A novel modeling and simulation approach for the prediction of human thermophysiological comfort

Abstract

Thermophysiological comfort is one of the most important aspects of wear comfort. Currently, there are no software solutions available for the combined consideration of material physical and mechanical characteristics, fit, and thermophysiological behavior. Thus, a laborious empirical process is typically required to determine an appropriate design matching new textile materials to pattern cuts as well as changing climatic conditions. A detailed wear trial in a climatic chamber supports this process. The objective of this research is to analyze the thermal comfort of clothing with different thermal characteristics through the simulation of heat regulation in the human body, microclimate, clothing, and environment.

Keywords

three-dimensional body scanning fitting simulation human model thermal simulation thermal comfort

Thermal comfort has always been a crucial factor in garment design; it can be defined as the state of mind that expresses satisfaction with the thermal environment and is assessed by subjective evaluation.¹ Human beings face different environmental conditions throughout their daily routine. Since the environmental parameters (ambient temperature, humidity, and air velocity) that affect thermal comfort are continuously changing, clothing plays a very important role within this scenario of ever-changing external factors.

The human body always tries to maintain its core body temperature within a narrow temperature range of 37 ± 0.5℃, whereby this temperature is achieved by balancing internal heat production and heat loss to the environment. Due to its metabolism, the human body continuously produces heat. The metabolic rate is dependent on the activity level of the human. When resting, energy is only required for basic body functions, that is, respiration and heart functions providing oxygen and nutrients to body cells. However, during phases of activity, body muscles need more oxygen and nutrients to perform mechanical work, thus resulting in an increase in the metabolic rate. Excess energy that is not consumed by muscles is then liberated outside the body by conduction, convection, radiation, evaporation, and respiration. The heat balance equation can be written as²
$S = (M - W) - (E + C + R + B)$

Here, heat storage (S) is the difference between heat production (metabolism M corrected for work W) and heat loss (due to evaporation E, convection and conduction C, radiation R, and respiration B). A body achieves heat balance once the rate of heat storage is zero (S = 0). If heat storage increases continuously, the rate of heat production is greater than the rate of heat loss. As the human body has a limited thermal storage capacity, the core body temperature increases as a result. In contrast, if the rate of heat loss is greater than the rate of heat production, the core body temperature decreases. Therefore, the human thermoregulation system has thermal sensors and diverse actuators, such as vasomotion (vasodilation and vasoconstriction), sweating, and shivering thermogenesis, to maintain the human body’s thermal equilibrium.³ With respect to increasing core temperature, vasodilation and sweating help the human body to release heat efficiently. However, if the core temperature decreases, vasoconstriction and shivering processes are activated to preserve heat within the body or to generate heat to maintain the core body temperature.²

Clothing plays a key role in achieving or maintaining the heat balance of the human body with its environment, because it significantly affects heat and mass transfer between the body and the environment. Researchers generally agree that air permeability, thermal insulation, and water vapor resistance of clothing are the most decisive factors in terms of thermal comfort.^4,5 Thermal comfort⁶ is modulated by the yarn and fabric structure, the thickness and number of fabric layers, the drape behavior of fabrics, and most importantly, air gaps between the clothing layers and the body.⁷

A microclimate is developed in the air gaps between the clothing and the human body. Due to its proximity of being immediately adjacent to the human body, the comfort/discomfort levels depend very much on the characteristics of this microclimate. Microclimate characteristics are influenced by numerous factors, for example, the metabolic rate of the human body, environmental conditions, material thermal properties, product design, and drapability of the worn clothing.^8,9 In the past, many experimental techniques have been used to study the air gap distribution and its effect on heat transfer across clothing. Results revealed that clothing reduces thermal conduction by trapping air within the clothing structure and between the body and the clothing layers.¹⁰ Other researchers came to the conclusion that garment fit and the resultant air volume play very important roles in characterizing the thermal properties of any clothing system.^11–13 An experimental study has been conducted to investigate the effects of heterogeneous and homogeneous air gaps on dry heat loss through the garment.¹⁴ The results have shown that the folds in garments cause more heat loss from the body as compared to a homogeneous air gap. It has also been experimentally concluded that with the increases in air gap thickness, the relative water vapor permeability of the air gap decreases and evaporative resistance increases.¹⁵ However, this research was limited only for 0, 2, and 4 mm air gap thickness. The three-dimensional (3D) body scanning method has been found by several researchers to be the most accurate tool in determining the air gap volume^16–19 and is therefore widely used in this context. To measure the thickness/volume of the air gap, 3D scans of the nude and dressed body were aligned, and the air gap thickness was calculated.^17,20,21 However, this process may lead to inaccuracies due to post-processing discrepancies, especially during cleaning and closing the holes of scanned surfaces as well as superimposing scanned nude and dressed human models. To avoid the non-ideal post-processing of scanned data, the use of 3D fit simulation software is presented as another option for air gap analysis.^8,22,23 Importantly, the air layer underneath the clothing is not uniform, but it was assumed to be either uniform or non-existent in many published clothing models.^24–27 Therefore, the assumption of a uniform air layer during the prediction of thermal clothing comfort might lead to errors in results.

