ORIGINAL PAPER

International Journal of Occupational Medicine and Environmental Health 2017;30(3):379395

https://doi.org/10.13075/ijomeh.1896.00761

EVALUATION OF OCCUPATION HOT EXPOSURE

IN INDUSTRIAL WORKPLACES
IN ASUBTROPICAL COUNTRY
YU-CHIAO YANG1,2, MING-CHI WEI3, and SHOW-JEN HONG1
1
Kaohsiung Medical University, Kaohsiung, Taiwan
Department and Graduate Institute of Pharmacology
2
Kaohsiung Medical University Hospital, Kaohsiung, Taiwan
Department Medical Research
3
Chia Nan University of Pharmacy and Science, Tainan, Taiwan
College of Sustainable Environment, Department of Applied Geoinformatics

Abstract
Objectives: Theobjective of this study has been to evaluate theoccupational heat exposure of12workers at5plants in
asubtropical country. Material and Methods: Theheat stresses and strain on workers in5plants were assessed by theIn-
ternational Organization for Standardization(ISO)7243index (wet bulb globe temperatureWBGT) and theISO7933in-
dex (maximum allowable exposure timeDlim). Results: Results indicated that42%of thesubjects (5workers) surpassed
theWBGTlimits. According to theDlim,42%of thesubjects could not continue working in thehot environments. There-
lationships between thevarious heat stress indices and theWBGTindex were also correlated. However, further studies
from different heat environments and more subjects should be performed. Conclusions: Thesensitive dependence of skin
temperature on meteorological and physiological indices for each subject was clearly observed. Obviously, theheart rate
response to metabolic rate was much greater than that caused by environmental heat alone. Theexponential relationship
between workers duration-limited exposure time, predicted by various estimated criteria, andWBGT were also found.
IntJOccup Med Environ Health2017;30(3):379395

Key words:
Skin temperature, Heart rate, Heat stress, Wet bulb globe temperature, Maximum allowable exposure time, Metabolic rate

INTRODUCTION limits. Workers are often exposed to severe environmental

Heat stress is prevalent in some industrial workplaces, heat stresses, which may deteriorate work efficiency and
such as themetal, glass, mining, ceramic, construction productivity and may even be life threatening. TheInter-
industries as well as engine or boiler rooms; workers are national Organization for Standardization(ISO)7243in-
frequently exposed to heat stresses above conventional dex (wet bulb globe temperature WBGT) [1,2] and

Funding: the article was financially supported by the National Science Council of Taiwan, Kaohsiung Medical University (Office of Research and Development, Center
for Researchs and Development) (KMU-M106034 & M104020), grant manager: Yu-Chiao Yang, Ph.D., and Chuang Song Zong Pharmaceutical Co., Ltd. (Kaohsiung,
Taiwan) (S 102035 & S 102036).
Received: August 4, 2015. Accepted: March 1, 2016.
Corresponding author: M.-C.Wei, Chia Nan University of Pharmacy and Science, College of Sustainable Environment, Department of Applied Geoinformatics, No.60,
sec.1, ErrenRd., Rende District, Tainan71710, Taiwan (e-mail:s120702@mail.cnu.edu.tw).

Nofer Institute of Occupational Medicine, d, Poland 379

 ORIGINAL PAPER Y.-C. YANG ET AL.

the ISO 7933 index (maximum allowable exposure Comparison different heat-stress indices has shown that
timeDlim)[3,4] are recommended by theInternational theISO7933standard[3] affords thebest prediction of
Organization for Standardization (ISO) as the interna- strain caused by muscular work and heat whether strain
tional standard for assessing heat load. Theaim of both is measured by sweat rate, heart rate or core body tem-
standards is to regulate chronic exposure to heat stresses, perature[14]. Some data supports theuse of arequired
and the World Health Organization (WHO) [5] has ad- sweat rate as atrustworthy indicator of response to heat
opted acore body temperature of38C as its upper limit. stress, while other data supports aconflicting conclusion.
TheWBGTindex is by far themost commonly used heat Kampmann and Piekarski [15] delineated the consider-
stress index throughout theworld (e.g.,it is used in Aus- able discrepancies between theactual sweat rate and
tralia, China, India, Japan, the United Kingdom (UK), thesweat rate predicted by themethod as well as between
theUnited States of America(USA), and theEuropean theactual and predicted core body temperature. Work
Union). It was developed by Yaglou and Minard [6] to by Khknen[16] showed similar problems when testing
reduce theincidence of heat illness during military train- themethod across several industries. Furthermore, sever-
ing, and it gained popularity mainly due to its relative al studies have shown that theISO7933standard[3] is not
simplicity and convenience of use. This index has also valid in conditions where protective clothing, high levels
been adopted by other organizations to prevent heat in- of radiation, high humidity, high air velocity or saturated
juries[79]. Despite thealmost ubiquitous application of clothing are involved[17].
theWBGTindex, evidence suggests that theindex has Mehnert et al. [18] demonstrated that the prediction
several significant limitations. Theestimation of metabol- of mean skin temperature used for the ISO 7933 stan-
ic rate, variations in skin temperature or skin dampness, dard[3] purposes was not valid in places with high levels
theeffects of air velocity, and theeffects of clothing all of radiation or humidity. For thebest prediction of strain,
limit theefficacy of theindex. Nevertheless, several stud- thepredicted heat strain(PHS) model was developed to
ies have used theWBGTindex as asafety index for work- predict therequired sweat rate as well as rectal tempera-
ers across various vocations[1012]. tures [19,20], and thevalues predicted by this model are
If theWBGTvalues of thehot environment exceed comparable to the ISO 7933 standard [3,4]. The PHS
theWBGTreference values, or if amore detailed anal- has been entirely replaced by therevised reversion of
ysis is required, then the ISO 7933 standard [3,4] pro- theISO7933standard[3]. The2004revised standard[4]
poses ananalytical method of evaluating heat stress. includes amendments to the algorithms (PHS index)
TheISO7933standard[3] is derived from theBelding and that predict changes in sweat rate and internal body
Hatch heat stress indexes(HSI) criteria[13], which repre- temperature in response to heat and amendments to
sents thedetermination and interpretation of heat stress themaximum allowable exposure times(Dlim). TheDlim is
using calculation of therequired sweat rate(SWreq) and is reached when either thecumulated water loss or therec-
derived from thethermal equilibrium of thehuman body tal temperatures reaches thecorresponding maximum
through abalance between metabolic heat production and values. Comparisons between thecumulated water loss
heat dissipation from thebody. TheISO7933standard[3] and therectal temperatures have found that thecumu-
determines therequired sweat rate and is considered more lated water loss is auseful indicator for theinstruction of
detailed risk than theWBGTindex, and therefore, it is workers in extremely hot environments and for ensuring
abetter method for predicting uncompensable heat-stress. their health and safety.

