Preprint version of the paper published in Building and Environment 01/2013; 59:644-653 DOI: 10.1016/j.buildenv.2012.10.001 The influence of clothing distribution and local discomfort on the assessment of global thermal comfort Antonio Simone, Sabrina Della Crociata, Francesco Martellotta Dipartimento di Scienze dell’Ingegneria Civile e dell’Architettura Politecnico di Bari, via Orabona 4 - 70125 Bari (Italy). Abstract A field study was carried out in a hypermarket located in Southern Italy in order to evaluate the environmental comfort in large-scale retail trade buildings. Global and local thermal comfort conditions were studied by collecting exposure data and subjective responses of employees using both questionnaires and simultaneous physical measurements. In a few cases discrepancies appeared between subjective ratings and the corresponding PMV-index, suggesting that the latter could be unable to reliably estimate thermal comfort when clothing insulation is unevenly distributed on human body. Analysis of collected data confirmed this, with particular reference to female workers wearing skirts. In addition, such effects may be further emphasized by local thermal discomfort, which may finally influence global thermal perception. Analysis of the results showed that cold floor and radiant temperature asymmetry caused by warm ceiling played a major role in emphasizing the negative effects due to uneven clothing distribution, confirming that this particular combination of events may bias PMV model accuracy. 1. Introduction Fanger’s PMV-index [1], described in ISO 7730 [2], is the most used parameter to evaluate thermal comfort. Its reliability was widely demonstrated in air-conditioned environments [3-4] including, in particular, hypermarkets [5]. However, only a few researches discuss how this index works when environmental conditions deviate from reference model. Several studies were carried out in order to evaluate clothing insulation (Icl), taking into account the aspects which could modify the Icl-value (e.g. air movement, vapour resistance, further insulation provided by chair seat). However, in a lot of cases (e.g. for steady and light work activities and for more frequently worn clothing combination) the improvement in Icl accuracy is not as high to justify the more cumbersome calculations[6].