The thermal comfort of clothing can be evaluated by objective and subjective measuring methods that are very demanding and time-consuming. In the past few decades, researchers have been trying to replace these empirical methods by reliable simulation methods so that numerous thermophysiological human models could be developed. The most significant and established models include a simple human model by Gagge et al.,²⁸ a thermal comfort model by Fanger,²⁹ a detailed human model by Wissler,³⁰ a popular human model by Stolwijk et al.³¹ developed for NASA, a reliable and significant human model by Fiala et al.,^32,33 and a human model by Tanabe.³⁴ Since the outputs of these models are the core body and skin temperatures, which are the main parameters for the evaluation of thermal comfort and sensation, all thermal comfort models are associated with any human thermophysiological model.³⁵ These human models have been successfully applied in simulation and modeling to predict realistic situations in different areas and industries, for example for the development of the universal climate index (UTCI)³⁶ or the prediction of thermophysiological responses of anesthetic patients during open-heart surgery,³⁷ the development of thermophysiological human simulators for clothing research by coupling the thermophysiological model with a thermal manikin,^3,38,39 and the prediction of human thermal responses for building climatology.^40,41

Despite all these developments, the application of thermophysiological models to the clothing industry, as part of the methodology for predicting thermal comfort, is still proving to be a challenge for the scientific community. This is related to the fact that the thermal simulation of the human body–clothing–environment is a very complex phenomenon and is influenced by many physical, physiological, and non-physical parameters,³⁵ which are not fully defined nor considered for every thermophysiological model or simulation platform (software) due to their limitations. Consequently, the clothing industry does not fully rely on these models or simulation techniques during their product development process. However, there is a great opportunity for scientists to find solutions that can fully replace the empirical, time-consuming development process with an advanced, fast, and detailed simulation technique.

The main objective of this research is the application of simulation techniques to generate a comprehensive solution for evaluating the thermal comfort of different textile materials, while considering climatic conditions, non-uniform air gaps, fit, and the mechanical and thermal properties of clothing. This simulation process offers enormous potential for sectors involving outdoor, work, and safety clothing as well as sportswear.

Methodology

The presented study evaluated the thermal comfort of clothing constructed from different textile materials; simple clothing ensembles, that is, a long sleeve shirt and trekking trousers, were selected for this research. In case A, materials F1 and F2 were used for the long sleeve shirt and the trekking trousers, respectively. In case B, materials F3 and F4 were chosen for the long sleeve shirt and the trekking trousers, respectively. All textile materials selected for the clothing are shown in Table 1.
Table 1.
Selection of textile materials for clothing

Case Clothing Fabric code Fabric type Specifications Mass per unit area [g/m²]

A Long sleeve shirt F1 Knitted 98% Rec. PES/2% PES 161

Trekking trousers F2 Woven 100% PA 125

B Long sleeve shirt F3 Knitted 100% PES hybrid yarn 92

Trekking trousers F4 Fleece 100% PES 168

PES: polyester; PA: polyamide.

The selected materials were comprehensively characterized. The physical, mechanical, and thermal properties measured are listed in Table 2. A Zwick tensile strength tester was used for measuring the stress–strain behavior. The results obtained were prepared according to the requirements of the fit simulation software; in this research project, Modaris V8⁴² was used. This software allows a realistic clothing simulation by defining the material characteristics based on two different fabric evaluation systems: the Fabric Assurance by Simple Testing (FAST) system⁴³ and the Kawabata Evaluation System (KES).⁴⁴ Therefore, fabric elongations were calculated according to the FAST system against three different loads of 5, 20, and 100 N/m in both warp (wale) and weft (course) directions as well as diagonally (at an angle of 45°) against the load of 5 N/m.
Table 2.
Characterization of the materials

Characteristics Instruments Textile material

F1 F2 F3 F4

Physical

Fabric thickness [mm] Karl Schröder KG Material Testing Machine 0.46 0.29 0.41 1.55

Average mass per unit area of the fabrics [g m^–2] GSM Cutter, weighing balance 161 125 92 168

Thermal

Water vapor resistance [m² Pa W^–1] Sweat guarded hot plate 1.69 3.18 2.98 5.57

Thermal resistance [m² K W^–1] Sweat guarded hot plate 0.005 0.009 0.024 0.063

Specific heat [Jkg^–1K^–1] Calculated 493 295 170 1640

Mechanical

Bending stiffness [µN m] (warp, weft, 45°) Bending stiffness tester Cantilever ACPM 200 1.18 0.74 0.63 3.22 10.32 4.82 1.58 1.28 0.42 2.87 2.88 2.91

Elongation [%]