380 IJOMEH 2017;30(3)

 OCCUPATIONAL HOT EXPOSURE IN INDUSTRIAL WORKPLACES ORIGINAL PAPER

To assess themaximum allowable exposure time, themaxi MATERIAL AND METHODS

mum water loss is set at7.5%of thebody mass (Dmax50, aver Test locations and subject descriptions
age or median subject) and5%of thebody mass (Dmax95, Four heavy-oil power plants and1coal-fueled power plant
most susceptible subject). Thus, theDlim may be calculated were surveyed (Table 1). All 4 heavy-oil power facilities
for anaverage subject on thebasis of amaximum water had similar heat sources but varied in thesize of theboiler
loss of7.5%of thebody mass and on thebasis of5%of and thetype of production (designated as plantsA,B,C,
the body mass in order to protect 95% of the work- and D). Plant E (paper plant) was a coal-fueled power
ing population. However, compared to theprevious re- plant, which had thehigh radiant heat generated from
quired sweat rate index(ISO79331989)[3], the2004re- theboiler. All5plants were in thecity of Tainan (southern
vised ISO 7933 standard [4], there are also such similar part of Taiwan), and this region was extremely hot during
limitations in conditions where protective clothing, high thesummer months. Thedry air temperatures ranged fre-
levels of radiation and high humidity are involved[21,22]. quently3040C in July and August.
Moreover, many different models, such as theenviron- Monitoring of both area and personal heat stress was
mental stress index(ESI)[23], discomfort index(DI)[24], performed during thesummer months to try to establish
modified discomfort index(MDI)[25], effective tempera- thehottest working conditions. Prior to beginning theex-
ture(ET)[26], corrected effective temperature(CET)[27], perimental trials, the supervisors and/or employees of
equivalent temperature (Teq) [28] and operative tempera- each plant were interviewed to identify those employees at
ture(To)(ISO7730)[29] have been proposed as alterna- thehighest risk for heat strain. Based on these interviews,
tives to rational methods of heat-stress evaluation. Thedif- test locations were selected around themobility zones of
ferent methods for evaluating heat stress all have advan- workers and their respective refreshment rooms as well
tages and disadvantages. However, thedifferences between as outside thebuildings in theshade. Adetailed descrip-
thevarious heat-stress indices and theWBGT are rare in tion of each test location is also provided in theTable1.
theliterature. In addition, heat stress is generally related All12workers in this study were men ranging2751years
to meteorological parameters and physiological variables. old (34.510.6 years) who had been employed in these
During exposure to hot environments, various physi- facilities for at least 6 months. Therefore, they were all
ological responses have been used as criteria for assess- considered experienced and acclimatized workers. Their
ing thephysiological heat stress, e.g.,required sweat average heightstandard deviation and weightstan-
rate(SWreq)[3,4], rectal temperature[5,30,31], oral tem- dard deviation were 168.45.9 cm and 68.67.3 kg,
perature [3133], skin temperature [18,32,33] and heart respectively.
rate[33]. To thebest of theauthors knowledge, there has
been little scrutiny of therelationship between meteoro- Meteorological measurements
logical parameters and physiological variables. This research was done in July and theselected plants
To address this gap, this study correlates thevarious evalu- were in thecity of Tainan, Taiwan. Environmental pa-
ation criteria with theWBGTindex in4heavy-oil power rameters including dry-bulb temperature (Ta [C]),
plants and1coal-fueled power plant. Thecorrelations be- natural wet-bulb temperature (Tnw [C]), wet bulb tem-
tween thevarious heat-stress indices and thephysiological perature (Tw [C]), globe temperature (Tg [C]), and
responses, such as sweat rate, skin temperature, metabolic air velocity (Va [m/s]), were measured. All instruments
rate, oral temperature and heart rate, are also discussed. compliant with the ISO 7726 [34] and ISO 7243 [2]

IJOMEH 2017;30(3) 381

 ORIGINAL PAPER Y.-C. YANG ET AL.

Table1. Test locations in thestudy evaluating theoccupational heat exposure of12workers at5plants in asubtropical country

Location Description
PlantA boiler room of textile plant, heat source:15t/hheavy oil-boiler,1worker
1 approximately2m from theboiler front
2 passageway
3 rest room
4 outside of building in shade
PlantB dye vat department of textile plant, heat source:15t/hheavy oil-boiler,4workers
1 dye vat stove, side
2 passageway
3 dye vat stove, side
4 furnace, side
5 boiler, side
6 outside of building in shade
PlantC boiler room of forage plant, heat source:9.6t/hheavy oil-boiler,1worker
1 boiler, side
2 boiler, side
3 approximately2m from theboiler front
4 rest room
5 outside of building in shade
PlantD tinplate division department of tinplate plant, heat source:20t/h heavy oil-boiler,5workers
1 passageway
2 toaster, side
3 toaster, side
4 package area
5 rest room
6 boiler, side
7 outside of thebuilding in theshade
PlantE boiler room of paper (pulp) plant, heat source:120t/hcoal-boiler,1worker
1 boiler, side
2 boiler, front
3 boiler, side
4 boiler, rear
5 boiler, feed-side
6 rest room
7 outside of building in shade

requirements. Experimental instrumentation, includ- and a hot wire anemometer (model KA22, Kanomax,
ing themercury-in-glass thermometers, Assmann psy- Japan), were calibrated prior to each trial. All measure-
chrometer (KP-25, Komatsu Factory Co., Ltd., Japan) ments were taken at a height of 110 cm (chest height

382 IJOMEH 2017;30(3)

 OCCUPATIONAL HOT EXPOSURE IN INDUSTRIAL WORKPLACES ORIGINAL PAPER

of theworker) to represent heat exposure to thetrunk, The Belding and Hatch heat stress index (HSI) was
and themeasurements were recorded every15min be- calculated according to theBelding and HatchHSIcri-
tween 9 a.m. and 3 p.m. Instruments were left at each teria [13]. Effective temperature (ET) and corrected
location for20min to reach equilibrium. effective temperature (CET) were estimated from psy-
chrometric charts using anAssmann psychrometer.
Physiological measurements In addition, several indices, including operative tem-
At each of thesurveyed plants,1observer followed1work- perature (To), equivalent temperature (Teq), Oxford in-
er for theentire work shift (time-motion study). Theop- dex(wetdry indexWD), discomfort index(DI), fighter
eration characteristics, work time, rest time, clothing and index of thermal stress(FITS), modified discomfort in-
volume of water intake were monitored during thework dex (MDI), environmental stress index (ESI), wet-bulb
shift. During thesummer, workers wore underpants, dry temperature (WBDT), and relative humidity dry
light-weight trousers, short-sleeve shirts or T-shirts, socks temperature (RHDT), were also estimated based on
and shoes assuming anapproximate clothing insula- thepublished literature[13,2329,37].
tion(Icl) of0.6clo (where1clo=0.155m2C/W). Several According to theworkers operation characteristics, work
times during each work shift, theobserver would survey time and rest time, thetime-weighted averages of heart
thepost-exercise recovery heart rate of theworker. When- rates were obtained to estimate thetime-weighted aver-
ever possible, these measurements were made at times ages (TWAs) of heat-stress indices. The time-weighted
when theworker would normally stop to rest. Theheart averages of theemployees metabolic rates were also com-
rate(HR) of each worker was also counted for1min after pared using themetabolic rate tables available in thelit-
specific tasks [32,35,36]. Heart rates were taken at least erature [1,7,9,38]. The calculated time-weighted aver-
twice to confirm thevalues. Workers were seated for these ages (TWAs) of heat-stress indices were compared with
measurements. recommendedTWAs and upper limits from thepublished
literature[1,3,7,9].
Calculations
TheWBGTindex and thevariables of theISO7933stan- Statistical analysis
dard were calculated according to theISO7243(2003)[2] Environmental conditions during thetest days were av-
andISO7933(2004)[4], respectively. Mean radiant tem- eraged for 5 h each working day (9 a.m. 3 p.m.), and
perature(Tr) was calculated according tothe ISO7726[34], data is presented in this study as means standard de-
using thefollowing equation: viations (MSD). Data analysis included constructing
linear models which were used for defining the correla-
Tr=[(Tg+273)4+2.5108Va0.6(TgTa)]0.25273 (1) tion between theWBGT and various evaluation criteria.
Correlations and regressions between 3 different physi-
where: ological variables (Tsk,HR, andSWreq) and various indi-
Tr mean radiant temperature, ces were determined for all data points. Theanalysis of
Tg globe temperature, variance (ANOVA) was carried out using Tukeys method
Va air velocity, with asignificance level ofp<0.05 using2010Microsoft
Ta dry-bulb temperature. Office Excel (Microsoft Co., USA) and Origin software
All temperatures are inC andVais inm/s. version6.1 (Origin LabCo.,USA).

