Abstract
Despite the growing relevance of customization as a source of competitive advantage, the make-to-order (MTO)/engineer-to-order (ETO) manufacturing strategies have been neglected in the literature. Companies following these strategies deal with simultaneous customer-oriented projects that compete for and share resources, while coordinating interdependent engineering and production activities. It becomes relevant understanding the impact that different development projects and production variables have on the manufacturing system performance. For this, we propose a hybrid multi-dimensional simulation model, using System Dynamics (SD), Discrete Event Simulation (DES) and Agent-based simulation (ABS) for MTO/ETO performance assessment. (C) 2018 The Authors. Published by Elsevier B.V.

Abstract
Fingerprint recognition has been widely studied for more than 45 years and yet it remains an intriguing pattern recognition problem. This paper focuses on the foreground mask estimation which is crucial for the accuracy of a fingerprint recognition system. The method consists of a robust cluster-based fingerprint segmentation framework incorporating an additional step to deal with pixels that were rejected as foreground in a decision considered not reliable enough. These rejected pixels are then further analysed for a more accurate classification. The procedure falls in the paradigm of classification with reject option- a viable option in several real world applications of machine learning and pattern recognition, where the cost of misclassifying observations is high. The present work expands a previous method based on the fuzzy C-means clustering with two variations regarding: i) the filters used; and ii) the clustering method for pixel classification as foreground/background. Experimental results demonstrate improved results on FVC datasets comparing with state-of-the-art methods even including methodologies based on deep learning architectures.

Abstract
The spread of biometric applications in mobile devices handled by untrained users opened the door to sources of noise in mobile iris recognition such as larger extent of rotation in the capture and more off-angle imagery not found so extensively in more constrained acquisition settings. As a result of the limitations of the methods in handling such large degrees of freedom there is often an increase in segmentation errors. In this work, a new near-infrared iris dataset captured with a mobile device is evaluated to analyse, in particular, the rotation observed in images and its impact on segmentation and biometric recognition accuracy. For this study a (manually annotated) ground truth segmentation was used which will be published in tandem with the paper. Similarly to most research challenges in biometrics and computer vision in general, deep learning techniques are proving to outperform classical methods in segmentation methods. The utilization of parameterized CNN-based iris segmentations in biometric recognition is a new but promising field. The results presented show how this CNN-based approach outperformed the segmentation traditional methods with respect to overall recognition accuracy for the dataset under investigation.

Abstract
In this paper, we investigate two variants of association rules for preference data, Label Ranking Association Rules and Pairwise Association Rules. Label Ranking Association Rules (LRAR) are the equivalent of Class Association Rules (CAR) for the Label Ranking task. In CAR, the consequent is a single class, to which the example is expected to belong to. In LRAR, the consequent is a ranking of the labels. The generation of LRAR requires special support and confidence measures to assess the similarity of rankings. In this work, we carry out a sensitivity analysis of these similarity-based measures. We want to understand which datasets benefit more from such measures and which parameters have more influence in the accuracy of the model. Furthermore, we propose an alternative type of rules, the Pairwise Association Rules (PAR), which are defined as association rules with a set of pairwise preferences in the consequent. While PAR can be used both as descriptive and predictive models, they are essentially descriptive models. Experimental results show the potential of both approaches.

Abstract
Carcass data were collected from 24 kids (average live weight of 12.5 +/- 5.5 kg; range 4.5 to 22.4 kg) of Jarmelista Portuguese native breed, to evaluate bioelectrical impedance analysis (BIA) as a technique for prediction of light kid carcass and muscle chemical composition. Resistance (Rs, Omega) and reactance (Xc, Omega), were measured in the cold carcasses with a single frequency bioelectrical impedance analyzer and, together with impedance (Z, Omega), two electrical volume measurements (Vol(A) and Vol(B), cm(2)/Omega), carcass cold weight (CCW), carcass compactness and several carcass linear measurements were fitted as independent variables to predict carcass composition by stepwise regression analysis. The amount of variation explained by Vol(A) and Vol(B) only reached a significant level (P < 0.01 and P < 0.05, respectively) for muscle weight, moisture, protein and fat-free soft tissue content, even so with low accuracy, with VolA providing the best results (0.326 <= R-2 <= 0.366). Quite differently, individual BIA parameters (Rs, Xc and Z) explained a very large amount of variation in dissectible carcass fat weight (0.814 <= R-2 <= 0.862; P < 0.01). These individual BIA parameters also explained a large amount of variation in subcutaneous and intermuscular fat weights (respectively 0.749 <= R-2 <= 0.793 and 0.718 <= R-2 <= 0.760; P < 0.01), and in muscle chemical fat weight (0.663 <= R-2 <= 0.684; P < 0.01). Still significant but much lower was the variation in muscle, moisture, protein and fat-free soft tissue weights (0.344 <= R-2 <= 0.393; P < 0.01) explained by BIA parameters. Still, the best models for estimation of muscle, moisture, protein and fat-free soft tissue weights included Rs in addition to CCW, and accounted for 97.1% to 99.8% (P < 0.01) of the variation observed, with CCW by itself accounting for 97.0% to 99.6% (P < 0.01) of that variation. Resistance was the only independent variable selected for the best model predicting subcutaneous fat weight. It was also selected for the best models predicting carcass fat weight (combined with carcass length, CL; R-2 = 0.943; P < 0.01) and intermuscular fat weight (combined with CCW; R-2 = 0.945; P < 0.01). The best model predicting muscle chemical fat weight combined CCW and Z, explaining 85.6% (P < 0.01) of the variation observed. These results indicate BIA as a useful tool for prediction of light kids' carcass composition.

Abstract
This paper aims to analyse the bi-directional relationship between technical efficiency, as a measure of companies' performance, and capital structure, under the agency cost theory as well as the pecking order and trade-off theory, to explain the capital structure decisions. The technical efficiency was estimated by the DEA method and corrected by using a suitable bootstrap to obtain statistical inferences. To test the agency cost hypothesis, asymmetric information hypothesis, risk-efficiency hypothesis and franchise value hypothesis (under pecking order and trade off theories framework), two models were applied using some determinants of capital structure such as size, profitability, tangibility, liquidity as control and explanatory variables through a truncated regression with bootstrapping. From an initial sample of 1024 small and medium sized companies from the interior of Portugal, for the period 2006-2009, a subsample of 210 SMEs from secondary and tertiary sectors was selected. The results suggest that medium sized companies have higher average bias-corrected efficiency than small companies; that short-term leverage is positively related to efficiency and that the companies in the sample follow pecking order theory.