Search

A key aspect for the forklifts is the state-of-health (SoH) assessment to ensure the safety and the reliability of uninterrupted power source. Forecasting the battery SoH well is imperative to enable preventive maintenance and hence to reduce the costs. This paper demonstrates the capabilities of gradient boosting regression for predicting the SoH timeseries under circumstances when there is little prior information available about the batteries. We compared the gradient boosting method with light gradient boosting, extra trees, extreme gradient boosting, random forests, long short-term memory networks and with combined convolutional neural network and long short-term memory networks methods. We used multiple predictors and lagged target signal decomposition results as additional predictors and compared the yielded prediction results with different sets of predictors for each method. For this work, we are in possession of a unique data set of 45 lithium-ion battery packs with large variation in the data. The best model that we derived was validated by a novel walk-forward algorithm that also calculates point-wise confidence intervals for the predictions; we yielded reasonable predictions and confidence intervals for the predictions. Furthermore, we verified this model against five other lithium-ion battery packs; the best model generalised to greater extent to this set of battery packs. The results about the final model suggest that we were able to enhance the results in respect to previously developed models. Moreover, we further validated the model for extracting cycle counts presented in our previous work with data from new forklifts; their battery packs completed around 3000 cycles in a 10-year service period, which corresponds to the cycle life for commercial Nickel-Cobalt-Manganese (NMC) cells., Comment: 16 pages, 10 figures and 10 tables

International audience; Intelligent products are an undeniable asset for efficient product lifecycle management (PLM), providing ways to capture events related to physical objects at various locations and times. Today and more than ever before, PLM tools and systems must be built upon standards for enhancing interoperability among all product stakeholders and developing tools independent of specific vendors, applications, and operating systems. Based on this observation, this paper develops strategies to improve “information sustainability” in PLM environments using standardized communication interfaces defined by a recent standard proposal named quantum lifecycle management (QLM) messaging standards. More concretely, data synchronization models based upon QLM standards are developed to enable the synchronization of product-related information among various systems, networks, and organizations involved throughout the product lifecycle. Our proposals are implemented and assessed based on two distinct platforms defined in the healthcare and home automation sectors.

The context where European manufacturers of industrial systems operate has dramatically changed over recent years: the pressure of emerging countries they have to face, policy makers' environmental laws and industrial companies' interests are pushing towards sustainable manufacturing and a holistic view of industrial systems. Designers and system engineers are the main actors involved, because they have high influence on product life cycle costs and environmental impacts. However they need tools to pursue a holistic view. The aim of this paper is to propose a closed loop framework to improve life cycle performances of industrial systems, focusing on the automotive sector. (C) 2015 The Authors. Published by Elsevier B.V.

International audience; In recent years, some scholars claimed the usage of intelligent products to make systems more efficient throughout the Product Life Cycle (PLC). Integrating intelligence and information into products themselves is possible with, among others, auto-ID technologies (barcode, RFID,...). In this paper, a new kind of intelligent product is introduced, referred to as "communicating material" paradigm. Through this paradigm, a product is i) capable of embedding information on all or parts of the material that it is made of, ii) capable of undergoing physical transformations without losing its communication ability and the data that is stored on it. This new material is used in our study to convey information between the different actors of the PLC, thus improving data interoperability, availability and sustainability. Although "communicating materials" provide new abilities compared to conventional products, they still have low memory capacities compared to product databases that become larger and larger. An information dissemination framework is developed in this paper to select the appropriate information to be stored on the product, at different stages of the PLC. This appropriateness is based on a degree of data relevance, which is computed by taking into account the context of use of the product (actor's expectations, environment,...). This framework also provides the tools to split information on all or parts of the material. A case study is presented, which aims at embedding context-sensitive information on "communicating textiles".

In our vision of the Internet of Things, it is possible to easily query and set up information flows between any kinds of products, devices, computers, users and information systems in general. However, for the moment it is not possible due to the lack of sufficiently generic and standardized application-level interfaces for exchanging the kind of information required by an Internet of Things. The Quantum Lifecycle Management (QLM) messaging standards have been developed and are proposed as a standard application-level interface that would fulfill such requirements. The article presents the design principles and provides a high-level description of QLM messaging. Real-life implementations illustrate why such a messaging interface is needed and compares it with related standards.