IJOMEH 2017;30(3) 383

 ORIGINAL PAPER Y.-C. YANG ET AL.

RESULTS

Tg globe temperature; Ta dry-bulb temperature; Tnw natural wet-bulb temperature; Tw wet bulb temperature; Va air velocity; Pa partial water vapour pressure in theenvironment.
min.max
23.529.01 0.402.63 0.111.25 3.070.50 2.323.36
36.511.33 33.9837.57 34.703.45 30.2436.91 27.981.71 24.5629.15 26.811.50 25.4528.23 0.191.50 0.110.37 2.930.33 2.483.22
36.093.35 31.5039.59 33.782.90 30.2937.77 27.081.42 24.8328.41 25.821.64 24.2328.49 0.251.64 0.110.51 2.810.33 2.473.23
35.562.35 31.3138.26 33.152.06 30.0734.99 25.741.13 24.1926.82 26.041.22 23.0226.98 0.151.31 0.110.21 2.560.22 2.302.89
42.018.73 26.0350.58 34.975.46 25.9939.91 27.353.02 20.9829.71 26.003.61 18.8929.67 0.143.61 0.110.18 3.000.58 1.913.63
34.692.52 33.7736.10 31.121.80 29.5533.99 25.851.37 24.1927.21 25.241.31 25.1428.12 0.440.48 0.111.25 3.070.21 2.853.36
Restroom 30.962.95 26.0333.98 29.572.03 25.9930.30 23.991.70 20.9825.07 23.022.48 18.8925.45 0.160.05 0.110.22 2.390.35 1.912.85
37.565.24 26.0350.58 34.323.42 25.9939.91 27.041.99 20.9829.71 26.152.29 18.8929.67 0.210.22 0.111.25 2.850.42 1.913.63
The evaluation of meteorological parameters
From theTa andTwdata, therelative humidity, and ambi-

[kPa]
Pa
ent water vapor pressure(Pa) were determined with apsy-

MSD
chrometric chart. Ranges and means standard devia-
tions for each variable at each measurement site are given

min.max
in theTable2. All averages correspond to theaverage of

Table2. Environmental parameters in thestudy evaluating theoccupational heat exposure of12workers at5plants in asubtropical country
measurements taken from9a.m. to3p.m. These ambient

[m/s]
Va
temperatures and relative humidity are readily available

MSD
and thus tempting metrics for heat stress assessment[39].
TheTa,Tnw,Tw, andTg in thecoal-fueled power plant
(plantE) were higher than thetemperatures in the heavy-

min.max
oil power plants (plantsA,B,C andD). TheTg of plantE
ranged 40.4550.58C due to the higher radiant energy

[C]
Tw

36.693.50 32.9939.72 35.373.53 31.0238.34 27.341.63 25.0728.62 27.402.63

load. Notably, the temperature outside thebuilding in

MSD
the shade ranged 29.5533.99C for Ta, 24.1927.21C
forTnw, 25.1428.12C forTw, and33.7736.10C forTg dur-
ing this study. Theresults reveal that theweather in sub-
min.max

tropical climates is relatively hot and humid(68.362.54%)

during thesummer months. Additionally, as it may be seen
[C]
Tnw

from theTable2, it may be speculated that therest areas

MSD

were air-conditioned and had significantly lower tempera-

M mean; SD standard deviation; min. minimal value; max maximal value.

tures than work areas.
min.max

Relationships between theWBGTindex

and various heat-stress indices
[C]
Ta

Using thedata in theTable2 and themean radiant tem-

MSD

perature (Tr) (equation (1)), the values of various heat-

stress indices were determined for each of thework po-
sitions studied. To appraise therelationships between
min.max

thevarious heat-stress indices and theWBGTindex, this

Outside thebuilding in theshade.

study correlated the11indices with theWBGTindex, and

[C]
Tg

these results are tabulated in the Table 3. All statistical

MSD

contrasts were accepted at the p < 0.05 level of signifi-

cance. All statistical analyses were performed using anEx-
cel spreadsheet and anOrigin version6.1.
Location

a
Outside
PlantD

This table provides the numbers of observations (N),

PlantA

PlantC

PlantE
PlantB

Overall

corresponding regression equations, and correlation

a

384 IJOMEH 2017;30(3)

 OCCUPATIONAL HOT EXPOSURE IN INDUSTRIAL WORKPLACES ORIGINAL PAPER

Table3. Regression analysis results between wet bulb globe temperature(WBGT) and various heat stress indices in thestudy
evaluating theoccupational heat exposure of12workers at5plants in asubtropical country (N=29)

Index Regression equation Correlation coefficient(r2)

Effective temperaturea(ET) 0.684WBGT+8.283 0.83
Equivalent temperature (Teq)
b
1.609WBGT11.957 0.94
Corrected effective temperature (CET) c
0.856WBGT+4.280 0.95
Operative temperatured(To) 1.620WBGT11.829 0.93
Oxford index (WD)
e
0.729WBGT+5.370 0.74
Discomfort index (DI)
f
0.892WBGT+3.326 0.84
Fighter index of thermal stress (FITS)
g
0.941WBGT+10.413 0.79
Modified discomfort index (MDI) h
0.832WBGT+4.813 0.79
Environmental stress index (ESI)i
0.808WBGT+5.333 0.73
Wet-bulb dry temperature (WBDT)
j
0.939WBGT+2.740 0.86
Relative humidity dry temperature (RHDT) k
0.845WBGT+10.563 0.79
a
ET was estimated from psychrometric charts using an Assmann psychrometer.
b
Teq = 0.55Ta+0.45Tr+[(0.240.75Va 0.5)(36.5Ta)]/(1+Icl), Generally, during summer, the workers might be wearing light trousers and
short-sleeve shirt or T-shirt, with approximate clothing insulation(Icl) of0.4clo.
c
CETwas estimated from psychrometric charts using an Assmann psychrometer, orCET=0.786WBGT+6.0.
d
To=(0.45+0.25Va)Ta+[1(0.45+0.25Va)]Tr.
e
WD=0.85Tnw+0.15Ta.
f
DI=0.5Tnw+0.5Ta.
g
FITS=0.83Tnw+0.35Ta+5.08.
h
MDI=0.75Tnw+0.3Ta.
i
ESI=0.63Ta0.03RH(%)+0.002SR+0.0054(TaRH(%))0.073/(0.1+SR), whereRH(%) is relative humidity andSR is solar radiation.
j
WBDT=0.4Tnw+0.6Ta.
k
RHDT=0.9Ta+0.1RH(%).
Other abbreviations as inTable2.

