According to recent studies, unmanned aerial vehicles can play a game-changing part in cost reduction and speed increase address the last-mile delivery problem attend emergencies. Last-mile services are becoming more relevant, especially when times social distance is required. However, systemic view of how provide feasible solution enable this application as vertical market urban context still inexistent. Given scenario, article contributes by proposing cyber-physical system roadmap...

Unmanned Aerial Vehicle (UAV) swarm, also named drone has been the study object of many types research due to its potential improve applications such as monitoring, surveillance, and search missions. With several drones flying simultaneously, challenge is increase their level automation intelligence while avoiding collision, reducing communication with these entities, improving strategical organization accomplish a specific task. In this sense, we propose solution coordinate UAV swarm using...

The connection between Singular Spectrum Analysis (SSA) decomposition and short-term market movements is investigated. Since SSA a non-parametric approach, suitable to decompose general time-series into meaningful components, such as trends, oscillations noise, it proposed new oscillator-type Technical Indicator, replacing popular ones. New Trading Rules (TTRs) are designed applied some major global stock indexes illustrate the benefits in terms of revealing movements. performance evaluated...

Drones can play a game-changing role in reducing both cost and time the context of last-mile deliveries.This paper addresses delivery problem from complex system viewpoint, where collective performance drones is investigated.We consider with tradable permit model (TPM) for airspace use.Typically, other research works regarding drones, fully cooperative centralized scenario contemplated.In our approach, due to TPM, agents (i.e.drones) need compete permits distributed manner.We simulate...

This paper proposes a trading system (TS) based on multicriteria decision aid (MCDA) process known as TODIM, (Multicriteria Interactive Decision Making) Portuguese acronym for interactive and decision-making. MCDA has been employed to solve financial questions because of its ability deal with complex environment populated different sorts criteria alternatives, such markets. The aim is propose general adaptive tool supporting the strategies investors market practitioners in an environment....

In this paper, we introduce a novel method to select the components of portfolio securities. This is based on reinforcement learning technique known as automata. Several heuristic solutions for weights selection problem have been introduced in literature. The point that these applications assumes are given. difference our work propose some order instead weights. terms heuristic, automata because its ability solve complex systems such optimal components. We test use financial indicators...

A new trading system (TS) is proposed in this paper, based on singular spectrum analysis (SSA) and a decision-making process that relies performance measures adjusted by profit risk. The aim to contribute the literature of technical (TA) proposing general adaptive tool for supporting strategies investors market practitioners. Recent advances rules SSA are pooled together complimentary fashion, with intention group outperforms its individual constituents, as occur ensembles. Simulations...

Drone swarms are being utilized in several areas, e.g., monitoring, surveillance, security, and search, demonstrating great potential. In this paper, we considered a specific number of drones flying simultaneously (using computational intelligence techniques between them to avoid collisions) increased the effectiveness conducting search tasks. The main contribution work is propose technique coordinate swarm using methods based on A* algorithm its heuristics (i.e., Euclidean Manhattan...

A comparison between Moving Averages (MA) and two versions of Singular Spectrum Analysis (SSA) methodology - the Caterpillar Toeplitz is presented. had already been studied in this manner but same not true for SSA. SSA assumes stationarity time-series, which means that process needs to be mean-reverting. However, such assumption a necessary condition In paper both approaches are applied trend estimation problem order used as an indicator trend-following technical rules design. Similarities...

Over the past years, there has been a growing interest in cryptocurrency markets. In this context, price forecasting initiatives that aid decision-making process of investors and market participants have emerged drawn academia financial technology industry. paper, we present machine learning classification model forecasts direction - top, modeled as 1, or neutral bottom, 0 Litecoin (LTC) over forecast horizon equivalent to volume-wise samples 100 thousand LTC. For modeling, adopt random...

