作者: Fernando Piedade , Marcia Baptista , Paulo Chaves
DOI: 10.1016/J.PROMFG.2020.02.030
关键词: Factory (object-oriented programming) 、 Welding 、 Manufacturing engineering 、 Computer science 、 Universal Plug and Play 、 Digital manufacturing 、 Wireless sensor network 、 Process (engineering) 、 Quality (business) 、 Enterprise resource planning
摘要: Abstract The mass production of metal molds for glass bottles is a highly automated and technologically demanding sector activity. Mold manufacturing involves several operations one the most critical related to mold surface welding. main goal here insert more durable into specific areas create regions with enhanced characteristics (e.g. improved wear resistance) in order improve quality durability mold. Such an operation typically three steps, namely: i) preheating, ii) manipulation iii) Overall, this rather complex process, tight control monitoring its variables. Our industrial partner, medium-size manufacturer bottles, already had extensively process. However, there were frequent reports processing flaws during welding stage deposition, fissures, pores) being lost some cases. reasons behind reported difficult discern from available sensors methods. From our early assessment it became clear that range was insufficient communication between machines inside cell either too simple or disconnected rest factory. This resulted absence integrated record In view this, necessary develop implement digitalization plan, which we describe paper, responsible As result planning, additional introduced cell, pressure, temperature, humidity, flow, voltage. intent have variables could be used diagnose and/or prognosticate problems Also, since lacked integration, network established using IoT technologies supported by PlugThings Framework together Universal Plug Play (UPnP) technology. PlugThings, platform developed partner iTime, acted as mediator sensor Web server. also allowed perform vertical integration on Enterprise Resource Planning (ERP) system. Finally, initiative enabled access charting tools data form XML files. Currently, system undergoing exhaustive tests before expanding other cells shop floor. Future steps will involve use analytics artificial intelligence predict failures process better streamline production.