coefficients (r2). Significant correlations were found
between various heat-stress indices and theWBGTin-
dex, withr2 values of 0.730.95. Highly significant cor-
relations were observed between theWBGTindex and
theTeq(r2=0.94), CET(r2=0.95), andTo(r2=0.93). In
addition, there was also a highly significant relationship
between theET andDIvalues(r2=0.93), as illustrated in
theFigure1. This was consistent with previous results[37].
Correlation between heat-stress indices
and meteorological parameters
To investigate theeffects of meteorological parameters
on heat-stress indices, theWBGT,Teq,CET, andTovalues
were related to environmental parameters(Ta,Tr,Pa,Va).
Thedata in theTables2and3was subjected to theregres-
sion analysis, and these results are listed in the Table 4.
Correlations and regressions between the 4 different
meteorological parameters and theheat-stress indices
(p<0.05) were performed as described in theStatisti-
cal analysis. Clearly, theWBGT andCETvalues were
primarily responsive to theTa (r2 = 0.86 and r2 = 0.87,
respectively).
The linear relationships for theWBGT andCETvalues to
theTr (r2=0.81 andr2=0.8, respectively) were less pro-
nounced than forTa, but they were still evident. Also clear-
ly evident was that theTeq andTovalues were primarily
responsive to theTr (r2=0.91 andr2=0.94, respectively),
while, even though thecorrelations of theTeq andTo values

IJOMEH 2017;30(3) 385

 ORIGINAL PAPER Y.-C. YANG ET AL.

withTa (r2=0.83 andr2=0.78, respectively) were less pro-

38
DI [C]

nounced than forTr, they were still evident(Table4). Am-

35
bient water vapor pressure was not as highly correlated
r2 = 0.93 (N = 29)
DI = 0.1.250 ET 5.370 with these indices, withr2values of0.440.62. It was also
32 evident that, despite thewide range in these indices, no di-
rect relationship or correlation between these indices and
29 theVa was found(Table4).

26
Workers time-weighted average heat-stress indices
The time-motion study was done to understand theopera-
tion characteristics (work time, rest time, clothing, etc.)
23
of each worker. At each of thesurveyed plants, theop-
eration characteristics were monitored during thework
shift to calculate thetime-weighted average heat level and
20
20 23 26 29 32 35 38
ET [C]
thetime-weighted average metabolic rate to which each
r2 correlation coefficient. worker was exposed. The Table 5 summarizes the val-
N number of samples.
ues of thetime-weighted average heart rate, metabolic
Fig.1. Regression between thediscomfort index(DI) rate,WBGT,CET andHSI.
and effective temperature(ET) values at each measurement
site in thestudy evaluating theoccupational heat exposure The mean values of theWBGT,CET andHSI, were28.86
of12workers at5plants in asubtropical country 1.15C, 29.031.04C, and 107.9950%, respectively.

Table4. Regression analysis results between various heat stress indices and environmental parameters in thestudy evaluating
theoccupational heat exposure of12workers at5plants in asubtropical country (N = 29)

Index Parameters Regression equation Correlation coefficient (r2)

WBGT Ta[C] 0.769Ta+3.774 0.860
Tr[C] 0.389Tr+14.689 0.810
Pa[kPa] 4.659Pa+16.861 0.490
Va[m/s] 0.654Va+30.396 0.003
CET Ta[C] 0.6754Ta+6.931 0.870
Tr[C] 0.336Tr+16.770 0.800
Pa[kPa] 4.557Pa+17.118 0.620
Va[m/s] 0.550Va+30.239 0.002
Teq Ta[C] 1.247Ta15.608 0.830
Tr[C] 0.684Tr+9.371 0.910
Pa[kPa] 7.387Pa+15.501 0.450
Va[m/s] 1.958Va+36.992 0.008
To Ta[C] 1.227Ta5.101 0.780
Tr[C] 0.703Tr+9.057 0.940
Pa[kPa] 7.394Pa+15.928 0.440
Va[m/s] 2.192Va+37.484 0.010

386 IJOMEH 2017;30(3)

 OCCUPATIONAL HOT EXPOSURE IN INDUSTRIAL WORKPLACES ORIGINAL PAPER

Table4. Regression analysis results between various heat stress indices and environmental parameters in thestudy evaluating
theoccupational heat exposure of12workers at5plants in asubtropical country (N = 29) cont.

Index Parameters Regression equation Correlation coefficient (r2)

HSIa Ta[C] 9.267Ta203.631 0.640
Tr[C] 4.530Tr65.936 0.560
Pa[kPa] 52.183Pa34.892 0.310
Va[m/s] 42.707Va+105.376 0.060

WBGT wet bulb globe temperature; HSI heat stress index; Tr mean radiant temperature. Other abbreviations as inTable3.
a
Light activity,M=213Watt (recommended bythe International Organization for StandardizationISO7243[2]).

Thelimited values of theWBGT andCET were judged
from thepublished literatures[1,2,9]. According to Broth-
erhood[13], anHSI of100is considered to be thelimit
that anaverage person can work for8h without danger of
heat strain. This data(Table5) shows that42% (5work-
ers, WBGTTWA>WBGTlimit), 42%(5workers,CETTWA>
CETlimit) and58%(7workers,HIS>100%) of thesubjects
exceeded therecommended limits of theWBGT,CET
andHSI, respectively. Thecomparison of different
heat stress indices (WBGT, CET and HSI) showed that
theWBGTindex appeared to match theCETvalue, while
theheat strain was overestimated by theHSI.
Furthermore, thelimits for acclimatized subjects, Dlimloss50
(maximum water loss of 7.5% of the body mass) are
reported in theFigure2a, where theduration of thelim-
ited exposure time is plotted against thetime-weighted
averageWBGTindex(WBGTTWA). Amodel has also been
proposed by Holmr who used theWBGTindex for pre-
dicting theallowable exposure time[40]. To make acom-
parison, theduration of thelimited exposure time predict-
ed by theHolmr model (WBGTmodel) and that predict-
ed by theHISmodel[13] are presented in theFigure2b.
In these conditions, according to theWBGTmod-
el 42% of the workers should not work continuously,
and 58% and 42% of the subjects should not continue
working in thehot environments, as predicted by theHSI
and PHS index (ISO 7933 index, 2004) [4], respectively.
However, further studies on different heat environments

Table5. Time-weighted average parameters for each worker in thestudy evaluating theoccupational heat exposure of12workers
at5plants in asubtropical country

Parameter MSD Range

HR [bpm] 9110.2 83108
MTWAa [Watt] 247.9368.75 153.52312.86
WBGTTWA [C] a
28.861.15 27.6230.81
CETTWA [C] a
29.031.04 27.5430.75
HSITWA [%]
a
107.9950.00 49.30182.68

HR heart rate; TWA time-weighted average. Other abbreviations as in Tables3and4.

a
ATWA is thetime-weighted average parameters:

A TWA 
 A  ti
t i
wheretiis theexposure time of worker at every test point andtiis thetotal exposure time of worker.