This work intends to replicate the interactions among intelligent agents in a double auction environment, leading an artificial market. Aiming at making able take their own decisions, different methodologies for intelligence are proposed, ranging from zero reinforcement learning. Regarding study of financial markets, can rely on myriad techniques making-decision process. In this manuscript, Fundamental and/or Technical Analysis tools used by decisions. A set parameters is assigned each...

Singular Spectrum Analysis (SSA) is a nonparametric approach that can be used to decompose time series as trends, oscillations and noise. In online applications, the SSA algorithm must recalculated for each new sample available. The so called Causal have been in this context. manuscript, an alternative version of SSA, Recursive proposed technical indicator. Based on "forgetting factor" parameter, it possible control amount previous samples are algorithm. This functionality may confer...

The central theme of this paper is the prioritization failing equipment using a Multicriteria Decision Aid method. Maintenance decision criteria needed in productive processes with several similar machines. This based on inherent characteristics production process and failed equipment. PROMETHÉÉ II method proposed paper, has characteristic being non-compensatory allows pairwise comparison alternatives, resulting an order alternatives. used were previously defined by specialists along...

This manuscript introduces a Matlab computational framework for multiagent system simulation of financial markets. The main motivation using is that this platform very familiar among researches in the area engineering. could motivate researchers to investigate area. aim our replicate interactions agents through double auction environment, leading an artificial market. Every agent assigned with unlimited credit, and short selling allowed. These two degrees freedom might ensure particular able...

Artificial multi-agent systems can be applied to replicate complex processes, including the stock market. In this manuscript, an artificial market was simulated. Heterogeneous agents are set trade by using different investiment strategies through a double auction The main innovative aspect of work is evaluate performance and effectiveness with investment based on combination Technical/Fundamental Analysis. According simulations performed, results showed that time series resulting from...

Recently, there has been an increasing interest in applying cognitive modelling approaches towards the understanding of behaviour agents complex systems. In this paper we employ a approach representative instance such systems namely El Farol bar problem. We establish relationship between and strategies, by defining degree belief correctness devise concept rationality inspired emotion theories, conceptualizing threshold which must change its current strategy. investigate whether larger memory...

In this paper we introduce a novel trading strategy for in continuous-double auctions risky assets. The principle underlying our is an optimising of the level leverage using numerical methods. Previous studies have shown that similar strategies can perform well when tested against theoretical models asset price processes such as geometric Brownian motion, or back-tested historical empirical data. However, former approach fails to account phenomena non-Gaussian return distributions which are...

The objective of this work is to analyze the productivity different Brazilian research promotion agencies by using Method Interactive Decision Making and Multicriteria (Acronym TODIM in Portuguese), which based on Prospect Theory. proposed model classify 22 (two federal twenty state agencies) under six criteria. result shows that some small are more efficient than bigger ones terms funding/scholarship ratio. This suggests winner could be further investigated promoted among other country.

Singular Spectrum Analysis (SSA) is a non parametric approach that can be used to decompose time series into trends, oscillations and noise. In online applications, the SSA algorithm must recalculated for each new sample available. The so called Causal have been in this context. This paper aims contribute application of as an indicator trading rules design by means comparing them previously proposed ones based on classic indicators. Using historical from some major Brazilian stocks,...

The Foreign Exchange (Forex) is the largest market in world and has a daily trading volume of approximately 3.2 trillion dollars. price movements are influenced by many exogenous factors, thereby it difficult to predict. As result, modeling its movement would enable high profitable investment strategies. Aiming at setting up model that helps practitioner make better decisions for on Forex, Machine Learning algorithms (Random Forest SVM) adopted this work. Classic technical indicators, like...

The stock price movements result from many factors that are often difficult to be detected and modelled. investigation of trends the use information available evaluate investments identify trading opportunities can promising. However, financial data non-stationary, i.e., their statistical characteristics constantly change. Therefore, market is a challenging environment for application Machine Learning techniques, since they only make reliable predictions consistent with what have seen...