Forecasting human entrances at a commercial store using facial recognition data

Dissertação de mestrado integrado em Engenharia e Gestão de Sistemas de Informação

Forecasting time series combining Holt-Winters and bootstrap approaches

Publicado em "AIP Conference Proceedings", Vol. 1648

Influence of ageing on the properties of bitumen from asphalt mixtures with recycled concrete aggregates

The reuse of recycled concrete aggregates in new hot-mix asphalt can be a more sustainable method of production, but these mixtures may need a heat treatment before compaction to improve their water sensitivity performance. A direct consequence of this treatment is an increase in the hot-mix asphalt resilient modulus. The aim of this paper is to analyse the effect of ageing on the stiffness of asphalt mixtures with different amounts of recycled concrete aggregates, before and after a heat treatment, which was analysed through the assessment of its bitumen properties. Moreover, this paper also aims to analyse whether the rolling thin-film oven test is able to simulate the ageing effect of the heat treatment. In the laboratory work, a paving grade bitumen B50/70 has been used to produce asphalt mixtures with 0% and 30% recycled concrete aggregates, and the bitumen was later characterised (using penetration, softening point, dynamic viscosity and dynamic shear rheometer tests) in various situations, such as when using virgin bitumen, short-term aged bitumen, aged bitumen after heat treatment (simulated with 4 h of rolling thin-film oven test) and bitumen samples recovered from asphalt mixtures with different production mixes (0% and 30% recycled concrete aggregate) and heat treatment conditions (0 and 4 h of curing time in the oven). Based on the results obtained, it could be concluded that the ageing resulting from the heat treatment is the primary cause of the hot-mix asphalt's increased stiffness, while recycled concrete aggregate content has a small influence. Moreover, it could be concluded that when there is no curing time, the recycled concrete aggregate protects the bitumen against ageing. Additionally, it could be stated that the rolling thin-film test is able to adequately simulate the ageing effect of the heat treatment. Thus, this test is useful for determining the ageing suffered by the bitumen when the recycled concrete aggregate mixture is manufactured using a heat treatment.

Integrated management systems as complex adaptive systems

It is a difficult task to avoid the “smart systems” topic when discussing smart prevention and, similarly, it is a difficult task to address smart systems without focusing their ability to learn. Following the same line of thought, in the current reality, it seems a Herculean task (or an irreparable omission) to approach the topic of certified occupational health and safety management systems (OHSMS) without discussing the integrated management systems (IMSs). The available data suggest that seldom are the OHSMS operating as the single management system (MS) in a company so, any statement concerning OHSMS should mainly be interpreted from an integrated perspective. A major distinction between generic systems can be drawn between those that learn, i.e., those systems that have “memory” and those that have not. These former systems are often depicted as adaptive since they take into account past events to deal with novel, similar and future events modifying their structure to enable success in its environment. Often, these systems, present a nonlinear behavior and a huge uncertainty related to the forecasting of some events. This paper seeks to portray, for the first time as we were able to find out, the IMSs as complex adaptive systems (CASs) by listing their properties and dissecting the features that enable them to evolve and self-organize in order to, holistically, fulfil the requirements from different stakeholders and thus thrive by assuring the successful sustainability of a company. Based on the revision of literature carried out, this is the first time that IMSs are pointed out as CASs which may develop fruitful synergies both for the MSs and for CASs communities. By performing a thorough revision of literature and based on some concepts embedded in the “DNA” of the subsystems implementation standards it is intended, specifically, to identify, determine and discuss the properties of a generic IMS that should be considered to classify it as a CAS.

Using customer lifetime value to improve the prediction of bank deposit subscription in telemarketing campaigns

Customer lifetime value (LTV) enables using client characteristics, such as recency, frequency and monetary (RFM) value, to describe the value of a client through time in terms of profitability. We present the concept of LTV applied to telemarketing for improving the return-on-investment, using a recent (from 2008 to 2013) and real case study of bank campaigns to sell long- term deposits. The goal was to benefit from past contacts history to extract additional knowledge. A total of twelve LTV input variables were tested, un- der a forward selection method and using a realistic rolling windows scheme, highlighting the validity of five new LTV features. The results achieved by our LTV data-driven approach using neural networks allowed an improvement up to 4 pp in the Lift cumulative curve for targeting the deposit subscribers when compared with a baseline model (with no history data). Explanatory knowledge was also extracted from the proposed model, revealing two highly relevant LTV features, the last result of the previous campaign to sell the same product and the frequency of past client successes. The obtained results are particularly valuable for contact center companies, which can improve pre- dictive performance without even having to ask for more information to the companies they serve.

A proactive intelligent decision support system for predicting the popularity of online news

"Lecture notes in computer science series, ISSN 0302-9743, vol. 9273"

Strategic planning of electricity systems: Integrating renewable energies

Tese de Doutoramento - Programa Doutoral em Engenharia Industrial e Sistemas (PDEIS)

Exploração de séries temporais em processos de previsão de vendas

Dissertação de mestrado em Engenharia de Sistemas

Consumption, wealth, stock and housing returns: evidence from emerging markets

We test the predictive ability of the transitory deviations of consumption from its common trend with aggregate wealth and labour income, cay, for both future equity and housing risk premia in emerging market economies. Using quarterly data for 31 markets, our country-level evidence shows that forecasting power of cay vis-à-vis stock returns is high for Brazil, China, Colombia, Israel, Korea, Latvia and Malaysia. As for housing returns, the empirical evidence suggests that financial and housing assets are perceived as complements in the case of Chile, Russia, South Africa and Thailand, and as substitutes in Argentina, Brazil, Hong Kong, Indonesia, Korea, Malaysia, Mexico and Taiwan. Using a panel econometric framework, we find that the cross-country heterogeneity observed in asset return predictability does not accrue to regional location, but can be attributed to differences in the degree of equity market development and in the level of income.

Exploração de dados para soluções de retalho em business intelligence

Dissertação de mestrado integrado em Engenharia de Gestão e Sistemas de Informação

Previsão de cheias em Barcelos

Dissertação de mestrado integrado em Engenharia Civil

Random walk forecast of urban water in Iran under uncertainty

There are two significant reasons for the uncertainties of water demand. On one hand, an evolving technological world is plagued with accelerated change in lifestyles and consumption patterns; and on the other hand, intensifying climate change. Therefore, with an uncertain future, what enables policymakers to define the state of water resources, which are affected by withdrawals and demands? Through a case study based on thirteen years of observation data in the Zayandeh Rud River basin in Isfahan province located in Iran, this paper forecasts a wide range of urban water demand possibilities in order to create a portfolio of plans which could be utilized by different water managers. A comparison and contrast of two existing methods are discussed, demonstrating the Random Walk Methodology, which will be referred to as the â On uncertainty pathâ , because it takes the uncertainties into account and can be recommended to managers. This On Uncertainty Path is composed of both dynamic forecasting method and system simulation. The outcomes show the advantage of such methods particularly for places that climate change will aggravate their water scarcity, such as Iran.

Optimização do processo de recolha de resíduos: desenvolvimento de ferramentas de investigação operacional para o problema de orientação de equipas com multi-restrições

Tese de Doutoramento em Engenharia Industrial e de Sistemas.