Abstract
X-by-wire applications have extremely demanding reliability requirements that are increasingly being addressed through the adoption of distributed and fault-tolerant architectures. The development of these applications is facilitated by the availability of high-level services such as agreement or reliable broadcast (RB). Some dependable communication buses, e.g., TTP-C, already provide these services, whereas FlexRay does not. In this paper, we present an approach to evaluate the reliability of a family of RB protocols implemented both on top of FlexRay and on top of ordinary time-division multiple access (TDMA). In particular, we evaluate the impact of the acknowledgment policy on the reliability of these protocols. We express the reliability as the probability of violation of the agreement and validity properties of the protocol during a mission. For that, we develop an analytical model based on discrete-time Markov chains, which considers a comprehensive set of faults (permanent, transient, omissive, and asymmetric) affecting both nodes and channels, and their effects on the protocol execution. The structure of the model is quite flexible and easily adaptable to other TDMA-based protocols. To assess the sensitivity of the protocol to both internal and external factors, we carried out a large number of experiments considering several network configurations and fault rates. The results show that for FlexRay, the negative-acknowledgment policy provides the same reliability as the positive-acknowledgment policy. However, for TDMA-based protocols that lack FlexRay's ability to distinguish silence from the loss of a message, the negative-acknowledgment policy leads to lower reliability, and its fitness for safety-critical applications depends on the system configuration and environment conditions.

Abstract
Current technological and market challenges increase the need for development of intelligent systems to support decision making, allowing managers to concentrate on high-level tasks while improving decision response and effectiveness. A Racing based learning module is proposed to increase the effectiveness and efficiency of a Multi-Agent System used to model the decision-making process on scheduling problems. A computational study is put forward showing that the proposed Racing learning module is an important enhancement to the developed Multi-Agent Scheduling System since it can provide more effective and efficient recommendations in most cases.

Abstract
Aggressive marketing causes rapid changes in consumer behavior and some significant impact in the retail business. In this context, the sales forecasting at the SKU level can help retailers to become more competitive by reducing inventory investment and distribution costs. Sales forecasts are often obtained combining basic univariate forecasting models with empirical judgment. However, more effective forecasting methods can be obtained by incorporating promotional information, including price, percentage of discount (direct discount or loyalty card discount), calendar events and weekend indicators not only from the focal product but also from its competitors. To deal with the high dimensionality of the variable space, we propose a two-stage LASSO regression to select optimal predictors and estimate the model parameters. At the first stage, only focal SKUs promotional explanatory variables are included in the Autoregressive Distributed Lag model. At the second stage, the in-sample forecast errors from the first stage are regressed on the explanatory variables from the other SKUs in the same category with the focal SKU, and to use that information more effectively three different approaches were considered: select the five top sales SKUs, include all raw promotional information, and preprocess raw information using Principal Component Analysis. The empirical results obtained using daily data from a Portuguese retailer show that the inclusion of promotional information from SKUs in the same category may improve the forecast accuracy and that better overall forecasting results may be obtained if the best model for each SKU is selected.

Abstract
Recommender Systems are an important tool in e-business, for both companies and customers. Several algorithms are available to developers, however, there is little guidance concerning which is the best algorithm for a specific recommendation problem. In this study, a metalearning approach is proposed to address this issue. It consists of relating the characteristics of problems (metafeatures) to the performance of recommendation algorithms. We propose a set of metafeatures based on the application of systematic procedure to develop metafeatures and by extending and generalizing the state of the art metafeatures for recommender systems. The approach is tested on a set of Matrix Factorization algorithms and a collection of real-world Collaborative Filtering datasets. The performance of these algorithms in these datasets is evaluated using several standard metrics. The algorithm selection problem is formulated as classification tasks, where the target attribute is the best Matrix Factorization algorithm, according to each metric. The results show that the approach is viable and that the metafeatures used contain information that is useful to predict the best algorithm for a dataset. © Springer International Publishing AG 2016.

Abstract
The Ada language has for long provided support for the development of reliable real-time systems, with a model of computation amenable for real-time analysis. To complement the already existent multiprocessor support in the language, an ongoing effort is underway to extend Ada with a fine-grained parallel programming model also suitable for real-time systems. This paper overviews the model which is being proposed, pointing out the main issues still open and road ahead. © 2015 IEEE.

Abstract
The restructuring of power systems induced new challenges to generation companies in terms of adequately planning the operation of power stations in order to maximize their profits. In this scope, hydro resources are becoming extremely valuable given the revenues that their operation can generate. In this paper we describe the application of the Matlab (R) Linprog optimization function to solve the Short Term Hydro Scheduling Problem, HSP, admitting that some stations are installed in the same cascade and that some of them have pumping capabilities. The optimization module to solve the HSP problem is then integrated in an iterative process to take into account the impact that the operation decisions regarding the hydro stations under analysis have on the market prices. The updated market prices are then used to run again the HSP problem thus enabling considering the hydro stations as price makers. The developed approach is illustrated using a system based on the Portuguese Douro River cascade that includes 9 hydro stations (4 of them are pumping stations) and a total installed capacity of 1485 MW.