Abstract
Window functions are a sub-class of analytical operators that allow data to be handled in a derived view of a given relation, while taking into account their neighboring tuples. Currently, systems bypass parallelization opportunities which become especially relevant when considering Big Data as data is naturally partitioned. We present a shuffling technique to improve the parallel execution of window functions when data is naturally partitioned when the query holds a partitioning clause that does not match the natural partitioning of the relation. We evaluated this technique with a non-cumulative ranking function and we were able to reduce data transfer among parallel workers in 85% when compared to a naive approach.

Abstract
Sales forecasts gained more importance in the retail industry with the increasing of promotional activity, not only because of the considerable portion of products under promotion but also due to the existence of promotional activities, which boost product sales and make forecasts more difficult to obtain. This study is performed with real data from a Portuguese consumer goods retail company, from January 2012 until April 2015. To achieve the purpose of the study, dynamic regression is used based on information of the focal product and its competitors, with seasonality modelled using Fourier terms. The selection of variables to be included in the model is done based on the lowest value of AIC in the train period. The forecasts are obtained for a test period of 30 weeks. The forecasting models overall performance is analyzed for the full period and for the periods with and without promotions. The results show that our proposed dynamic regression models with price and promotional information of the focal product generate substantially more accurate forecasts than pure time series models for all periods studied.

Abstract
In the past few decades, ecological and environmental issues have dominated the forest industry worldwide, but economic aspects have been much less studied in this dynamic period. However, a sustainable and efficient forest biomass supply is critical for socio-economic development in many regions, particularly in rural areas. Nature protection efforts have contributed to reduced harvesting quotas, which have resulted in an imbalance of the environmental functions of the forests and forest management, particularly wood supply. Considering the size and distribution of forest production management units and the forest stands that compose those units, there is a clear need for improved decision-making tools that help forest managers in planning harvest sequences. The optimization of harvest scheduling should consider economic and spatial factors, which may reduce production costs by increasing the logistic efficiency. Moreover, incorporating maximum harvesting opening size constraints into planning can help preserve biodiversity. This article presents a new spatial harvest scheduling model based on the integer programming method; it was developed using real data from a forest production unit located in the northern part of the southeast region of Brazil. The goal of the proposed scheduling approach is to maximize the net present value and concentrate the harvesting locations in each period. In spite of the fact that the object of the study is plantation forest under management different to common conditions in Europe or North America, the model is flexible and can be used in management of forest in Central Europe. © 2016 Jan Kašpar et al.

Abstract
The future of robotics is now trending for home servicing. Nursing homes and assistance to elder peopleare areas where robots can provide valuable help in order to improve the quality of life of those who need it most. Calling a robot,for a person of age,can be a daunting task if the voice is failing and any resort to battery operated devices failsto comply. Using a simple mechanical apparatus,such as aClick trainerfordogs, a person can call a robot by pressing thebutton of a powerless device. The high pitch sound produced by this device can be captured and tracked down in order to estimate the person’s location within a room. This paper describes a method that provides good accuracy and uses simple and low cost technology,in order to provide an efficient positional value for an assistance robot to attend its caller. The robot does not need to search for the person in aroom as it can directly travel towards the Click’s sound source.

Abstract
The electricity markets environment has changed completely with the introduction of renewable energy sources in the energy distribution systems. With such alterations, preventing the system from collapsing required the development of tools to avoid system failure. In this new market environment competitiveness increases, new and different power producers have emerged, each of them with different characteristics, although some are shared for all of them, such as the unpredictability. In order to battle the unpredictability, the power supplies of this nature are supported by techniques of artificial intelligence that enables them crucial information for participation in the energy markets. In electricity markets any player aims to get the best profit, but is necessary have knowledge of the future with a degree of confidence leading to possible build successful actions. With optimization techniques based on artificial intelligence it is possible to achieve results in considerable time so that producers are able to optimize their profits from the sale of Electricity. Nowadays, there are many optimization problems where there are no that cannot be solved with exact methods, or where deterministic methods are computationally too complex to implement. Heuristic optimization methods have, thus, become a promising solution. In this paper, a simulated annealing based approach is used to solve the portfolio optimization problem for multiple electricity markets participation. A case study based on real electricity markets data is presented, and the results using the proposed approach are compared to those achieved by a previous implementation using particle swarm optimization.

Abstract