Accepted Article SAFE: Smart helmet for Advanced Factory Environment Angelo Altamura 1 | Francesco Inchingolo 1 | Gianvito Mevoli 1 | Pietro Boccadoro 1,2 1 Department of Electrical and Information Engineering (DEI), Politecnico di Bari, Bari, Italy, Email: a.altamura6@studenti.poliba.it, f.inchingolo1@studenti.poliba.it, g.mevoli1@studenti.poliba.it, pietro.boccadoro@poliba.it 2 CNIT, Consorzio Nazionale Interunivesitario per le Telecomunicazioni, Politecnico di Bari, Bari, Italy. Summary Safety is a leading theme in industrial applications such as constructions sites and production chains. Several solutions are already in use to protect workers from structural failures or traumatic events. These work instruments have proven their effectiveness, lowering the potentially hurtful, if not fatal, consequences of accidents on workplaces. In this context, the Industrial Internet of Things (IIoT) paradigm can boost the development of smart wearable systems that do not impede movements and/or operations. This work proposes Smart helmet for Advanced Factory Envi- ronment (SAFE), an autonomous and integrated wearable IIoT device. Thanks to integrated sensors, SAFE is able to perform continuous monitoring activities of the worker and to promptly send of alerts in case of alarms and/or accidents. KEYWORDS: Industrial Internet of Things, wearable, safety, eHealth 1 INTRODUCTION AND MOTIVATION Safety is probably one of the most continuously investigated field in several applications, from driving to health-care 1,2,3,4,5,6 . Over time, workplace safety procedures and appliances have been increasingly enhanced to ensure the state-of-the-art require- ments 1,2,3,4 . This is not just a matter of comfort (i.e., lighting or seats) as it can be applied to evacuation procedures, as well as the safety clothing 1 . These solutions are of even higher relevance in those workplaces that require the physical commitment of the worker, exposing him/her to high risks of cuts (as in sawmills and forges), crushing (as for those working with the presses), bumps and falls (as in construction sites and scaffolding) 2 . A way to improve worker safety 3,4 is to enrich the usual equipment with smart solutions, able to detect and report accidents that can occur in construction sites, in the depths of wells and tunnels or on freight elevators. For instance, wearing a helmet in these situations is mandatory in order to absorb shocks and other mechani- cal actions 1 . Specifically, in the context of constructions sites, a survey on the use of safety helmets has shown the importance of comfort in the security work suits. A study conducted in the U.K. demonstrated that in more than 86% of the sample construction sites, the usage of safety helmets was not mandatory but self-help 5 . The IIoT is now unleashing thrilling opportunities to fur- ther improve wearable devices, safety protocols and eHealth. This study proposes SAFE, an integrated smart object exploiting wireless communication technologies, leveraging the potential of the IIoT paradigm. It creates a security shelter that provides continuous monitoring of the worker’s biometrics together with hazardous gas and flame sensors. The system (Figure 1 ) is dis- tributed and crowd-sourced by design, since it is equipped with signaling devices allowing both workers and administrations to promptly react to critical situations. The system proposes to manage generic hazardous cases. Considering mining safety, differ- ent available works 3,6 use specific sensors to detect hazardous environment situations without taking into account the workers’ vital state. On the other hand, in industrial safety there are different systems that provide worker fatigue detection 2 , but do not consider the environmental state. When compared to other solutions (Table 1 ), SAFE represents a proof-of-concept of a sys- tem that enables continuous monitoring of both vital parameters and environmental state without begin invasive or affecting the This article is protected by copyright. All rights reserved. This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as doi: 10.1111/itl2.86