5 N/m Zwick Tensile Strength Tester

Warp 1.15 0.42 2.81 0.42

Weft 2.56 0.05 4.11 1.27

20 N/m

Warp 4.34 1.09 7.61 2

Weft 10 0.21 14 6.4

100 N/m

Warp 10.88 2.51 19.48 7.17

Weft 19 0.98 50.7 28.2

Shear stiffness [N/m] 40.46 124.24 23.21 57.48

A test person was selected and scanned with the help of a 3D body scanner, VITUS (3D body scanning of the test person was completed by the Hohenstein Institut für Textilinnovation GmbH)⁴⁵ to conduct further simulation processes. As a consequence of scanning, a considerable amount of point cloud data was collected, which was further used to produce the polygon model by triangulation. It is a fact that the scanning data always requires post-processing processes of refining and repairing the scanned surface in order to be suitable for further tasks. Thus, the software Geomagic Studio 10⁴⁶ was used for this purpose and was used to prepare the test person virtual model for the further fit and thermal simulation. Figure 1(a) represents the polygon model of the test person. The surface reconstruction of the polygon model was carried out to develop a non-uniform rational B-splines (NURBS) model (Figure 1(b)), which is defined by combinations of vertices, node vectors, and transition conditions. The weight factors represent an additional design element to describe the complex body surface in a realistic manner. In modeling the geometric surface, NURBS is an industry-standard tool⁴⁷ that is widely used for parametric representations.
Figure 1.
(a) Polygon model and (b) non-uniform rational B-splines model of the test person.

Body measurements were taken from the scanned data according to the norms ISO 8559 and DIN EN ISO 7250. With the help of the software Modaris V8, these measurements were used to develop the two-dimensional (2D) patterns (Figure 2) for the trekking trousers and the long sleeve shirt.
Figure 2.
Two-dimensional patterns for (a) trekking trousers and (b) the long sleeve shirt.

The 3D fit simulation of these garments was carried out by means of the software Modaris V8. To investigate the fabric draping properties during fit simulation, a material database was generated for each fabric selected according to FAST by entering various material properties into the software, including elongation, bending and shearing stiffness, mass per unit area, and fabric thickness. Thus, the fit simulations were performed according to the selected materials for two different cases (A and B, as mentioned in Table 1). Figure 3 illustrates the process chain of the 3D fit simulation.
Figure 3.
Process chain of three-dimensional fit simulation (defining of seams, assigning material properties, positioning of pattern cuts, and simulation results).

As the air gaps between the clothing and the body depend on both the draping properties of the fabric and the body shape, the thickness of air gaps was analyzed for both cases A and B (Table 1) using Geomagic software.⁴⁶ For this purpose, the results of the 3D fit simulation were imported into Geomagic, from which the air gap thicknesses were examined. Figure 4 shows the small variations in the thickness of the air gaps that occur at the upper and lower parts of the body for different fabrics F1 and F3 and F2 and F4, respectively. These variations are expected to increase with loose-fitting clothing, which can be observed in the cross-sectional views (CS-1, CS-2, CS-3, and CS-4) of the clothing and the body. As the clothing on the lower part of the body is more loosely fitted compared to the upper body, more variations can be observed in CS-3 and CS-4 compared to CS-1 and CS-2. The following reasons might cause these variations:
the draping properties (bending and tensile stiffness, shear rigidity, mass per unit area, thickness of the fabric) of the materials are not identical;

the standing posture of the right- and the left-hand sides of the body typically differ slightly.

Figure 4.
Visualization of air gaps between the clothing and the body regarding different materials.

These air gaps influence the thermoregulation of the human body, therefore also affecting thermal comfort.

The thermal simulation of the system body–clothing–environment was conducted by use of the software Theseus-FE.⁴⁸ The 3D scanned models that were imported into Theseus-FE represent only the anatomical shape of a test person. However, for thermal simulations, it is essential that a model can imitate human body reactions and perform thermophysiological responses for changing parameters directly affecting thermoregulation, for example environmental conditions, clothing properties, and activity levels. Theseus FE allows using the manikin Fiala-FE,⁴⁹ which can perform all types of thermophysiological regulations that occur in the human body, that is, the controlling active system (shivering, vasomotion, and sweating) as well as the passive system (convection, conduction, radiation, sweat evaporation, and breathing). To perform these thermophysiological regulations, the manikin Fiala-FE uses two different models: the shell and solver internal manikin models, as shown in Figure 5. A shell model is used for pre- and post-processing in a GUI (graphic user interface), for example for visualizing simulation results and applying boundary conditions and the clothing layer. A solver internal manikin model is the Fiala Model,⁵⁰ which consists of a half-sphere for the head and solid cylinders for the remaining body parts, as shown in Figure 5. Both models are connected via heat fluxes. The shell model calculates the heat exchange between the manikin and the surrounding environment and then gives signals (heat flux) to the solver internal manikin. The manikin Fiala-FE manikin uses finite elements that build up the layers, sectors, and body elements to simulate the transportation of metabolic heat from the inner body to the outer surfaces.⁴⁹ Moreover, the results of the 3D fit simulation were also imported into this software.
Figure 5.
(a) Scanned data of the test person, (b) discretization of the thermophysiological manikin Fiala-FE,⁴⁹ and (c) thermophysiological model of the test person.

