Abstract
The concept of Industry 4.0 promises to transversally revolutionise industries. Simulation, as one of the main pillars of Industry 4.0, allows improvements in the organisational and production processes of companies. This research work develops a decision support tool based on system dynamics, that address the problem of car dealership sales forecast and evolution depending on the commercial strategies adopted. This decision support tool considers main variables that are expected to influence car sales in Portugal. To develop this tool several interviews were conducted with the people responsible for the commercial sector of different dealerships while considering existing literature on the subject. This allowed us to parameterize a system dynamics model with the most influential sales factors. The developed tool is expected to contribute to car dealerships to evaluate their commercial policies and define adjustments to these to improve profitability.

Abstract
Breast cancer locoregional treatment causes alterations to the physical aspect of the breast, often negatively impacting the self-esteem of patients unaware of the possible aesthetic outcomes of those treatments. To improve patients' self-esteem and enable a more informed choice of treatment when multiple options are available, the possibility to predict how the patient might look like after surgery would be of invaluable help. However, no work has been proposed to predict the aesthetic outcomes of breast cancer treatment. As a first step, we compare traditional computer vision and deep learning approaches to reproduce asymmetries of post-operative patients on pre-operative breast images. The results suggest that the traditional approach is better at altering the contour of the breast. In contrast, the deep learning approach succeeds in realistically altering the position and direction of the nipple.

Abstract
Background: Diet and exercise interventions have been associated with improved body composition and physical fitness. However, evidence regarding their combined effects in older adults is scarce. This study aimed to investigate the impact of a combined 12-week Mediterranean diet-based sustainable healthy diet (SHD) and multicomponent training (MT) intervention on body composition, anthropometry, and physical fitness in older adults. Methods: Diet intervention groups received a weekly SHD food supply and four sessions, including a SHD culinary practical workshop. The exercise program included MT 50 min group session, three times a week, on non-consecutive days. Body composition and physical fitness variables were assessed through dual X-ray absorptiometry, anthropometric measurements, and senior fitness tests. Repeated measures ANOVA, with terms for group, time, and interaction, was performed. Results: Our results showed that a combined intervention significantly lowered BMI and total fat. Also, significant differences between assessments in all physical fitness tests, except for aerobic endurance, were observed. Adjusted models show significant differences in BMI (p = 0.049) and WHR (p = 0.037) between groups and in total fat (p = 0.030) for the interaction term. Body strength (p < 0.001), balance tests (p < 0.001), and aerobic endurance (p = 0.005) had significant differences amongst groups. Considering the interaction term, differences were observed for upper body strength (p = 0.046) and flexibility tests (p = 0.004 sit and reach, p = 0.048 back scratch). Conclusions: Our intervention study demonstrates the potential of implementing healthy lifestyle and sustainable models to promote healthy and active aging.

Abstract
Time series forecasting is a challenging task with applications in a wide range of domains. Auto-regression is one of the most common approaches to address these problems. Accordingly, observations are modelled by multiple regression using their past lags as predictor variables. We investigate the extension of auto-regressive processes using statistics which summarise the recent past dynamics of time series. The result of our research is a novel framework called VEST, designed to perform feature engineering using univariate and numeric time series automatically. The proposed approach works in three main steps. First, recent observations are mapped onto different representations. Second, each representation is summarised by statistical functions. Finally, a filter is applied for feature selection. We discovered that combining the features generated by VEST with auto-regression significantly improves forecasting performance in a database composed by 90 time series with high sampling frequency. However, we also found that there are no improvements when the framework is applied for multi-step forecasting or in time series with low sample size. VEST is publicly available online.

Abstract
Optical tweezers are an interesting tool to enable single cell analysis, especially when coupled with optical sensing and advanced computational methods. Nevertheless, such approaches are still hindered by system operation variability, and reduced amount of data, resulting in performance degradation when addressing new data sets. In this manuscript, we describe the deployment of an automatic and intelligent optical tweezers setup, capable of trapping, manipulating, and analyzing the physical properties of individual microscopic particles in an automatic and autonomous manner, at a rate of 4 particle per min, without user intervention. Reproducibility of particle identification with the help of machine learning algorithms is tested both for manual and automatic operation. The forward scattered signal of the trapped PMMA and PS particles was acquired over two days and used to train and test models based on the random forest classifier. With manual operation the system could initially distinguish between PMMA and PS with 90% accuracy. However, when using test datasets acquired on a different day it suffered a loss of accuracy around 24%. On the other hand, the automatic system could classify four types of particles with 79% accuracy maintaining performance (around 1% variation) even when tested with different datasets. Overall, the automated system shows an increased reproducibility and stability of the acquired signals allowing for the confirmation of the proportionality relationship expected between the particle size and its friction coefficient. These results demonstrate that this approach may support the development of future systems with increased throughput and reliability, for biosciences applications.

Abstract
Demand Driven Material Requirements Planning (DDMRP) assumes that a production order is generated for replenishment when the inventory position, given by the net flow equation, is below a given level. Literature on this production planning and control system suggests prioritizing open orders on the shop floor based on the inventory buffer status. However, the performance of buffer-oriented priority dispatching largely remains unknown. Using discrete event simulation, this study suggests that buffer-oriented dispatching based on the net flow equation outperforms due date-oriented dispatching rules and first-come-first-served. The performance impact depends, however, on the reorder quantity associated with the production orders. These results have important implications for industrial practice.