In smart factories, guaranteeing shopfloor-synchronised and real-time decision-making is essential to be responsive to the ever-changing internal environment, namely the shopfloor of the production system and assets. At operational level, decisions should balance counter acting objectives of maintenance and production; there- fore, their decision-making processes should be joint and coordinated, to fulfil production requirements considering the health state of the assets. The knowledge of the current state is promoted by the application of Prognostics and Health Management (PHM) as an aid to support informed decision-making. Nevertheless, PHM- purposed information is usually not complete in terms of production requirements. To support joint maintenance and production management decisions, an ontological approach is proposed. The ontology, called ORMA (Ontology for Reliability-centred MAintenance), has a modular structure, including formalisation of asset, pro- cess, and product knowledge. Via suitable relationships, rules, and axioms, ORMA can infer product feasibility based on the current health state of the assets and their functional units. ORMA is implemented in a Flexible Manufacturing Line at a laboratory scale. Therein, an integrated solution, involving a health state detection algorithm that interacts with the ontology, supports human decision-making via a web-based dashboard; joint maintenance and production management decisions can be then taken, relying on diversified information pro- vided by the PHM algorithm as well as the augmentation via ontology reasoning. The proposed ontology-based solution represents a step towards reconfigurability of smart factories where human and automated decision- making processes work in synergy.
keywords: هستی شناسی | استدلال | پیشگویی و مدیریت بهداشت | phm | نگهداری | تولید | Ontology | Reasoning | Prognostics and health management | PHM | maintenance | production

Article history:Received 25 February 2016Revised 12 August 2016Accepted 26 August 2016Available online 9 September 2016Keywords:Computational histopathology ClassificationUnsupervised feature learning Sparse feature encoderClassification of histology sections in large cohorts, in terms of distinct regions of microanatomy (e.g., stromal) and histopathology (e.g., tumor, necrosis), enables the quantification of tumor composition, and the construction of predictive models of genomics and clinical outcome. To tackle the large technical vari- ations and biological heterogeneities, which are intrinsic in large cohorts, emerging systems utilize either prior knowledge from pathologists or unsupervised feature learning for invariant representation of the underlying properties in the data. However, to a large degree, the architecture for tissue histology classi- fication remains unexplored and requires urgent systematical investigation. This paper is the first attempt to provide insights into three fundamental questions in tissue histology classification: I. Is unsupervised feature learning preferable to human engineered features? II. Does cellular saliency help? III. Does the sparse feature encoder contribute to recognition? We show that (a) in I, both Cellular Morphometric Fea- ture and features from unsupervised feature learning lead to superior performance when compared to SIFT and [Color, Texture]; (b) in II, cellular saliency incorporation impairs the performance for systems built upon pixel-/patch-level features; and (c) in III, the effect of the sparse feature encoder is correlated with the robustness of features, and the performance can be consistently improved by the multi-stage ex- tension of systems built upon both Cellular Morphmetric Feature and features from unsupervised feature learning. These insights are validated with two cohorts of Glioblastoma Multiforme (GBM) and Kidney Clear Cell Carcinoma (KIRC).© 2016 Elsevier B.V. All rights reserved.
Keywords: Computational histopathology | Classification | Unsupervised feature learning | Sparse feature encoder