IJOMEH 2017;30(3) 387

 ORIGINAL PAPER Y.-C. YANG ET AL.

a) and more subjects should be performed. Furthermore,

550
DLE [min]

the tolerance times predicted by the ISO 7933 stan-

dard [4,41], HSI [13] and WBGT model [40] were also
440 related to thedifference between thetime-weighted aver-
DLE predicted by:
ageWBGTindex(WBGTTWA) and theWBGTlimit.
ISO 7933 (2004, Dlimloss50) [4]
Correlations and regressions between thetolerance times
330
predicted by the 3 models and thedifference between
thetime-weighted averageWBGTindex(WBGTTWA) and
220 theWBGTlimit (p<0.05) were performed as described in
theStatistical analysis. Apositive difference between
thetime-weighted averageWBGTindex(WBGTTWA) and
110 theWBGTlimit indicates that thejob may not be performed
continuously. A negative value (WBGTTWAWBGTlimit)
indicates that theheat load was below therecommended
0
27 28 29 30 31 values. For each model, theexponential regression equa-
WGBTTWA [C]
tion and correlation coefficients(r2) are given in theTa-
b)
550 ble6. For the3indices examined, theHISvalue,PHSin-
DLE [min]

dex(ISO7933,2004)[4] andWBGTmodel, these correla-

tion coefficients(r2) were0.93,0.61and0.71, respectively.
440
DLE predicted by:
HSI Correlation between the various heat-stress indices
WBGT (Holmr (1990) [40])
330 and thephysiological responses
Using thedata in theTable5, thedata of time-motion study
and theTr(equation(1)), thevalues of time-weighted av-
220
erage heat-stress indices were determined. Correlations
and regressions between thetime-weighted average heat-
stress indices and thetime-weighted averageHR were
110
carried out using Tukeys method with asignificance level
ofp<0.05using2010Microsoft Office Excel and Origin
0 version6.1software. Thecorrelation of thevarious heat-
27 28 29 30 31
WGBTTWA [C] stress indices with thetime-weighted averageHR(bpm)
WBGT wet bulb globe temperature; TWA time-weighted average; for each subject is given in theTable7.
DLE duration-limited exposure time; PHS predicted heat strain; The heart rate was not significantly correlated with
HIS heat stress indexes. these meteorological indices, withr2values of0.290.57.
Fig.2. Thetime-weighted averageWBGT index for each Theworkers duration-limited exposure times (predicted
worker in relation to theDLE estimated from various models: byISO7933[4]) obtained in this study were in accordance
a)DLEwas predicted byPHSindex; b)DLEwas predicted
byHSI andWBGT in thestudy evaluating theoccupational to some extent with those obtained by HR (r2 = 0.58).
heat exposure of12workers at5plants in asubtropical country Asignificant correlation was also found between theHR

388 IJOMEH 2017;30(3)

 OCCUPATIONAL HOT EXPOSURE IN INDUSTRIAL WORKPLACES ORIGINAL PAPER

Table6. Regression analysis results betweenWBGTTWAWBGTlimita and workersDLE predicated byHSIcriteria,ISO7933standardb

andWBGTcriteriac in thestudy evaluating theoccupational heat exposure of12workers at5plants in asubtropical country

Index Regression equation Correlation coefficient (r2)

HSI DLE[min]=74.836e0.444 (WBGTTWAWBGTlimit) 0.93
ISO7933 DLE[min]=199.098e 0.231 (WBGTTWAWBGTlimit)
0.61
WBGT DLE[min]=92.617e0.442 (WBGTTWAWBGTlimit) 0.71
a
WBGTlimit recommended bythe International Organization for StandardizationISO7243[2].
b
Obtained fromISO7933(2004)[4], Dlimloss50(maximum water loss of7.5%of thebody mass).
c
Proposed by Holmr(1998)[40].
DLE duration limited exposure time, HSI heat stress index.
Other abbreviations as in Tables4and5.

and thedifference betweenWBGTTWA andWBGT lim-
it(r2=0.86)(Table7). Furthermore, thetime-weighted
average skin temperature predicted by theISO7933[4]
foreach subject was also related to thetime-weighted
average heat-stress indices. Theanalysis of variance
(ANOVA) was carried out using Tukeys method with
asignificance level ofp<0.05using2010Microsoft Of-
fice Excel and Origin version6.1software.
The Table8shows thecorrelations of thevarious heat-stress
indices with thetime-weighted average skin temperature
predicted by theISO7933[4] for each subject. Themean
skin temperature was strongly responsive to these me-
teorological indices(WBGT,ET,CET,Teq,To)(r2=0.77
0.96), indicating that themean skin temperature increased
as themeteorological index increased. Increasing thetime-
weighted average HIS (r2 = 0.89) and SWreq (r2 = 0.93)
values also led to higher mean skin temperatures (Ta-
ble 8). In this study, there was also a highly significant
relationship between tolerance times and mean skin tem-
peratures (r2 = 0.76) (Table 8). Additionally, there was

Table7. Correlation of various heat stress indices with thetime-weighted average heart rate(HR[bpm]) for each worker in thestudy
evaluating theoccupational heat exposure of12workers at5plants in asubtropical country

Index Regression equation Correlation coefficient (r2)

WBGT[C] HR=15.625 WBGT352.641 0.33
ET[C] HR=16.994 ET381.678 0.29
CET[C] HR=16.949 CET394.695 0.29
Teq[C] HR=9.346 Teq225.869 0.56
To[C] HR=10.417 To266.979 0.57
Tsk[C] HR=40.000 Tsk1318.960 0.64
HSI[%] HR=0.222 HSI+74.203 0.84
SWreq[g/h] HR=0.019 SWreq+81.576 0.52
DLE (ISO7933)[min]
a
HR=0.072 DLE (ISO7933)+122.049 0.58
WBGTTWAWBGT limit
[C]
b
HR=4.167 (WBGTTWAWBGTlimit)+103.646 0.86

SWreq required sweat rate. Other abbreviations as in Tables36.

a
Obtained fromthe International Organization for StandardizationISO7933(2004)[4], Dlimloss50(maximum water loss of7.5%of thebody mass).
b
WBGTlimit recommended byISO7243[2].

IJOMEH 2017;30(3) 389

 ORIGINAL PAPER Y.-C. YANG ET AL.