International audience; Intelligent products may be considered as information vectors, conveying their own information from a phase of its life cycle to another. Considering that such a product could be made with "communicating material", in previous work, a data dissemination process has been developed with the aim of selecting information from a database, that should be stored on the product according to the current stage of its life cycle. This selection is based on an indicator of relevance which depends on a multitude of criteria, whose weights are evaluated by a group of experts. The set of expert opinions must be formalized in accordance with a mathematical theory and then, synthesized by a suitable aggregation method. In this framework, three aggregation methods are introduced and compared. This comparison is carried out based on the scenario of data dissemination when using intelligent products. Results enable to assess the goodness of fit of the three aggregation methods.

Developments in the area of radio frequency identification (RFID) and sensor network technologies have created new possibilities for product lifecycle management (PLM). These technologies are enabling building blocks upon which various business applications for product management can be built. A significant aspect in the through-life management of products is the gathering and management of data related to the product during the various phases of its lifecycle. Both RFID and wireless sensor technologies have created novel levels of product status visibility and automatic identification with granularity to the level of individual components. We consider three approaches that leverage automatic identification technologies and miniaturised wireless sensors to support PLM strategies at the product instance level based on a product centric computing paradigm. The paper draws upon a number of case studies focusing on the various phases of a product's lifecycle: Beginning of Life, Middle of Life and End of Life. We used the case studies to extract technical, capability and information requirements needed to support PLM strategies. We compared the ability and suitability of existing architectural approaches to meet the key requirements we identified. Furthermore we assessed them to evaluate their level of support for managing product instances at different lifecycle phases. Finally, we present an extension to significantly improve the ability of the most promising architecture for supporting PLM strategies.

The significance of tracking in operations management has been overshadowed, theoretically by the concept of supply chain visibility, and practically by the attention grabbing radio frequency identification (RFID) technology. This paper describes how uses of tracking for operations management, specifically, to improve the management of project and service supply chains, were explored and developed in a ten-year research program. Key results of the program are synthesized and linked to the relevant academic discourse in operations management using design science methodology.

Item-centric tracking is an opportunity to increase visibility and control in different operations of a company. The economical feasibility of item-centric tracking is based on recent technological developments for monitoring the material flow on the item-level instead of the material type-level. It enables companies to track and trace assets over their lifecycle and manage supply chain operations across organizational boundaries. However, the challenge in practice is to find the right areas to start the development of business applications. In this viewpoint paper a means-ends framework for introducing item-centric tracking in business is proposed based on realized track and trace cases. Based on the framework a roadmap for introducing tracking in businesses operations is outlined. The proposed roadmap circumvents the most challenging business cases for introducing tracking and RFID-technology in the supply chain and focuses on asset management as a more accessible route for business. © 2008 Elsevier B.V. All rights reserved.

Reinforcement learning is based on exploration of the environment and receiving reward that indicates which actions taken by the agent are good and which ones are bad. In many applications receiving even the first reward may require long exploration, during which the agent has no information about its progress. This paper presents an approach that makes it possible to use pre-existing knowledge about the task for guiding exploration through the state space. Concepts of short- and long-term memory combine guidance by pre-existing knowledge with reinforcement learning methods for value function estimation in order to make learning faster while allowing the agent to converge towards a good policy.

A manufactured product usually has a great amount of design knowledge and other product information associated with it in CAD drawings and other technical documentation. Today, such product information mainly enables preventive maintenance of products. By collecting usage, performance and maintenance information about the product, it could be possible to perform more predictive maintenance. The key issue is to document product information in a way that would enable analysing the relation between usage parameters and actual product failures. Semantic networks are here proposed as an intermediate knowledge representation format for combining such static and dynamic product information.

The management of single-item level information has grown to be one of the greatest challenges of supply network management. The main reasons for this are the increasing product customisation and increasingly complex supply networks. Product customisation adds to the quantity and specificity of product-related information. Current mainstream solutions, based on integrating product information databases, are not suitable for complex, short-term supply networks. This paper presents an approach and a proposed system for managing single-item level information. The approach we call product centric information management is based on centralising information to the individual products. The proposed system uses software agents, peer-to-peer information sharing, and a coding mechanism that utilises the domain names of the Internet. We also review applications of the system in tracking and logistics control.

International projects present really difficult logistics challenges. In international investment projects, a vast number of individualised deliveries has to be managed through a large supply network in a tight timeframe. This article investigates how the logistic challenges of international projects can be solved by utilising advanced web technologies and product identification. The paper presents a control system being built at Helsinki University of Technology, which is based on distributed programming, and wireless identification technologies. The aim of the system is to change the controlling mechanisms of project deliveries by giving the deliveries themselves the means with which to control their route. This enables the material flows in the project delivery chains to be controlled from the inside of the material flow itself.