Energy and Buildings 139 (2017) 732–746 Contents lists available at ScienceDirect Energy and Buildings j ourna l ho me pa g e: www.elsevier.com/locate/enbuild Indoor air quality and thermal comfort optimization in classrooms developing an automatic system for windows opening and closing Francesca Stazi a,∗ , Federica Naspi b , Giulia Ulpiani c , Costanzo Di Perna c a Dipartimento di Scienze, Ingegneria della Materia, dell’Ambiente e Urbanistica (SIMAU), Facoltà di Ingegneria, Università Politecnica delle Marche, Via Brecce Bianche, 60131 Ancona, Italy b Dipartimento di Ingegneria Civile, Edile e Architettura (DICEA), Facoltà di Ingegneria, Università Politecnica delle Marche, Via Brecce Bianche, 60131 Ancona, Italy c Dipartimento di Ingegneria Industriale e Scienze Matematiche (DIISM), Facoltà di Ingegneria, Università Politecnica delle Marche, Via Brecce Bianche, 60131 Ancona, Italy a r t i c l e i n f o Article history: Received 13 June 2016 Received in revised form 15 November 2016 Accepted 7 January 2017 Available online 13 January 2017 Keywords: Natural ventilation Indoor air quality Thermal comfort Window opening Adaptive actions School building automation a b s t r a c t Thermal comfort and indoor air quality in school classrooms are essential requirements to promote students’ productivity and reduce health symptoms. This paper presents the development of an automatic system for window openings, based on thermal comfort and indoor air quality correlations. The research was carried out in two adjacent classrooms. The initial phase aimed at assessing environmental conditions in classrooms, testing objective and subjective comfort models and establishing trigger parameters for window opening events; the second phase regarded the implementation of an adaptive control algorithm in an automatic system piloting windows with the aim of maintaining a satisfactory environment both in terms of IAQ and thermal comfort. The main results show that: (1) the IAQ is a relevant issue in school classrooms, because students usually suffer high CO 2 levels; (2) the stronger driving force for undertaking adaptive actions is thermal comfort, while the need to improve the air quality is a secondary constraint; (3) the mechanized system ensures a good quality in terms of IAQ, thermal comfort and users’ satisfaction. © 2017 Elsevier B.V. All rights reserved. 1. Introduction The main target to be achieved in school classrooms is preserv- ing students’ attention, efficiency and health while attending the lessons. From an environmental perspective it can be translated into reaching and maintaining a satisfactory perception in terms of air temperature and indoor air quality. Indeed, students spend almost all day in indoor environments and mainly inside class- rooms with high occupant density [1–3]. Several studies show that inadequate ventilation rates can lead to a decrease in users’ perfor- mances and growing absenteeism [4,5] and that adaptive actions, to restore comfortable conditions, usually occur when the situation is already critical, especially in relation to CO 2 concentration [6,7]. During the recent years different adaptive control algorithms (named ACA in the paper as in [8]) have been developed with the aim of predicting human behaviours, usually in relation to a spe- cific action (e.g. windows or blind use, light-switching) [9–12]. The ∗ Corresponding author. E-mail addresses: f.stazi@univpm.it (F. Stazi), f.naspi@staff.univpm.it (F. Naspi), g.ulpiani@pm.univpm.it (G. Ulpiani), c.diperna@univpm.it (C. Di Perna). “adaptive” term means that these algorithms are based on the adap- tive comfort theory, stating that: “if a change occurs such as to produce discomfort, people react in ways which tend to restore their comfort” [13]. This principle focuses on the human-building interaction, since it asserts that if people perceive the ambient as uncomfortable, they will try to improve their thermal comfort, operating both on building controls (e.g. using doors, windows, blinds and fans) and on their personal condition (e.g. changing their clothes or taking hot or cold drinks). Many authors demonstrated that the adaptive model is the most proper and realistic approach for naturally ventilated buildings. Humphreys and Nicol, assessing the validity of the Fanger steady-state comfort model [14,15], con- cluded that a ISO 7730 [16] compliant approach is more likely to bring to erroneous evaluations of thermal discomfort because the model poorly reflects the human thermal adaptation [17,18]. Simi- lar results were obtained by de Dear and Brager [19] and confirmed by many other researches [20,21]. This active-user comfort approach is based on the relationship between the indoor temperature and the comfort one. Humphreys and Nicol assessed that the comfort temperature is a function of the outdoor average temperature and proposed two equations to derive it in free-running buildings [22]. The proposed equations http://dx.doi.org/10.1016/j.enbuild.2017.01.017 0378-7788/© 2017 Elsevier B.V. All rights reserved.