Table8. Relationship between various heat stress indices and thetime-weighted average skin temperature(Tsk[C]) for each worker
in thestudy evaluating theoccupational heat exposure of12workers at5plants in asubtropical country

Index Regression equation Correlation coefficient (r2)

WBGT[C] Tsk=0.309 WBGT+26.500 0.80
ET[C] Tsk=0.303 ET+26.819 0.86
CET[C] Tsk=0.346 CET+25.373 0.77
Teq[C] Tsk=0.221 Teq+27.762 0.96
To[C] Tsk=0.250 To+26.648 0.94
HSI[%] Tsk=6.70510 HSI+34.689
3
0.89
SWreq[g/h] Tsk=3.79010 SWreq+35.043
3
0.93
DLE (ISO7933)[min]
a
Tsk=38.0010.645 Ln (DLE (ISO7933)) 0.76
WBGTTWAWBGT limit
[C]
b
Tsk=0.128 (WBGTTWAWBGTlimit)+35.569 0.88

Abbreviations as in Tables37.
a
Obtained fromthe International Organization for StandardizationISO7933(2004)[4], Dlimloss50(maximum water loss of7.5%of thebody mass).
b
WBGTlimit recommended byISO7243[2].

asignificant relationship between themean skin tempera-
ture and theWBGTTWAWBGTlimit(r2=0.88)(Table8).
These results indicate that skin temperature may be used
for predicting whether agiven environment is suitable for
continuous work.
DISCUSSION
This paper describes astudy of occupational heat exposure
in5plants in Taiwan (subtropical country). TheWBGTval-
ues ranged 22.535.8C for each of the work positions
studied. TheWBGTvalues observed in all locations in
the heavy-oil power plants (plants A, B, C and D), ex-
cept for therestrooms and reference locations, indicated
that heat stress was moderate, with themeanWBGT be-
ing30.73C(87.31F). Compared with theheavy-oil power
plants, in thecoal-fueled power plant(plantE), heat stress
was severe, with the mean WBGT of 34.08C (93.34F).
Notably, outside thebuilding in theshade, thedry-bulb
air temperatures varied from 29.55C to 33.99C. They
ranged 27.1329.88C (80.8385.8F), with a mean level
of 28.5C (83.3F) WBGT. The results indicate that the
heat stress that occurs naturally due to thehot climate in
coastal areas of Taiwan during thesummer months was
considerable for theworkers. Furthermore, thehot cli-
matic conditions in summer directly influence theoccur-
rence of heat-related illnesses such as heat edema, heat
syncope, heat cramps, heat exhaustion and heat stroke.
Epstein and Moran[37] reported that thevalues calculated
by theWD,DI,FITS,MDI, andWBDT indices strong-
ly correlated with theWBGTindex, withr2values that
ranged0.930.967. In this study, thecoefficients of determi-
nation(r2) evaluated for the5indices were less than those re-
ported in theliterature (withr2values of0.740.86) (Table3).
According to Wallaceetal.[42], theRHDTindex, due to its
easily attainable components, would also be very beneficial
for training situations outside themilitary. As it may be seen
from theTable3, theRHDTindex showed asignificant cor-
relation(r2=0.79) with theWBGTindex in this study.
Moran and Epstein[23] presented evidence that applying
theESI to thepooled Israeli databases from various en-
vironments showed a strong correlation (r2 = 0.960.99)
with the WBGT. Our value (r2 = 0.73) was smaller than
those found by other authors. Additionally, theETindex
was obtained by referring measured dry bulb temperature,

