 ± 0.99, p = 0.02) and wooden chairs (4.56 ± 0.64, p = 0.00).

Subject characteristic parameters and pressure index

Figure 5 shows the cluster analysis used to divide the 14 male subjects into three groups and the 11 female subjects into three groups according to their body characteristics, represented by male groups 1, 2, 3 and female groups 1, 2, 3, respectively. The mean value and standard deviation for each group is listed in Table 5. Significant differences existed among most of the physical characteristic groups. In order to further analyze the relationship between the bamboo sheet chair and body characteristics, Table 6 lists the total contact area and the contact areas in low, middle-to-high and high-pressure threshold ranges. The human-chair interface pressure exponents, maximum pressure, and average pressure are shown in Fig. 6.

The results presented in Table 6 show that the backrest St and Sl for female subjects decreased with shorter statures and buttocks-popliteal lengths, while the seat pan St decreased with lower weight and hip breadths. The seat pan Sh for female group 2 (246.11 cm²) was larger than that of female group 1 (204.44 cm², p = 0.02) and female group 3 (167.11 cm², p = 0.00). The backrest Sh of male group 3 was 0 cm², while the seat pan Sh was 376.46 cm², which was larger than that of the male group 1 (209.07 cm², P = 0.00) and male group 2 (250.31 cm², P = 0.00).

For subjects with similar weights (differences within 2 kg), such as those in female groups 1 and 3, the seat pan pressure exponent and average pressure decreased with taller statures and longer buttocks-popliteal lengths, while the backrest pressure exponent and average pressure increased with greater statures and buttocks-popliteal lengths. The average seat pan pressure was greater than 4 kPa for all groups, and the average seat pan pressure for male group 3 was greater than 9.33 kPa.