Chapter
To read the full-text of this research, you can request a copy directly from the authors.

Abstract

Besides pursuing the economic goals of low costs and high profits, companies are becoming more and more aware of the environmental and social impact of their actions. Companies striving for the integrated optimization of environmental and economic perspectives within their production processes, need to be supported by tools helping to understand the effects of the decision making process. In this context, this paper describes the Artificial Intelligence developed for a Decision Support System (DSS) which enables the early identification of problems occurring on manufacturing. The decision making process beneath the DSS starts from the aggregation of production lines sensors data in Key Performance Indicators (KPI). The data are then processed by means of an Artificial Neural Networks (ANN) based knowledge system which enables to suggest preventive maintenance interventions. The proposed maintenance activities, elaborated throughout a scheduling engine, are integrated within the weekly production schedule, according to the selected optimization policy.

No full-text available

Request Full-text Paper PDF

To read the full-text of this research,
you can request a copy directly from the authors.

... Cost estimation and maintenance planning 21 [5,[20][21][22][23][24][25][26][27][28][29][30][31][32][33][34][35][36][37][38][39]] Joint scheduling and planning 14 [40][41][42][43][44][45][46][47][48][49][50][51][52][53][54] Multi-state and multi-component systems optimization 12 [55][56][57][58][59][60][61][62][63][64][65][66] Electronics 2021, 10, x FOR PEER REVIEW 5 of 20 ...
... The problem was formulated and solved in the semi-Markov decision process framework in order to minimize the long-run expected average cost per unit time. Cinus et al. [46] propose a decision support system that processes sensor data and Key Performance Indicators (KPIs) using an artificial neural network (ANN)-based knowledge system and integrates the maintenance actions within the weekly production schedule. Mourtzis et al. [47,48] propose an augmented reality mobile application, interfaced with a shop-floor scheduling tool, in order to enable the operator to decide on immediately calling AR remote maintenance or scheduling maintenance tasks for later along with production tasks. ...
Article
Full-text available
Decision-making for manufacturing and maintenance operations is benefiting from the advanced sensor infrastructure of Industry 4.0, enabling the use of algorithms that analyze data, predict emerging situations, and recommend mitigating actions. The current paper reviews the literature on data-driven decision-making in maintenance and outlines directions for future research towards data-driven decision-making for Industry 4.0 maintenance applications. The main research directions include the coupling of decision-making with augmented reality for seamless interfacing that combines the real and virtual worlds of manufacturing operators; methods and techniques for addressing uncertainty of data, in lieu of emerging Internet of Things (IoT) devices; integration of maintenance decision-making with other operations such as scheduling and planning; utilization of the cloud continuum for optimal deployment of decision-making services; capability of decision-making methods to cope with big data; incorporation of advanced security mechanisms; and coupling decision-making with simulation software, autonomous robots, and other additive manufacturing initiatives.
... Najjar, and Jacobsson, 2013;Xu et al., 2013;Guo et al., 2013;Mendes et al., 2014; de Jonge et al., 2015;Gopalakrishnan et al., 2015;Xu et al., 2015;Terkaj et al., 2015;Wan et al., 2015;Nadj et al., 2016;Yildirim et al., 2016;Said et al., 2016;Fitouri et al., 2016;Ghosh et al., 2017 Reliability-and Degradation-based Decision Making 12Le et al., 2014;Hong et al., 2014;Song et al., 2014;Tang et al., 2015a;Tang et al., 2015b;Do et al., 2015;Lin et al., 2015;Park et al., 2016;Drumheller et al., 2017;He et al., 2017;Animah, and Shafiee, 2017; Zan et al., 2018 Joint Optimization 12 Lee, andNi, 2013;Kouedeu et al., 2015;Gan et al., 2015;Jafari, and Makis, 2015;Jiang et al., 2015a;Van Horenbeek, and Pintelon, 2015;Cinus et al., 2016;Bousdekis et al., 2017a;Bousdekis et al., 2017b;Gu et al., 2017;Mourtzis et al., 2017 Multi-State and Multi-Component Systems Optimization11 ...
Article
Full-text available
The emergence of Industry 4.0 has led to a wide use of sensors which have facilitated manufacturing operations. Predictive maintenance has significantly benefited from these technological advancements with the use of real-time detection and prediction algorithms regarding future failures. During the last years, there is also an increasing interest on decision making algorithms triggered by failure predictions. The current paper reviews the literature on decision making in predictive maintenance in the context of smart manufacturing. Moreover, it discusses the results, identifies the existing research gaps and outlines a research agenda on the field.
