Abstract
In many companies in the automotive industry there are challenges in some key processes in their logistic departments, mainly in internal logistics. These challenges happen due to poorly defined rules for the transportation of goods, resulting in a great cost associated with the time lost in the process. Also, the optimization of these processes, incrementing the efficiency of internal logistics can bring competitive advantages to the companies. For that matter, this study was developed at a major tire manufacturing company and proposes a model for the optimization of in-bound logistics, viewed as an online vehicle routing problem with soft deadlines (OVRPSD), using multiple depots. The main goal of this study is the increase of efficiency in logistic, optimizing the number of vehicles to supply the machines in order to reduce the stopping time of machines due to the lack of tires to consume. © 2019, Springer International Publishing AG, part of Springer Nature.

Abstract
Today's power systems are subject to various challenges arising from the large-scale integration of renewable energy sources (RES), especially wind energy production. System flexibility, or the capability of a system to address deviations in variable RES production, is becoming more and more relevant. This paper aims to provide a systematic approach to evaluate the level of flexibility of a power system by unequivocally considering fast-ramping units (FRU), hourly demand response (DR) and energy storage (ES). In addition, to research the flexibility role in power system operation, an "online" index is considered to evaluate the technical aptitude of the FRU, hourly DR and ES system to deliver the required flexibility. The mathematical representation of day-ahead scheduling, with the added modeling of an online flexibility index, is a mixed-integer nonlinear program (MINLP). This paper presents a method to convert this MINLP into a mixed-integer linear program without loss of accuracy. The adapted 6-bus and IEEE 118-bus systems are employed to assess the suggested models and flexibility metric, demonstrating the proficiency of the online flexibility index.

Abstract
Heart sounds are difficult to interpret due to events with very short temporal onset between them (tens of milliseconds) and dominant frequencies that are out of the human audible spectrum. Computer-assisted decision systems may help but they require robust signal processing algorithms. In this paper, we propose a new algorithm for heart sound segmentation using a hidden semi-Markov model. The proposed algorithm infers more suitable sojourn time parameters than those currently suggested by the state of the art, through a maximum likelihood approach. We test our approach over three different datasets, including the publicly available PhysioNet and Pascal datasets. We also release a pediatric dataset composed of 29 heart sounds. In contrast with any other dataset available online, the annotations of the heart sounds in the released dataset contain information about the beginning and the ending of each heart sound event. Annotations were made by two cardiopulmonologists. The proposed algorithm is compared with the current state of the art. The results show a significant increase in segmentation performance, regardless the dataset or the methodology presented. For example, when using the PhysioNet dataset to train and to evaluate the HSMMs, our algorithm achieved average an F-score of 92% compared to 89% achieved by the algorithm described in [D.B. Springer, L. Tarassenko, and G. D. Clifford, "Logistic regressionHSMM-based heart sound segmentation," IEEE Transactions on Biomedical Engineering, vol. 63, no. 4, pp. 822-832, 2016]. In this sense, the proposed approach to adapt sojourn time parameters represents an effective solution for heart sound segmentation problems, even when the training data does not perfectly express the variability of the testing data.

Abstract
Rheumatic heart disease remains a major burden in the developing countries. The World Heart Federation proposed guidelines for the echocardiographic detection of the disease, in which the mitral leaflets' morphology assessment is a key indicator. The drawback is that these guidelines are dependent on the clinician experience. To overcome this limitation, we propose an automatic segmentation of the mitral leaflets using a new method based on convolutional neural network, specifically the UNet architecture. The results indicate a median DICE coefficient of 0.74 in PLAX and 0.79 in A4C for the anterior mitral leaflet segmentation, while median DICE of 0.60 in PLAX and 0.69 A4C are met for the posterior leaflet. A visual evaluation of this segmentation approach by two cardiologists is in line with the numerical results. The false detection due to overestimation and artifacts remains an issue to be addressed in the future.

Abstract
The longevity of the population is the result of important scientific breakthroughs in recent years. However, living longer with quality, also brings new challenges to governments, and to the society as a whole. One of the most significant consequences will be the increasing pressure on the healthcare services. Ambient Assisted Living (AAL) systems can greatly improve healthcare scalability and reach while keeping the user in their home environment. The work presented in this paper specifies, implements, and validates a smart environment system that aggregates Automation and Artificial Intelligence (AI). The specification includes a reference architecture, composed by three modules, whose tasks are to automate and standardize the collection of data, to relate and give meaning to that data and to learn from it. The system is able to identify daily living activities with different levels of complexity using a temporal logic. It enables a real time response to emergency situations and also a long term analysis of the user daily routine useful to induce healthier lifestyles. The implementation addresses the applications and techniques used in the development of a functional prototype. To demonstrate the system operation three use cases with increasing levels of complexity are proposed and validated. A discussion on related projects is also included, specifically on automation applications, Knowledge Representation (KR) and Machine Learning (ML).

Abstract