390 IJOMEH 2017;30(3)


wet bulb temperature and air velocity to anomographs.
In this study, there was amoderate correlation(r2=0.83)
between theET andWBGT for all of thetreatments.
A previous study performed by Yaglou and Minard [6]
showed that theWBGTindex was developed as adirect
measurement of theCETindex. Furthermore, theJapan
Society for Occupation Health(JSOH) recommendation
contained thefollowing regression equation[9]:
CET=0.786WBGT+6(C) (2)
where:
CET corrected effective temperature,
WBGT wet bulb globe temperature.
The correlation equation predicted by this study was
in good agreement with that reported by theJapan So-
ciety for Occupation Health(Figure1). In addition, So-
haretal.[24] reported astrong correlation between theDI
and theWGBTindex(r2=0.95). Our value(r2 =0.84) timate theheat load(7workers).
was lower than theone reported by Sohar et al. [24].
In addition, Epstein and Moran [37] also stated that
theDIvalues were very similar to theETvalues. This
result was demonstrated in this study, as depicted in this
Figure1(r2=0.93).
The operative temperature (To) index was modified by
Gagge and Nishi[43] and was covered byISO7730[29].
In 1979, Madsen [28] modified the equivalent tempera-
ture (Teq) index, and wrote an equation that included
theinfluence of clothing on readings. In this study, theTo
andTeq indices were also found to be significantly correlat-
ed with theWGBTindex (r2=0.93and0.94, respectively),
as depicted in theTable3.
The relationships of various heat stress criteria to
the4factors (Ta,Tr,Pa,Va) in this study for each of experi-
mental conditions are shown in theTable4. Theinfluence
of relative humidity was not significant (r2 = 0.310.62)
but these indices increased as theTa (r2 = 0.490.87)
OCCUPATIONAL HOT EXPOSURE IN INDUSTRIAL WORKPLACES ORIGINAL PAPER
andTr (r2 = 0.560.91) increased. These indices had no
direct relationship or correlation withVa(Table4).
The Table 5 depicts the means and ranges of the time-
weighted averageWBGT,CET andHISvalues during
thestudy for each subject. This data indicates that42%of
thesubjects(5workers) surpassed thelimits of theWBGT
andCET[2,9], and58%of thesubjects(7workers) exceeded
thelimits of theHIS(100%). Applying thePHSmodel to
theassessment of heat load revealed that42%of theaccli-
matized subjects were undergoing unacceptable physiologi-
cal stress (water loss or body temperature). However, amore
extensive sample of workers will enhance thescientific rele-
vance for thedifferent categories as for example age and spe-
cific professions in industry. TheWBGTindex andHSI also
allowed for thecalculation of an acceptable working time, as
depicted in theFigure2. Comparisons between theWBGT
andPHSmodels suggested that theboth criteria had similar
effects on theheat risk. In addition, acomparison of theHSI
andPHSmodels reveals that theHIScriteria tend to overes-
However, themaximum allowable exposure time predict-
ed byDlimloss50 (maximum water loss of7.5%of thebody
mass) was thehighest among the 3 methods. Moreover,
Nag and Nag [44] stated that the relationship of toler-
ance time to theWBGTvalues was useful for estimating
aprotective limit for employees working in extremely hot
environments. Theexponential relationships between
theduration-limited exposure time predicted by vari-
ous criteria and theWBGT for each subject are given in
theTable6. Asignificant correlation was found between
duration-limited exposure time and theWBGTTWA
WBGTlimitindex (r2=0.610.93). Thetolerance times pre-
dicted by these models corresponded closely to thework-
rest schedules recommended by theAmerican Confer-
ence of Governmental Industrial Hygienists (ACGIH)
for moderate work. The3criteria overestimated thetol-
erance times for light metabolic demands but underesti-
mated thetolerance times for heavy metabolic rates.
IJOMEH 2017;30(3) 391
 ORIGINAL PAPER Y.-C. YANG ET AL.
The increase in heart rate due to heat strain was on aver-
age 33 beats per minute (bpm) per 1C increase in core
temperature; heart rate was adopted inthe ISO8996[38] as
thephysiological parameter monitored. TheAmerican Con-
ference of Governmental Industrial Hygienists also recom-
mends that atime-weighted average for theday(8-hwork
shift) should not be greater than115bpm. Despite thewide
range in theWBGT, time-weighted average heart rates(83
108bpm) were well below thelevel suggested as limiting
by theACGIH. However, theheart rate response to meta-
bolic rate was much greater than that caused by environ-
mental heat [32,33]. This study (Table 7) also found that
heart rate was some extent correlated with duration-limited
exposure times (predicted bythe ISO7933[4])(r2=0.58) erence values for42%of theworkers; therefore, amore
and thedifference betweenWBGTTWA andWBGTlimit (rec-
ommended byJSOH[9])(r2=0.86). It may be used for
estimating whether agiven environment is suitable for
continuous work.
Skin temperature is avery important heat transfer factor,
and it directly affects energy transfer by convection and
theevaporation of sweat[45]. Astudy by Iampietro[46]
showed that skin temperature is areliable and sensitive
measure of theeffect of hot and cold environments in hu-
mans. Additionally, according to therecommendations of
some investigators, mean skin temperature is primarily
afunction of theambient temperature and independent of
thelevel of work[46,47]. In this study, thetime-weight av-
erage skin temperature was predicted bythe ISO7933[4].
As expected, thevalues for skin temperature predicted
bythe ISO7933[4] increased with increasing heat-stress
indices(Table8). Astrong correlation exists between skin
temperature and theSWreq index (r2 = 0.93) (Table 8). The authors would like to thank KMU colleagues (Li-Mei An,
This result suggests that this increase in theSWreqindex
was probably due to theincrease in skin temperature.
However, a study by Iampietro [46] also confirmed that
adecrease in tolerance time is proportional to an increase
in mean skin temperature. This was clearly seen in our ex-
periments, as well(r2=0.76)(Table8).
392 IJOMEH 2017;30(3)
CONCLUSIONS
To assess heat stress and strain in heavy-oil and coal-
fueled power plants, the ISO 7243 index (WBGT) and
the ISO 7933 index (the maximum allowable exposure
time, Dlim) were used for assessing workers heat loads.
TheWBGTvalues for thecoal-fueled power plant were
higher than those for theheavy-oil power plants. From
thecombination of thetime-weight averageWBGTval-
ues and metabolic rates for comparison against heat-stress
limits in published literature, it was possible to evaluate
whether agiven environment was suitable for continuous
work, and this method may also be used for constructing
safe work-rest schedules. TheWBGTlevels exceeded ref-
detailed analysis was executed, providing thedata to de-
termine an acceptable working time.
Exposure durations and theresulting heat stress predicted
bythe ISO7933[4] produced that42%of theacclimatized
subjects were undergoing unacceptable physiological stress.
Theexponential relationship between workers duration-
limited exposure time, predicted by various estimated crite-
ria, andWBGT were also found. Therelationship between
tolerance time and theWBGTvalues was useful for thein-
struction of workers in extremely hot environments and for
ensuring their health and safety. Furthermore, therelation-
ships between thevalues of various heat-stress indices and
theWBGTindex were also correlated in this study. Not-
withstanding, further studies involving different hot envi-
ronments and more subjects should be performed.
ACKNOWLEDGEMENTS
Shui-Chin Lu, Tin-Hsin Hsiao, Wen-Jeng Su, Yen-Jung Lee,
Min-Yuan Hung) and students (Yi-Jing Jang, Mei-Jing Tseng,
Shao-Yuan Hao, Fang-Yu Lin, Yei Ling Chong, Wai-Lun Lam,
Chin-Wei Kuo) for technical and editorial assistance. Finally,
theauthors acknowledge theeditors and referees for their con-
structive comments and encouragement.