The air gap thickness has a considerable influence on the thermoregulation of the human body and is not uniformly distributed throughout the body. Therefore, the clothing was divided into air zones with respect to the thickness of air gaps and their location on the body (Figure 6). For example, the air gap above a body element has only one air zone if its thickness is less than or equal to 10 mm. If the air gap thickness varies between above and below 10 mm, it was divided into two air zones. One has an average thickness below 10 mm, whereas the other has an average thickness above 10 mm. The heat transfer coefficient of each air zone (assumed as a vertical cavity) was calculated with the help of the Nusselt number, which is the function of the Prandtl and Rayleigh numbers.⁵¹ The processes of coupling the manikin Fiala-FE manikin with the scanned data of the test person, dividing the air gaps into air zones, and calculating the heat transfer coefficient have already been discussed comprehensively in a previous publication.⁸
Figure 6.
Defining air zones by dividing air gaps with respect to their thickness and location on the body.

The following parameters and boundary conditions were defined prior to thermal simulation:
material properties: conductivity, specific heat, thickness, and mass per unit area for both the long sleeve shirt and trekking trousers;

environment temperature 23℃, 50% relative humidity (RH), and air velocity of 0.3 m/s;

metabolic rates during the simulation as mentioned in Table 3, which represent different phases of outdoor activities: sitting, walking at 4 km/h, and walking at 6 km/h.

Table 3.
Phases of activities during the simulation

Phase Activity Duration Metabolic rate (Met)

1 Sitting 15 min 1.2

2 Walking at 4 km/h 40 min 2.6

3 Sitting 15 min 1.2

4 Walking at 6 km/h 40 min 4.2

5 Sitting 15 min 1.2

The metabolic rate M of the test person was calculated by the following equation⁵²
$M = 1.5 W + 2.0 (W + L) (\frac{L}{W})^{2} + η (W + L) (1.5 v^{2} + 0.35 vG)$
(1)
where W is the body weight, L is the carried load, v is the walking speed, η is the nature of the terrain, and G is the walking grade.
The boundary conditions regarding the heat transfer due to the convection within air zones are defined with the help of a heat transfer coefficient for each air zone, whereas the convective heat transfer between the external environment and outer clothing surface was defined as a function of air velocity. The view factor cavity was also defined to determine heat transfer due to radiation.

The software does not take into account water vapor movement through the fabric. Therefore, the RH that was measured during the wear trial of an identical clothing system and test person⁵³ was assigned in the air zones against each step of time. Thus, the cooling effect on the skin due to evaporation was realized.

Results and discussion

The main parameters (core body and mean skin temperatures) that are responsible for thermal comfort are illustrated in Figure 7 for both cases A and B. The trend of core body temperature is identical in both cases due to the active system of the Fiala model that also functions the same in both cases. The main function of the active system is to maintain the core body temperature at a set point (37℃). Both graphs of mean skin temperature illustrate the same trend as well, although there are some minor variations in values. The mean skin temperature in case B (F3 and F4) is slightly higher compared to that of case A, which results from the higher thermal resistance of the fabrics in case B.
Figure 7.
Comparison of (a) mean skin temperatures and (b) core body temperatures during thermal simulation in both cases.

Once the simulation begins, the mean core and skin temperatures start to reduce due to the influence of the outside environment (23℃ and 50% RH). However, as soon as the second phase sets in, the activity level increases from 1.2 Met (0 km/h) to 2.6 Met (4 km/h) so that internal heat production also increases due to the rising metabolic rate. Hence, both temperatures (core and skin) start to increase in the middle of the second phase. Although the third phase involves resting (0 km/h), a slight increase in both temperatures can still be noticed. This phenomenon results from the large amount of energy produced by the test person in the second phase due to a high metabolic rate that must be released to the environment in order to achieve a neutral condition (production of heat = release of heat). With the beginning of the fourth phase (6 km/h), the metabolic rate increases and, therefore, a continuous increase in core temperature throughout this phase is noticed. In contrast, the skin temperature, which was increasing at the beginning, started to decrease after some time. This fall in temperature continued until the end of the first half of the fifth phase (0 km/h). This is due to sweating and the resulting heat release by evaporation from the skin – a very effective phenomenon for heat release that sets in if vasodilation becomes insufficient in maintaining the core body temperature.

Visual and graphical representations of temperature changes on the clothing surface, skin, and microclimate on different places across the body throughout the thermal simulation in both cases A and B are presented in Figures 8–10.
Figure 8.
Clothing surface temperature at different time steps during thermal simulation.

The results can be interpreted as follows.
The outside environment is kept at a constant temperature of 23℃ and 50% RH, ensuring a constant clothing surface temperature throughout the simulation (Figure 8).

The clothing surface temperature at the abdomen level is lower compared to other body regions (Figure 8). This is due to the overlapping of fabrics, thus producing extra air gaps between them; these gaps cause an increase in thermal resistance and hence prevent heat dissipation at the clothing outer surface. This prevention of heat dissipation also increases the microclimate and skin temperatures compared to the surrounding abdomen areas (Figures 9 and 10).