Article
Full-text available
This paper introduces the concepts of neural networks and presents an overview of the applications of neural networks in decision support systems (DSS). Neural networks can be viewed as supporting at least two types of DSS: data driven and model-driven. First, neural networks can be employed as data analysis tools for forecasting and prediction based on historical data in a data-driven DSS. Second, neural networks also can be viewed as a class of quantitative models to be used in a model-driven DSS. After describing the basics of neural networks, we present selected applications of neural networks in DSS. We then describe a web-based DSS built by us that employs a neural network. This DSS has been built to assist a Hollywood decision maker in making decisions on a movie’s parameters. The paper concludes with a list of issues to consider in employing a neural network for a DSS application.
Chapter
Full-text available
A number of retail firms use a “private-label” strategy in which merchandise is sold under a brand name exclusive to the retail firm, but manufactured by one or more independent vendors. While offering a number of benefits, this approach also poses a different set of supply chain challenges than manufacturer-brand-based retailing, in that the retail firm must take a more active role in organizing and coordinating the planning and materials management activities in a supply base that is often dispersed and heterogenous. This chapter describes a methodology for planning capacity commitments, scheduling shipments, and managing inventory for an assortment of private-label retail merchandise produced by multiple vendors. The vendors differ in their lead time requirements, costs, and production flexibility. Product demand is uncertain, and fluctuates over time. We develop an optimization model to choose the production commitments that maximize the retailer’s expected gross profit, given market demand forecasts and vendors’ capacity and flexibility constraints. The model has been incorporated into a PC-based decision support system called the Sourcing Allocation Manager (SAM). This was developed in collaboration with supply chain planners at a global retailer of seasonal and fashion merchandise, and has been tested by buyers at two major retailers.
Article
Full-text available
More than twelve years have elapsed since the first public release of WEKA. In that time, the software has been rewritten entirely from scratch, evolved substantially and now accompanies a text on data mining [35]. These days, WEKA enjoys widespread acceptance in both academia and business, has an active community, and has been downloaded more than 1.4 million times since being placed on Source-Forge in April 2000. This paper provides an introduction to the WEKA workbench, reviews the history of the project, and, in light of the recent 3.6 stable release, briefly discusses what has been added since the last stable version (Weka 3.4) released in 2003.
Article
Full-text available
PyBrain is a versatile machine learning library for Python. Its goal is to provide flexible, easyto-use yet still powerful algorithms for machine learning tasks, including a variety of predefined environments and benchmarks to test and compare algorithms. Implemented algorithms include Long Short-Term Memory (LSTM), policy gradient methods, (multidimensional) recurrent neural networks and deep belief networks. © 2010 Tom Schaul, Justin Bayer, Daan Wierstra, Yi Sun, Martin Felder, Frank Sehnke, Thomas Rückstieß and Jürgen Schmidhuber.
Conference Paper
Besides pursuing the economic goals of low costs and high profits, companies are becoming more and more aware of the environmental and social impact of their actions. Companies striving for the integrated optimization of environmental and economic perspectives within their production processes, need to be supported by tools helping to understand the effects of the decision making process. In this context, this paper describes a Decision Support System (DSS) enabling the early identification of problems occurring on manufacturing lines thus suggesting related recovery actions, together with the potential repercussions of their adoption, at economic and environmental level. The decision making process beneath the DSS starts from the aggregation of production lines sensors data in Key Performance Indicators (KPI). The data are then processed by means of an Artificial Neural Networks (ANN) based knowledge system which enables to suggest preventive maintenance interventions. The proposed maintenance activities, elaborated throughout a scheduling engine, are integrated within the weekly production schedule, according to the selected optimization policy. Preliminary tests have been carried out in manufacturing plants of IKEA industries and Brembo.
Article
Robotic learning on real hardware requires an efficient algorithm which minimizes the number of trials needed to learn an optimal policy. Prolonged use of hardware causes wear and tear on the system and demands more attention from an operator. To this end, we present a novel black-box optimization algorithm, Reward Optimization with Compact Kernels and fast natural gradient regression (ROCK⋆). Our algorithm immediately updates knowledge after a single trial and is able to extrapolate in a controlled manner. These features make fast and safe learning on real hardware possible. The performance of our method is evaluated with standard benchmark functions that are commonly used to test optimization algorithms. We also present three different robotic optimization examples using ROCK⋆. The first robotic example is on a simulated robot arm, the second is on a real articulated legged system, and the third is on a simulated quadruped robot with 12 actuated joints. ROCK⋆ outperforms the current state-of-the-art algorithms in all tasks sometimes even by an order of magnitude.
Article
A manager tries to put together the various resources under his control into an activity that achieves his objectives. A model of his operation can assist him but probably will not unless it meets certain requirements. A model that is to be used by a manager should be simple, robust, easy to control, adaptive, as complete as possible, and easy to communicate with. By simple is meant easy to understand; by robust, hard to get absurd answers from; by easy to control, that the user knows what input data would be required to produce desired output answers; adaptive means that the model can be adjusted as new information is acquired; completeness implies that important phenomena will be included even if they require judgmental estimates of their effect; and, finally, easy to communicate with means that the manager can quickly and easily change inputs and obtain and understand the outputs. Such a model consists of a set of numerical procedures for processing data and judgments to assist managerial decision making and so will be called a decision calculus. An example from marketing is described. It is an on-line model for use by product managers on advertising budgeting questions. The model is currently in trial use by several product managers.
