Abstract
The amount of data extracted from production processes has increased exponentially due to the proliferation of sensing technologies. When processed and analyzed, data can bring out valuable information and knowledge from manufacturing process, production system and equipment. In industries, equipment maintenance is an important key, and affects the operation time of equipment and its efficiency. Thus, equipment faults need to be identified and solved, avoiding shutdown in the production processes. Machine Learning (ML) methods have been emerged as a promising tool in Predictive Maintenance (PdM) applications to prevent failures in equipment that make up the production lines in the factory floor. However, the performance of PdM applications depends on the appropriate choice of the ML method. The aim of this paper is to present a systematic literature review of ML methods applied to PdM, showing which are being explored in this field and the performance of the current state-of-the-art ML techniques. This review focuses on two scientific databases and provides a useful foundation on the ML techniques, their main results, challenges and opportunities, as well as it supports new research works in the PdM field.

Abstract
The industry faces more and more the challenge of deploying and taking advantage of evidence-based strategic decisions to enhance profit gain. In this research, the possibility of having a fully integrated system composed by a simulator and an IoT platform with the capability of collecting real-time data from the shop floor and returning performance indicators to support decision making is evaluated. The suggested approach involves a Manufacturing Executing System (MES) producing a production schedule, an IoT Platform composed by a message broker and a real-time database, a Simulator including simulation software and a wrapper, and a user application serving as an interface between the user and the IoT Platform and Simulator integrated system. A detailed analysis of the functionalities and integration of the Simulator and the IoT Platform will also be explored. To evaluate the approach, one use case of a production line in the automotive industry is used. The application of the integrated IoT Simulation system permits its validation and consequent future work. © 2019, IEOM Society International.

Abstract
Additive Manufacturing (AM) is one of the most trending production technologies, with a growing number of companies looking forward to implementing it in their processes. Producing through AM not only means that there are no supplier lead times needed to account for, but also enables production closer to the end customer, reducing then the delivery time. This is especially true for companies with a wide range of low and variable demand products. This paper proposes a mixed integer linear programming (MILP) model for the optimal design of supply chains facing the introduction of AM processes. In the addressed problem, the 3D printers allocation to distribution centers (DC), that will make or customize parts, and the Suppliers-DC-Customers connections for each product need to be defined. The model aims at minimizing the supply chain costs, exploring the trade-offs between safety stock and stockout costs, and between buying and 3D printing a part. The main relevant characteristics of this model are the introduction of stock service levels as decision variables and the use of a linearization of the cumulative distribution function to account for demand uncertainty. A real-world problem from a maintenance provider is solved, showing the applicability of the model. © 2019, IEOM Society International.

Abstract
The continuous adoption of Additive Manufacturing (AM) can enhance Supply Chain's (SC) effectiveness, adaptability and competitiveness. AM allows for a decentralized SC, bringing production centres nearer to customers, increasing products availability and decreasing inventory level and lead time. However, the integration of SC and AM brings difficulties, leading to the need of a completely new SC design. This paper proposes an optimization model supporting the design of spare parts SCs operating under a Make-To-Order (MTO) strategy. The proposed approach considers the decision of deploying productive resources (3D printers) in locations of a spare parts SC. The problem is represented as a combination of the p-median and location-allocation optimization models, which are solved using a Mixed Integer Linear Programming (MILP). The approach is tested in two scenarios from a real-world use case of an elevator maintenance service provider. Obtained results demonstrated the promising capabilities of the proposed approach for handling the new design challenges arising from the forthcoming widespread use of 3D printers in manufacturing SCs.

Abstract
In the current industrial revolution, additive manufacturing (AM) embodies a promising technology that can enhance the effectiveness, adaptability, and competitiveness of supply chains (SCs). Moreover, it facilitates the development of distributed SCs, thereby enhancing product availability, inventory levels, and lead time. However, the wide adoption of AM in industrial SCs creates various challenges, leading to new difficulties for SC design. In this context, this article proposes a new design approach to AM SCs using optimization methods. More specifically, the proposed approach, comprising the p-median and mixed-integer linear programming models, considers the decision of deploying productive resources (3-D printers) in specific locations of generic spare part SCs. The approach was evaluated in a real-world use case of an elevator maintenance service provider. The obtained results demonstrated the promising capabilities of the proposed design approach in managing the challenges arising from the forthcoming widespread use of 3-D printers in manufacturing SCs.

Abstract
Distributed manufacturing systems represent a new paradigm in the industrial context, supported by new technologies provided by industry 4.0. In this paper, a model for dynamic allocation of Production Orders (PO) in the context of distributed additive manufacturing systems is proposed. The scheduling model performs a local optimization of PO allocation considering a production times forecasting model, fed by system state data obtained by means of an IoT platform, and transportation real-time data. A simulation-based experiment was conducted in a test case with real and simulated data collected from an elevator spare parts provider in Brazil. A significant reduction of 77.94% of the Average Waiting Time (AWT) was obtained, allowing for an increased efficiency of the additive manufacturing system, which supports the forthcoming pilot application. Copyright (C) 2020 The Authors.