The lower part of the trekking trousers has many folds and therefore produces more heterogeneous air gaps around the calf compared to other body regions. The garment folds cause more heat loss from the body¹⁴ and, therefore, the clothing surface and microclimate temperatures associated with these body regions are lower.

Figure 9.
Skin temperature at different time steps during thermal simulation.

Figure 10.
Microclimate temperature at different time steps during thermal simulation.

Figure 11 illustrates the behavior of skin and microclimate temperatures at five different locations of the body (chest left, upper arm right, shoulder blade right, front thigh right, calf left) during the simulation. It can be observed that material properties have a considerable influence on skin and microclimate temperatures at different body locations. As the thermal resistances of fabrics F3 and F4 of case B are higher than those of fabrics F1 and F2 of case A, all temperature curves of case B show higher values (Figure 11).
Figure 11.
The behavior of skin and microclimate temperatures at different body locations during thermal simulation.

The differences in microclimate temperatures are more significant compared to the skin temperatures in both cases A and B. This is also due to the active system of the Fiala model, which tries to maintain body temperatures at a constant level (Figure 11). Moreover, the microclimate temperatures of thicker air gaps are more significantly affected by changing clothing materials than thinner air gaps. For example, the microclimate temperature of air gaps measuring 8 mm or less than 8 mm is affected up to 0.8℃ by a change of material compared to air gaps larger than 8 mm (Figure 11 and Table 4). Therefore, the air gap at the calf (left) exhibits the largest difference in microclimate temperature (more than 1℃). The air gap thickness at shoulder blade right is almost the same as compared to the front thigh right (Table 4), but the microclimate temperature difference at the front thigh right is more significant. This is due to the many other influential factors, for example different thermophysiological reactions (skin temperature, sweating, vasomotion) of the different body parts and different thermal properties (thermal and water vapor resistance) of the fabrics.
Table 4.
Thickness of air gaps at specific body parts

Air zones Thickness (mm)

Case A Case B

Chest left 5.3 5.0

Shoulder blade right 7.0 7.5

Upper arm right 3.6 5.0

Front thigh right 7.0 6.0

Calf left 17.0 18.0

Thermal comfort assessments of the investigated pieces of clothing in terms of dynamic thermal sensation (DTS) and Zhang overall thermal sensation index are illustrated in Figure 12.
Figure 12.
Assessment of (a) dynamic thermal sensation and (b) Zhang overall thermal sensation for both pieces of clothing.

Figure 12(a) shows the DTS index on a seven-point ASHRAE-scale (–3: cold; –2: cool; –1: slightly cool; 0: neutral; +1: slightly warm; +2: warm; +3: hot). The DTS has been validated for the sudden change of environmental parameters as well as steady-state conditions such as 1 or 3 hours exposure. Fiala developed a function for the DTS index with the variables of core body temperature and mean skin temperature and its gradient. The results show that in both cases, the DTS index dropped suddenly to a value of –2 once simulation begins due to the ambient conditions. Subsequently, it increased continuously until the end of the second phase until it reaches a value of 0.3, which remains nearly constant throughout the sitting phase (third phase), that is, an almost neutral thermal sensation. When the fourth phase (6 km/h) begins, a sharp increase in the DTS index from 0.3 to 1 in addition to minor variations occur until the end of the simulation. This is due to the increase in core and skin temperatures at the beginning of the fourth phase, as shown in Figure 7. The skin temperature starts to decrease after some time intervals (6 min); therefore, the DTS index value stops rising and remains at approximately 1.

Hui Zhang⁵⁴ developed a powerful mathematical framework for the assessment of human thermal sensation and comfort, which is based on a large number of human experiments in a climate chamber, using regression analysis. In these experiments, the local skin temperature was measured as a major input variable for thermal sensation. The Zhang overall sensation (ZOTS) index for both clothing items on a nine-point scale (–4: very cold; –3: cold; –2: cool; –1: slightly cool; 0: neutral; +1: slightly warm; +2: warm; +3: hot; +4: very hot) can be seen in Figure 12(b). The results reveal that the ZOTS index in both cases shows almost the same trend apart from small variations. Due to the outside temperature (23℃), there is a sudden fall of –1 (slightly cool) in the ZOTS index and, subsequently, it remains between 0 (neutral) and –1 (slightly cool) until the end of the third phase (sitting). In contrast to the DTS index, there is a sharp decline in the ZOTS index with the start of the fourth phase (6 km/h) until it reaches –1.8, which is due to the decrease in mean skin temperature (Figure 7) being the major parameter of the ZOTS index. In the last phase of sitting, the ZOTS index increases to some extent, which is caused by the increasing mean skin temperature.