REFERENCES
1. International Organization for Standardization. ISO
7243:1989. Hot environments Estimation of the heat stress
on working man, based on the WBGT-index (wet bulb globe
temperature). Geneva: The Organization; 1989.
2. International Organization for Standardization. ISO
7243:2003. Hot environments Estimation of the heat stress
on working man, based on the WBGT-index (wet bulb globe
Temperature). Geneva: The Organization; 2003.
3. International Organization for Standardization. ISO
7933:1989. Hot environments Analytical determination
and interpretation of thermal stress using calculation of re-
quired sweat rate. Geneva: The Organization; 1989.
4. International Organization for Standardization. ISO
7933:2004. Ergonomics of the thermal environment Ana-
lytical determination and interpretation of heat stress using
calculation of the predicted heat strain. Geneva: The Orga-
nization; 2004.
5. World Health Organization. Health factors involved in
working under conditions of heat stress. Technical Report
Series No 412. Geneva: The Organization; 1969.
6. Yaglou CP, Minard D. Control of heat casualties at military
training centers. AMA Arch Ind Health. 1957;16(4):30216.
7. American Conference of Governmental Industrial Hygien-
ists. Threshold limit values and biological exposure indices
for chemical substances and physical agents. Cincinnati: The
Conference; 2003.
8. National Institute for Occupational Safety and Health.
Criteria for a recommended standard. Occupational expo-
sure to hot environment. Washington: The Institute; 1986.
p. 86113.
9. Japan Society for Occupational Health. Recommendation of
occupational exposure limits (20052006). J Occup Health.
2005;47:35470.
10. Altinsoy H, Yildirim HA. Labor productivity losses over
western Turkey in the twenty-first century as a result of al-
teration in WBGT. Int J Biometeorol. 2015;59(4):46371,
https://doi.org/10.1007/s00484-014-0863-z.
OCCUPATIONAL HOT EXPOSURE IN INDUSTRIAL WORKPLACES ORIGINAL PAPER
11. Bernard TE, Iheanacho I. Heat index and adjusted tem-
perature as surrogates for wet bulb globe temperature to
screen for occupational heat stress. J Occup Environ Hyg.
2015;12(5):32333, https://doi.org/10.1080/15459624.2014.
989365.
12. Das B. Assessment of occupational health problems and
physiological stress among the brick field workers of West
Bengal, India. Int J Occup Med Environ Health. 2014;27(3):
41325, https://doi.org/10.2478/s13382-014-0262-z.
13. Brotherhood JR. Heat stress and strain in exercise and sport.
J Sci Med Sport. 2008;11(1):619, https://doi.org/10.1016/
j.jsams.2007.08.017.
14. Peters H. Evaluating the heat stress indices recom-
mended by ISO. Int J Ind Ergon. 1991;7:19, https://doi.
org/10.1016/0169-8141(91)90053-O.
15. Kampmann B, Piekarski C. The evaluation of workplaces
subjected to heat stress: Can ISO 7933 (1989) adequately
describe heat strain in industrial workplaces? Appl Erg
on.2000;31(1):5971, https://doi.org/10.1016/S0003-6870(99)
00024-1.
16. Khknen E. Comparison and error analysis of instrumenta-
tion and methods for assessment of neutral and hot environ-
ment on the basis of ISO standard. Kuopio, Finland: Kuopio
University Publications; 1993.
17. Mairiaux P, Malchaire J. Comparison and validation of
heat stress indices in experimental studies. Ergonomics.
1995;38:5872, https://doi.org/10.1080/00140139508925085.
18. Mehnert P, Malchaire J, Kampmann B, Piette A, GriefahnB,
Gebhardt H. Prediction of the average skin temperature in
warm and hot environments. Eur J Appl Physiol. 2000;82
(12):5260, https://doi.org/10.1007/s004210050651.
19. Malchaire JB. Occupational heat stress assessment by the
Predicted Heat Strain model. Ind Health. 2006;44:3807,
https://doi.org/10.2486/indhealth.44.380.
20. Malchaire J, Piette A, Kampmann B, Mehnert P, Geb-
hardtH, Havenith G, et al. Development and valida-
tion of the predicted heat strain model. Ann Occup Hyg.
2001;45:12335, https://doi.org/10.1093/annhyg/45.2.123.
IJOMEH 2017;30(3) 393
 ORIGINAL PAPER Y.-C. YANG ET AL.
21. Wang F, Kuklane K, Gao C, Holmr I. Can the PHS
model (ISO7933) predict reasonable thermophysiologi-
cal responses while wearing protective clothing in hot en-
vironments? Physiol Meas. 2011;32(2):23949, https://doi.
org/10.1088/0967-3334/32/2/007.
22. Wang F, Gao C, Kuklane K, Holmr I. Effects of various
protective clothing and thermal environments on heat strain
of unacclimated men: The PHS (predicted heat strain)
model revisited. Ind Health. 2013;51(3):26674, https://doi.
org/10.2486/indhealth.2012-0073.
23. Moran DS, Epstein Y. Evaluation of the environmental stress
index (ESI) for hot/dry and hot/wet climates. Ind Health.
2006;44:399403, https://doi.org/10.2486/indhealth.44.399.
24. Sohar E, Birenfeld C, Shoenfeld Y, Shapiro Y. Description
and forecast of summer climate in physiologically signifi-
cant terms. Int J Biometeorol. 1978;22(2):7581, https://doi.
org/10.1007/BF01552886.
25. Moran DS, Shitzer A, Pandolf KB. A physiological strain in-
dex to evaluate heat stress. Am J Physiol. 1998;275:R12934.
26. Houghton FC, Yaglou CP. Determining equal comfort lines.
J Am Soc Heat Vent Eng. 1923;29:16576.
27. Vernon HM, Warner CG. The influence of the humidity of the
air on capacity for work at high temperatures. J Hyg. 1923;32(3):
43163, https://doi.org/10.1017/S0022172400018167.
28. Madsen TL. Measurement of thermal comfort and discom-
fort: Indoor climate. Copenhagen: Building Research Insti-
tute; 1979.
29. International Organization for Standardization. ISO
7730:1983. Ergonomics of the thermal environment Ana-
lytical determination and interpretation of thermal comfort
using calculation of the PMV and PPD indices and local
thermal comfort criteria. Geneva: The Organization; 1983.
30. Aughey RJ, Goodman CA, McKenna MJ. Greater chance of
high core temperatures with modified pacing strategy during
team sport in the heat. J Sci Med Sport. 2014;17(1):1138,
https://doi.org/10.1016/j.jsams.2013.02.013.
31. Mazerolle SM, Ganio MS, Casa DJ, Vingren J, Klau J.
Is oral temperature an accurate measurement of deep
394 IJOMEH 2017;30(3)
body temperature? A systematic review. J Athl Train.
2011;46(5):56673.
32. Fuller FH, Smith PE Jr. Evaluation of heat stress in a hot
workshop by physiological measurements. Am Ind Hyg As-
soc J. 1981;42(1):327, https://doi.org/10.1080/15298668191
419316.
33. Lu S, Zhu N. Experimental research on physiologi-
cal index at the heat tolerance limits in China. Build
Environ. 2007;42:401621, https://doi.org/10.1016/j.build
env.2006.06.029.
34. International Organization for Standardization. ISO
7726:2003. Ergonomics of the thermal environment In-
struments for measuring physical quantities. Geneva: The
Organization; 2003.
35. Wenzel HG. Pulse rate and thermal balance of man during
and after work in heat as criteria of heat stress. Bull World
Health Organ. 1968;38(4):65764.
36. Bernard TE. Occupational heat stress in USA: Whither we
go? Ind Health. 2014;52(1):14, https://doi.org/10.2486/ind-
health.100.
37. Epstein Y, Moran DS. Thermal comfort and the heat stress
indices. Ind Health. 2006;44:38898, https://doi.org/10.2486/
indhealth.44.388.
38. International Organization for Standardization. ISO
8996:2004. Ergonomics of the thermal environment De-
termination of metabolic rate. Geneva: The Organization;
2004.
39. Bernard TE, Iheanacho I, Heat index and adjusted tem-
perature as surrogates for wet bulb globe temperature to
screen for occupational heat stress. J Occup Environ Hyg.
2015;12(5):32333, https://doi.org/10.1080/15459624.2014.
989365.
40. Holmr I, Parson K, Alfano G, Griefahn B. Development
and validation of the predicted heat strain model. Ann Oc-
cup Hyg. 2001;45(2):12335, https://doi.org/10.1016/S0003-
4878(00)00030-2.
41. Rowlinson S, Jia YA. Application of the predicted heat
strain model in development of localized, threshold-based

heat stress management guidelines for the construction
industry. Ann Occup Hyg. 2014;58(3):32639, https://doi.
org/10.1093/annhyg/met070.
42. Wallace RF, Kriebel D, Punnett L, Wegman DH,
WengerCB, Gardner JW, et al. The effects of continuous
hot weather training on risk of exertional heat illness. Med
Sci Sports Exerc. 2005;37(1):8490, https://doi.org/10.1249/
01.MSS.0000150018.90213.AA.
43. Gagge AP, Nishi Y. Physical indices of the thermal environ-
ment. ASHRAE J. 1976;18:4751.
This work is available in Open Access model and licensed under a Creative Commons Attribution-NonCommercial 3.0 Poland License http://creativecommons.org/
licenses/by-nc/3.0/pl/deed.en.
OCCUPATIONAL HOT EXPOSURE IN INDUSTRIAL WORKPLACES ORIGINAL PAPER
44. Nag A, Nag PK. Heat stress of women doing manipula-
tive work. Am Ind Hyg Assoc J. 1992;53:7516, https://doi.
org/10.1080/15298669291360490.
45. Bailes BK, Reeve K. Prevention of heat-related illness.
J Nurse Pract. 2007;3:1618, https://doi.org/10.1016/j.nur-
pra.2006.12.001.
46. Iampietro PF. Use of skin temperature to predict tolerance
to thermal environments. Aerosp Med. 1971;42(4):3969.
47. Pandolf KB, Goldman RF. Convergence of skin and rectal
temperatures as a criterion for heat tolerance. Aviat Space
Environ Med. 1978;49(9):1095101.
IJOMEH 2017;30(3) 395
 Reproduced with permission of copyright owner.
Further reproduction prohibited without permission.