Article
This article discusses how Predictive Maintenance (PdM) technologies can be successfully applied to both GMP and non-GMP utility systems in the pharmaceutical industry. The discussion also demonstrates the clear benefits of PdM, including the use of a proactive approach to maintenance.
Conference Paper
Energy consumption of a machine tool increases dramatically as reliability decreases, which indicates performance degradation. From the perspective of near-zero downtime and energy conservation, preventive maintenance is usually carried out. This research aims at modeling the energy consumption of machine tools after they are maintained preventively. The relationship between reliability and energy consumption is modeled based on historical data and then the energy consumption of a machine tool is calculated. Reliability is modeled by Weibull distribution, the parameters of which are estimated utilizing the Maximum Likelihood Estimation method. Reliability after preventive maintenance is modeled by employing Malik's Proportional Age Reduction model. The carbon emission is assessed referring to the standard developed by the Climate Registry in North America.
Article
This chapter surveys the state of the art of decision support systems (DSSs) for operations management (OM) with the intent to highlight the important role that DSSs play in facilitating decision processes in OM. Organized along the temporal dimension of OM decisions, our discussion starts with DSSs for long-term OM decisions including operations strategy, capacity planning, location planning, product/process design and design/manufacturing integration. This is followed by a discussion on DSSs for medium- and short-term OM decisions on production planning and control that includes systems such as communication- oriented production information and control systems, material requirements planning, manufacturing resources planning, distribution requirements planning, enterprise resources planning, and operations/project scheduling. After a brief discussion on DSSs for service operations, we conclude our book chapter by providing some perspectives on future development in DSSs for OM.
Article
Manufacturing industries are in the midst of a new technological revolution, characterised by the increasing advances and application of information and communication technology for both information processing and automatic control. A unique manufacturing methodology for non-destructive, real-time process optimization and quality certification has been developed. This includes the design of a decision support system (DSS) for controlling the optimized process. The DSS comprises the capability of supporting both the process operator and manager in the decision-making process, allowing them to take advantage of the degrees of freedom of the physical system to make it work in a way where the optimized process control variables are under statistical control, resulting in optimized output factors that, in turn, secure higher productivity and improved quality.
Article
From the Publisher:How data mining delivers a powerful competitive advantage!Are you fully harnessing the power of information to support management and marketing decisions? You will,with this one-stop guide to choosing the right tools and technologies for a state-of-the-art data management strategy built on a Customer Relationship Management (CRM) framework. Authors Alex Berson,Stephen Smith,and Kurt Thearling help you understand the principles of data warehousing and data mining systems,and carefully spell out techniques for applying them so that your business gets the biggest pay-off possible. Find out about Online Analytical Processing (OLAP) tools that quickly navigate within your collected data. Explore privacy and legal issues. . . evaluate current data mining application packages. . . and let real-world examples show you how data mining can impact — and improve — all of your key business processes. Start uncovering your best prospects and offering them the products they really want (not what you think they want)!How data mining delivers a powerful competitive advantage! Are you fully harnessing the power of information to support management and marketing decisions? You will,with this one-stop guide to choosing the right tools and technologies for a state-of-the-art data management strategy built on a Customer Relationship Management (CRM) framework. Authors Alex Berson,Stephen Smith,and Kurt Thearling help you understand the principles of data warehousing and data mining systems,and carefully spell out techniques for applying them so that your business gets the biggest pay-off possible. Find out about Online Analytical Processing (OLAP) tools thatquickly navigate within your collected data. Explore privacy and legal issues. . . evaluate current data mining application packages. . . and let real-world examples show you how data mining can impact — and improve — all of your key business processes. Start uncovering your best prospects and offering them the products they really want (not what you think they want)!
Article
The paper describes the fundamental concepts of maintenance and the role that information technology can play in the support of maintenance activities. Function-Behaviour-State modelling is used to describe faults and deterioration of mechanisms in terms of user perception and measurable quantities. Fault diagnosis, repair and maintenance strategies are explained briefly. Model-based Maintenance is discussed in terms of inspection, monitoring, diagnosis and planning, based on functional, behavioural and state models. Some examples of mechanisms illustrate the advantage of using computer-based models as a reference for fault diagnosis. Underlying techniques like qualitative physics, tolerance analysis, and dynamic simulation are discussed. The proposed method is aiming at functionally robust designs, reduction of preventive maintenance cost and more effective fault diagnosis.