Both thermal sensation indices illustrate that the higher thermal resistance of the case B fabric combination yields warmer feeling garments compared to those of case A. A significant difference in both indices can be seen after starting the simulation, although it starts to reduce after 102 min; at one point, both indices are almost identical. This is due to the mean skin temperature decreasing more substantially in case B compared to case A, thus minimizing the mean skin temperature difference. The higher rate of decreasing mean skin temperature in case B results from the higher thermal resistance producing more sweat compared to case A. From the results of the DTS index, it can be assumed that until the end of the second phase (4 km/h), case B provides more thermal comfort, which starts to decrease afterward. This is due to the higher thermal resistance of case B, which prevents heat dissipation of extra heat to the environment produced during walking phases. Therefore, after the second phase, case A can provide more thermal comfort because of its lower thermal resistance values. In contrast, the ZOTS index shows that case B can provide more thermal comfort compared to case A. The differences in the prediction of thermal sensation for both indices are caused by different calculation approaches.

Conclusion

The method introduced in this paper includes clothing drapability, the human body shape, and the thermoregulation of the human body according to different activity levels. It offers a holistic solution to predict the thermal comfort of clothing. Four different types of materials were selected and comprehensively characterized in the laboratory. A test person was scanned, and 2D patterns were developed according to the selected design and size of the test person. Subsequently, a 3D fit simulation of the clothing was performed on a virtual model of the test person by considering the draping properties of the fabric. Air gaps, which occur between the body and the clothing, were analyzed for both cases and then divided into several air zones. The heat transfer coefficient for every air zone was calculated. Furthermore, software limitations that do not allow the moisture movement through the fabric was compensated by defining the RH in air zones against each step of time. These RH values of air gaps were measured during wear trials of the same clothing system and test person, which enabled the actual cooling effect on the skin due to evaporation and diffusion of the moisture. A Fiala-FE manikin was coupled with a virtual model of the test person in order to study the thermoregulation of the human body with its environment. In a final step, the thermal comfort of clothing in both cases was evaluated through simulation. The presented results lead to the following conclusions.
Since the shape and thickness of air gaps between the clothing and the human body are affected by the draping properties of the clothing, the thermal characteristics of fabrics as well as air gaps have a significant influence on the thermal comfort of clothing. Moreover, the temperatures of thicker microclimates are more significantly affected by changing clothing materials than thinner microclimates.

A change of clothing materials has a much greater influence on the microclimate temperature compared to local skin temperatures. This is due to the active system of the human body, which constantly tries to maintain its set-point temperatures for achieving optimum comfort.

According to the prediction of thermal sensation based on DTS and ZOTS indexes, the garment in case B feels warmer compared to that in case A. This is due to the higher thermal resistance of F3 and F4 (case B) compared with F1 and F2 (case A). Furthermore, the indices yield different predictions during thermal simulation, which is due to their fundamentally different index calculation approaches.

The presented method is very useful for designing sports and works garments, where thermal comfort plays an important role for the performance of the wearer.

Case	Clothing	Fabric code	Fabric type	Specifications	Mass per unit area [g/m²]
A	Long sleeve shirt	F1	Knitted	98% Rec. PES/2% PES	161
Trekking trousers	F2	Woven	100% PA	125
B	Long sleeve shirt	F3	Knitted	100% PES hybrid yarn	92
Trekking trousers	F4	Fleece	100% PES	168

Characteristics	Instruments	Textile material
Physical
Fabric thickness [mm]	Karl Schröder KG Material Testing Machine	0.46	0.29	0.41	1.55
Average mass per unit area of the fabrics [g m^–2]	GSM Cutter, weighing balance	161	125	92	168
Thermal
Water vapor resistance [m² Pa W^–1]	Sweat guarded hot plate	1.69	3.18	2.98	5.57
Thermal resistance [m² K W^–1]	Sweat guarded hot plate	0.005	0.009	0.024	0.063
Specific heat [Jkg^–1K^–1]	Calculated	493	295	170	1640
Mechanical
Bending stiffness [µN m] (warp, weft, 45°)	Bending stiffness tester Cantilever ACPM 200	1.18 0.74 0.63	3.22 10.32 4.82	1.58 1.28 0.42	2.87 2.88 2.91
Elongation [%]
5 N/m	Zwick Tensile Strength Tester
Warp		1.15	0.42	2.81	0.42
Weft		2.56	0.05	4.11	1.27
20 N/m
Warp		4.34	1.09	7.61	2
Weft		10	0.21	14	6.4
100 N/m
Warp		10.88	2.51	19.48	7.17
Weft		19	0.98	50.7	28.2
Shear stiffness [N/m]		40.46	124.24	23.21	57.48

Phase	Activity	Duration	Metabolic rate (Met)
1	Sitting	15 min	1.2
2	Walking at 4 km/h	40 min	2.6
3	Sitting	15 min	1.2
4	Walking at 6 km/h	40 min	4.2
5	Sitting	15 min	1.2

Air zones	Thickness (mm)
Chest left	5.3	5.0
Shoulder blade right	7.0	7.5
Upper arm right	3.6	5.0
Front thigh right	7.0	6.0
Calf left	17.0	18.0

Footnotes

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the institution AiF. The IGF research project 19472 BG of the Forschungsvereinigung Forschungskuratorium Textil e. V. is funded through the AiF within the program for supporting the “Industrielle Gemeinschaftsforschung (IGF)” from funds of the Federal Ministry for Economic Affairs and Energy (BMWi) by a resolution of the German Bundestag.

