Output Decisions Research Articles

A Bayesian Approach to Estimate the Spatial Distribution of Crowdsourced Radiation Measurements around Fukushima

Citizen-led movements producing spatio-temporal big data are potential sources of useful information during hazards. Yet, the sampling of crowdsourced data is often opportunistic and the statistical variations in the datasets are not typically assessed. There is a scientific need to understand the characteristics and geostatistical variability of big spatial data from these diverse sources if they are to be used for decision making. Crowdsourced radiation measurements can be visualized as raw, often overlapping, points or processed for an aggregated comparison with traditional sources to confirm patterns of elevated radiation levels. However, crowdsourced data from citizen-led projects do not typically use a spatial sampling method so classical geostatistical techniques may not seamlessly be applied. Standard aggregation and interpolation methods were adapted to represent variance, sampling patterns, and the reliability of modeled trends. Finally, a Bayesian approach was used to model the spatial distribution of crowdsourced radiation measurements around Fukushima and quantify uncertainty introduced by the spatial data characteristics. Bayesian kriging of the crowdsourced data captures hotspots and the probabilistic approach could provide timely contextualized information that can improve situational awareness during hazards. This paper calls for the development of methods and metrics to clearly communicate spatial uncertainty by evaluating data characteristics, representing observational gaps and model error, and providing probabilistic outputs for decision making.

IGBT Open-Circuit Fault Diagnosis for MMC Submodules Based on Weighted-Amplitude Permutation Entropy and DS Evidence Fusion Theory

Modular multilevel converters (MMCs) have a complex structure and a large number of submodules (SMs). If there is a fault in one of the SMs, it will affect the reliable operation of the system. Therefore, rapid fault diagnosis and accurate fault positioning are crucial to ensuring the continuous operation of the system. However, the IGBT open-circuit faults in different submodules of MMCs have similar fault features, and the traditional fault feature extraction method cannot effectively extract the key features of the fault so as to accurately locate the faulty submodules. In response to this problem, this paper proposes a fault diagnosis method based on weighted-amplitude permutation entropy (WAPE) and DS evidence fusion theory. The simulation results show that WAPE has better feature extraction ability than basic permutation entropy, and the fused multiscale feature decision output has better diagnostic accuracy than the single-scale feature. Compared with traditional fault diagnosis methods, this method achieves the diagnosis of multiple fault types by collecting a single signal, which greatly reduces the number of samples and leads to higher diagnostic accuracy and faster diagnostic speed.

Deep convolutional tree-inspired network: a decision-tree-structured neural network for hierarchical fault diagnosis of bearings

The fault diagnosis of bearings is crucial in ensuring the reliability of rotating machinery. Deep neural networks have provided unprecedented opportunities to condition monitoring from a new perspective due to the powerful ability in learning fault-related knowledge. However, the inexplicability and low generalization ability of fault diagnosis models still bar them from the application. To address this issue, this paper explores a decision-tree-structured neural network, that is, the deep convolutional tree-inspired network (DCTN), for the hierarchical fault diagnosis of bearings. The proposed model effectively integrates the advantages of convolutional neural network (CNN) and decision tree methods by rebuilding the output decision layer of CNN according to the hierarchical structural characteristics of the decision tree, which is by no means a simple combination of the two models. The proposed DCTN model has unique advantages in 1) the hierarchical structure that can support more accuracy and comprehensive fault diagnosis, 2) the better interpretability of the model output with hierarchical decision making, and 3) more powerful generalization capabilities for the samples across fault severities. The multiclass fault diagnosis case and cross-severity fault diagnosis case are executed on a multicondition aeronautical bearing test rig. Experimental results can fully demonstrate the feasibility and superiority of the proposed method.

The moon, the ghetto and artificial intelligence: Reducing systemic racism in computational algorithms

Computational algorithms and automated decision making systems that include them offer potential to improve public policy and organizations. But computational algorithms based on biased data encode those biases into algorithms, models and their outputs. Systemic racism is institutionalized bias with respect to race, ethnicity and related attributes. Such bias is located in data that encode the results and outputs of decisions that have been discriminatory, in procedures and processes that may intentionally or unintentionally disadvantage people based on race, and in policies that may discriminate by race. Computational algorithms may exacerbate systemic racism if they are not designed, developed, and used–that is, enacted–with attention to identifying and remedying bias specific to race. Advancing social equity in digital governance requires systematic, ongoing efforts to assure that automated decision making systems, and their enactment in complex public organizational arrangements, are free from bias.