Article
Value-based Management (VBM) concepts are prevalent in theory and practice since shareholder value creation is commonly considered the paramount business goal. However. VBM mainly applies data-driven concepts to support decision-making, disregarding model-driven approaches. This paper develops a comprehensive approach to designing model-driven DSS for VBM. First, we derive a conceptual architecture for Integrated Business Planning (IBP) as the foundation for a model-driven approach to VBM. Second, we present a unified modeling approach for value-based performance and risk optimization that implements Value Added (xVA) performance metrics and applies robust optimization methods to mitigate risk impact.
Article
Clothing manufacturers’ direct investment and joint ventures in developing regions have grown rapidly in the last few decades. An inappropriate decision on site selection can cause adverse affects on productivity and increase manufacturing and logistical costs. Manufacturers often face with difficulties during the decision-making process, due to vague and subjective considerations. The process is hampered with variables that cannot be represented in terms of objective values, such as country risk and community facilities. The decision on plant location thus relies mostly on the subjective intuition and assessment of the manufacturer. This paper explains the development of a decision-making model for a clothing plant location, using the feed-forward neural network with an error back-propagation (EBP) learning algorithm and a fuzzy analytical hierarchy process (FAHP). Significant variables in the selection of a clothing plant location will be identified and input into the proposed decision-making model. A suitability index was computed for each site. The sites with higher suitability indices were more appropriate for the establishment of a clothing plant. It was also identified that those variables considered on plant location with lower evaluation weights can be neglected without affecting the prediction accuracy of the proposed decision model .
Article
This paper rigorously establishes that standard multilayer feedforward networks with as few as one hidden layer using arbitrary squashing functions are capable of approximating any Borel measurable function from one finite dimensional space to another to any desired degree of accuracy, provided sufficiently many hidden units are available. In this sense, multilayer feedforward networks are a class of universal approximators.
Article
The economic and political realities of the 1990s forced managers to reverse long-standing organizational cultures in order to reduce costs and energy expenditures in their organisations. For instance, these can be achieved, with respect to maintenance, by replacing a reactive repair-focused attitude by a proactive reliability-focused culture. Thereby far less (i) human effort is expended and (ii) energy would be wasted, both of which lead to increased profitability. (c) 2006 Published by Elsevier Ltd.
Article
The importance of maintenance function has increased due to its role in keeping and improving the system availability and safety but also the product quality. To support this new role, the maintenance concept has undergone through several major developments to lead to proactive considerations mainly based on prognosis process allowing normally to select the best maintenance action to carry out. In this paper the extension of a prognosis process by means of the integration of maintenance alternative impacts is proposed. The deployment of this extended prognosis process follows a methodology based both on probabilistic and event approaches. The feasibility and added value of this new prognosis is experimented on the manufacturing TELMA platform supporting the unwinding of metal bobbins.
Article
This paper presents a decision support system (DSS) for the modeling and management of project risks and risk interactions. This is a crucial activity in project management, as projects are facing a growing complexity with higher uncertainties and tighter constraints. Existing classical methods have limitations for modeling the complexity of project risks. For example, some phenomena like chain reactions and loops are not properly taken into account. This will influence the effectiveness of decisions for risk response planning and will lead to unexpected and undesired behavior in the project. Based on the concepts of DSS and the classical steps of project risk management, we develop an integrated DSS framework including the identification, assessment and analysis of the risk network. In the network, the nodes are the risks and the edges represent the cause and effect potential interactions between risks. The proposed simulation-based model makes it possible to re-evaluate risks and their priorities, to suggest and test mitigation actions, and then to support project manager in making decisions regarding risk response actions. An example of application is provided to illustrate the utility of the model.
Conference Paper
Production scheduling is a complex decision-making process that is becoming more important than ever in competitive markets today. However, effective production scheduling remains a very challenging task, and the current production scheduling systems tend to be inflexible and expensive. Although domain knowledge and information are very important to scheduling decisions, scheduling systems usually cannot effectively make use of them. In this paper, A multi-scenario analysis (MSA) model is proposed for distributed collaborative decision-making, based on which a multi-scenario analysis-based production scheduling platform (MSAPSP) is developed The platform comprises a toolkit (MSADev) and an engine (MSARE) powered by a script language (EXASL). It provides a flexible and cost-effective tool to exploit domain knowledge and information for scheduling and rescheduling decisions.
Advantages and disadvantages of using neural network for predictions. Ovidius University Annals, Economic sciences series 13
  • D Ciobanu
  • M Vasilescu
Understanding decision support systems and expert systems
  • E G Wallach
Advantages and disadvantages of using neural network for predictions Economic sciences series
  • D Ciobanu
  • M Vasilescu
Using Mixed-Integer Programming to Reduce
  • E Katok
  • D Ott
E. Katok and D. Ott, "Using Mixed-Integer Programming to Reduce," 2000.