ORCID iD

Muhammad Awais

References

ANSI/ASHRAE Standard 55-2013: 2010. Thermal environmental conditions for human occupancy.

Parsons

. Human thermal environments, 2nd ed. London and New York: Taylor and Francis Group, 2003.

Psikuta

. Development of an ‘artificial human’ for clothing research, Leicester: De Montfort University, 2009.

Das

Ishtiaque

. Comfort characteristics of fabrics containing twist-less and hollow fibrous assemblies in weft. J Text Apparel Technol Manag 2014; 3(4): 1–7–1–7.

Havenith

. Heat balance when wearing protective clothing. Ann Occup Hyg 1999; 43: 289–296.

Ogulata

. The effect of thermal insulation of clothing on human thermal comfort. Fibre Text East Eur 2007; 15: 67–72.

Duncan S, Mclellan T, Dickson EG. Improving comfort in military protective clothing. In: Song GBT-IC in C (eds) Woodhead Publishing Series in Textiles. Woodhead Publishing, 2011, pp. 320–369. DOI: 10.1533/9780857090645.3.320.

Awais

Krzywinski

Method development for modeling, designing, and digital representation of outdoor and protective clothing. In: Majumdar

Gupta

(eds). Functional textiles and clothing, Singapore: Springer Singapore, 2019, pp. 205–218.

Mayor

Couto

Psikuta

, et al. Advanced modelling of the transport phenomena across horizontal clothing microclimates with natural convection. Int J Biometeorol 2015; 59: 1875–1889.

10.

McCullough

Jones

. A comprehensive data base for estimating clothing insulation. ASHRAE Trans 1984; 2888: 29–29.

11.

Frackiewicz-kaczmarek J, Psikuta A, Bueno M. Effect of garment properties on air gap thickness and the contact area distribution. Text Res J 2015; 85: 1907–1918.

12.

McQuerry

DenHartog

Barker

. Analysis of air gap volume in structural firefighter turnout suit constructions in relation to heat loss. Text Res J 2017; 88: 2475–2484.

13.

Mert E, Psikuta A, Arévalo M, et al. A validation methodology and application of 3D garment simulation software to determine the distribution of air layers in garments during walking. Measurement 2018; 117: 153–164.

14.

Mert

Psikuta

Bueno

, et al. Effect of heterogenous and homogenous air gaps on dry heat loss through the garment. Int J Biometeorol 2015; 59: 1701–1710.

15.

Hes

de Araujo

. Simulation of the effect of air gaps between the skin and a wet fabric on resulting cooling flow. Text Res J 2010; 80: 1488–1497.

16.

Daanen H, Hatcher K, Havenith G. Determination of clothing microclimate volume. In: Tochihara Y and Ohnaka TBT-EEBS (eds) Environmental Ergonomics. Vol 3. Elsevier; 2005: 361–365. doi:https://doi.org/10.1016/S1572-347X(05)80057-6.

17.

Zhang

Z, Li J

. Volume of air gaps under clothing and its related thermal effects. J Fiber Bioeng Informatics 2011; 4: 137–144.

18.

Psikuta

Frackiewicz-Kaczmarek

Frydrych

, et al. Quantitative evaluation of air gap thickness and contact area between body and garment. Text Res J 2012; 82: 1405–1413.

19.

Daanen

HAM

Psikuta

10 - 3D body scanning. In: Nayak

Padhye

RBT-A

(eds). The Textile Institute Book Series, United Kingdom: Woodhead Publishing, 2018, pp. 237–252.

20.

Kim

IlY

Lee

, et al. Investigation of air gaps entrapped in protective clothing systems. Fire Mater 2002; 26: 121–126.

21.

Wang

Newton

, et al. Ease distribution in relation to the X-line style jacket. Part 1: development of a mathematical model. J Text Inst 2006; 97: 247–256.

22.

Mert E, Psikuta A, Arevalo M, et al. Quantitative validation of 3D garment simulation software for determination of air gap thickness in lower body garments. MS&E 2017; 254: 162007.

23.

Mert

Psikuta

Arévalo

, et al. A validation methodology and application of 3D garment simulation software to determine the distribution of air layers in garments during walking. Measurement 2018; 117: 153–164.

24.

Lotens

Havenith

. Calculation of clothing insulation and vapour resistance. Ergonomics 1991; 34: 233–254.

25.

Holcombe

. Mathematical simulation of heat and moisture transfer in a human-clothing-environment system. Text Res J 1998; 68: 389–397.

26.

Fan

Luo

. Heat and moisture transfer with sorption and condensation in porous clothing assemblies and numerical simulation. Int J Heat Mass Transf 2000; 43: 2989–3000.

27.

Berger

Sari

. A new dynamic clothing model. Part 1: heat and mass transfers. Int J Therm Sci 2000; 39: 673–683.

28.

