REVIEW J. Malchaire á B. Kampmann á G. Havenith P. Mehnert á H. J. Gebhardt Criteria for estimating acceptable exposure times in hot working environments: a review Received: 14 October 1999 / Accepted: 2 November 1999 Abstract Objectives: To revise the criteria used in the present ``Required Sweat Rate'' standard ISO 7933 (1989) for the prediction of the maximum duration of work in hot environments. Methods: Review of the lit- erature and in particular, of the bases for the present criteria. Results: A new method is proposed, to take into account the increase in core temperature associated with activity in neutral environments. The prediction of maximum wetness and maximum sweat rates are re- vised, as well as the limits for maximum water loss and core temperature. Conclusion: An improved set of maximum values and limits is described, to be used in the revised version of the ISO 7933 standard. Due to the major modi®cations to the ``Required Sweat Rate'' in- dex and in order to avoid any confusion, it is suggested that the revised model be renamed the ``Predicted Heat Strain'' (PHS) model. Key words Heat stress á Exposure limits á Physiological limits Introduction The ISO 9886 1992 standard (ISO 9886 1992) entitled ``Ergonomics of the working environment: physiologi- cal measurements'', proposes limit values for core temperatures and dehydration and heart rates for people working in hot environments. These limits are applicable to individuals, and dierent values are pro- posed, depending on whether or not heart rate and core temperature are monitored simultaneously and continuously. The point of view for setting limits must be dierent for heat stress indices such as those in the ``Required Sweat Rate'' standard (ISO 7933 1989). Indeed, these indices are intended to predict the risk of heat disorders from climate characteristics, the clothing of the subjects and their average metabolic rate, using predicted phys- iological responses as strain indicators. The indices as- sume an identical physiological response from all persons working under the same conditions and, there- fore, these predictions are applicable only to a group of workers. The reaction to heat varies considerably from one subject to another (Havenith 1997), thus the limit values for the strain indicators must be set to protect ``most'' of the workers. The percentage of protected workers might be set at 90%, 95% or 99%. Ideally, it should vary according to the severity of the eect that is being considered. Two eects are usually envisaged: dehydration and thermo-regulatory disturbances such as heat stroke. Obviously, the former (water loss of 7.5% of body mass, for instance), although severe, is less dangerous than the Int Arch Occup Environ Health (2000) 73: 215±220 Ó Springer-Verlag 2000 J. Malchaire (&) Unite HygieÁne et Physiologie du Travail, Universite Catholique de Louvain, Clos Chapelle-aux-Champs 3038, 1200 Brussels, Belgium e-mail: malchaire@hytr.ucl.ac.be Tel.: +32-2-7643229; Fax: +32-27643954 B. Kampmann Institut fuÈr Arbeitswissenschaften der RAG Aktiengesellschaft, HuÈlshof, Dortmund, Germany G. Havenith 1 Thermal Physiology Group, Department of Work Environment, TNO-Human Factors Research Institute, Soesterberg, The Netherlands P. Mehnert Institut fuÈr Arbeitsphysiologie an der UniversitaÈt Dortmund, Dortmund, Germany H. J. Gebhardt Institut fuÈr Arbeitsmedizin, Sicherheitstechnik und Ergonomie, Wuppertal, Germany Present address: 1 Human Thermal Environments Laboratory, Loughborough University of Technology, Loughborough, Leicestershire, UK