Gagge

Stolwijk

Nishi

. An effective temperature scale based on a simple model of human physiological regulatory response. SHRAE Trans 1971; 77: 247–262.

29.

Fanger

. Thermal comfort: Analysis and applications in environmental engineering, Danmarks Tekniske Hojskole: Danish Technical Press, 1970.

30.

Wissler

. Mathematical simulation of human thermal behavior using whole body models. Heat Transf Med Biol 1985; 1: 325–373.

31.

Stolwijk

Nadel

Wenger

, et al. Development and application of a mathematical model of human thermoregulation. Arch Sci Physiol (Paris) 1973; 27: 303–303.

32.

Fiala

Lomas

Stohrer

. A computer model of human thermoregulation for a wide range of environmental conditions: the passive system. J Appl Physiol 1999; 87: 1957–1972.

33.

Fiala

Lomas

Stohrer

. Computer prediction of human thermoregulatory and temperature responses to a wide range of environmental conditions. Int J Biometeorol 2001; 45: 143–159.

34.

Tanabe S, Kobayashi K, Nakano J, et al. Evaluation of thermal comfort using combined multi-node thermoregulation (65MN) and radiation models and computational fluid dynamics (CFD). Energy Build 2002; 34: 637–646.

35.

Katic

Zeiler

. Thermophysiological models and their applications: a review. Build Environ 2016; 106: 286–300.

36.

Fiala

Havenith

Bröde

, et al. UTCI-Fiala multi-node model of human heat transfer and temperature regulation. Int J Biometeorol 2012; 56: 429–441.

37.

Severens

NMW

Lichtenbelt

WD van M

Frijns

AJH

, et al. A model to predict patient temperature during cardiac surgery. Phys Med Biol 2007; 52: 5131–5145.

38.

Psikuta

Richards

Fiala

. Single-sector thermophysiological human simulator. Physiol Meas 2008; 29: 181–192.

39.

Psikuta A, Richards M and Fiala D. Single- and multi-sector thermophysiological human simulators for clothing research. In: 7th international thermal manikin and modelling meeting, Coimbra, Portugal, 3–5 September 2008, pp.1–5.

40.

Fiala D and Lomas KJ. Application of a computer model predicting human thermal responses to the design of sports stadia. In: conference proceedings (CIBSE ’99), Harrogate, UK, 3–5 October 1999, pp.492–499.

41.

Yang T, Cropper PC, Cook MJ, et al. A new simulation system to predict human-environment thermal interactions in naturally ventilated buildings. In: proceedings of the 10th international conference on building simulation, Beijing, China, 3–6 September, 2007, pp.751–756.

42.

Lectra. http://www.lectra.com/en (accessed 10 June 2017).

43.

De Boos AG, Tester DH. SiroFAST: Fabric Assurance By Simple Testing: A System for Fabric Objective Measurement and Its Application in Fabric and Garment Manufacture. CSIRO Division of Wool Technology; 1994. DOI: 10.25919/5d9e2029a7814.

44.

Kawabata

Niwa

. Fabric performance in clothing and clothing manufacture. J Text Inst 1989; 80: 19–50.

45.

Vitronic. https://www.vitronic.com/industrial-and-logistics-automation/sectors/3d-body-scanner.html (accessed 1 August 2016).

46.

D scanning, design and reverse engineering software from 3D Systems Geomagic. http://www.geomagic.com/en/ (accessed 7 September 2016).

47.

Rogers

. An introduction to NURBS with historical perspective, San Francisco, CA: Morgan Kaufmann, 2001.

48.

Theseus FE. http://www.theseus-fe.com/simulation-software/human-thermal-model (accessed 3 October 2017).

49.

Paulke S and Wagner S. Some considerations on global and local thermal comfort based on Fiala’s thermal manikin in Theseus-Fe. In: 6th EUROSIM congress on modeling and simulation, Ljubljana, Slovenia, 9–13 September 2007, pp.1–7.

50.

Fiala

. Dynamic simulation of human heat transfer and thermal comfort (thesis). Sustain Dev 1998; 45: 1–1.

51.

Incropera

DeWitt

Bergman

, et al. Fundamentals of heat and mass transfer, 6th ed. Chichester: John Wiley & Sons, 2007.

52.

Pandolf

Givoni

Goldman

. Predicting energy expenditure with loads while standing or walking very slowly. J Appl Physiol 1977; 43: 577–581.

53.

Krzywinski S, Wendt E and Awais M. Methodenentwicklung zur skalenübergreifenden Modellierung, Auslegung und digitalen Darstellung von Outdoor-Bekleidung auf Basis geometrischer, mechanischer und thermischer Kennwerte. TU Dresden: ITM, 2019.

54.

Zhang H. Human Thermal Sensation and Comfort in Transient and Non-Uniform Thermal Environments. PhD thesis, University of California, Berkley, 2003.

Air zones	Thickness (mm)
Air zones	Case A	Case B
Chest left	5.3	5.0
Shoulder blade right	7.0	7.5
Upper arm right	3.6	5.0
Front thigh right	7.0	6.0
Calf left	17.0	18.0