The economics of posted prices in a concentrated market where demand is uncertain

This paper analyses the theory of the optimal output decision for a firm whose policy is to post a non-negotiable price for a good or service in a concentrated market where the demand facing the firm is determined, in part, by a random variable. The theoretical findings are the opposite of those in competitive markets; Proposition 1 states that the optimal output of a risk-averse firm is expected to be larger than that of a risk-neutral firm if the expected payoff of its marginal profit is less than or equal to 1. Proposition 2 states that the optimal output of a risk-seeking firm is expected to be smaller than that of a risk-neutral firm if the expected payoff of its marginal profit is greater than 1.

Farm Efficiency and Productivity Growth: The Effect of Commodity Prices

Crop prices can affect farm productivity through input- output decisions. This study assesses the relationship between crop prices and productivity changes among a sample of Kansas farms. The changes in total factor productivity are evaluated using a nonparametric approach with a Malmquist productivity index and potential drivers of technical efficiency and productivity change are analyzed. Farms with higher leverage and greater diversification are likely to be more efficient and experience productivity change. Lower productivity occurred during years with higher crop prices, suggesting that innovation is more likely to occur when margins are tight.

Robust and resilient joint periodic maintenance planning and scheduling in a multi-factory network under uncertainty: A case study

The continuity of activity and competition in the global business market has urged organizations to move from centralized to decentralized structures and develop multi-factor production (MFP) networks. The design and implementation of a maintenance system are essential due to two reasons: first, increasing the life cycle of the equipment, and second, decreasing the probability of disruption in MFP networks. This study proposes a bi-objective optimization model for maintenance planning and scheduling in an MFP network. In the proposed model, how to perform, the performing agent of the maintenance operation, and the maintenance periods are determined based on the failure function in planning and scheduling phases, respectively. Besides, two strategies are proposed for MFP network resilience under disruption occurrence. The objective functions are minimizing maintenance costs and maximizing reliability. The augmented epsilon constraint method is utilized to obtain the Pareto front and trade off the objectives. Because of the parameters' inherent uncertainty, an effective robust programming approach is employed to control the uncertainty of input parameters and the conservatism level of output decisions. The CPLEX Solver can globally solve the proposed model in small and medium instances. A heuristic method based on the genetic algorithm is proposed for solving in the large-scaled sample. Finally, the proposed model is implemented for a case study of CNG stations, in which output results approve the model's applicability.

Path Following Optimization for an Underactuated USV Using Smoothly-Convergent Deep Reinforcement Learning

This paper aims to solve the path following problem for an underactuated unmanned-surface-vessel (USV) based on deep reinforcement learning (DRL). A smoothly-convergent DRL (SCDRL) method is proposed based on the deep Q network (DQN) and reinforcement learning. In this new method, an improved DQN structure was developed as a decision-making network to reduce the complexity of the control law for the path following of a three-degree of freedom USV model. An exploring function was proposed based on the adaptive gradient descent to extract the training knowledge for the DQN from the empirical data. In addition, a new reward function was designed to evaluate the output decisions of the DQN, and hence, to reinforce the decision-making network in controlling the USV path following. Numerical simulations were conducted to evaluate the performance of the proposed method. The analysis results demonstrate that the proposed SCDRL converges more smoothly than the traditional deep Q learning while the path following error of the SCDRL is comparable to existing methods. Thanks to good usability and generality of the proposed method for USV path following, it can be applied to practical applications.

Deep transfer learning based classification model for covid-19 using chest CT-scans

COVID-19 is an infectious and contagious virus. As of this writing, more than 160 million people have been infected since its emergence, including more than 125,000 in Algeria. In this work, We first collected a dataset of 4986 COVID and non-COVID images confirmed by RT-PCR tests at Tlemcen hospital in Algeria. Then we performed a transfer learning on deep learning models that got the best results on the ImageNet dataset, such as DenseNet121, DenseNet201, VGG16, VGG19, Inception Resnet-V2, and Xception, in order to conduct a comparative study. Therefore, We have proposed an explainable model based on the DenseNet201 architecture and the GradCam explanation algorithm to detect COVID-19 in chest CT images and explain the output decision. Experiments have shown promising results and proven that the introduced model can be beneficial for diagnosing and following up patients with COVID-19.

Fuzzy rule-based acceptance criterion in metaheuristic algorithms

Metaheuristic algorithms are solution approaches to solve optimization problems by repeating some algorithmic steps while searching the solution space. The strategy of the metaheuristic includes two basic tactics; exploration for escaping the local optimum and exploitation for the global optimum. The number of solutions while exploring the solution space can be used to classify metaheuristics. If the metaheuristic uses only one solution to generate a new solution, we call it the single-solution-based metaheuristic. Simulated annealing, iterated local search, adaptive large neighborhood search, iterated greedy, local search, and tabu search are examples of single-solution-based metaheuristics. Most of these metaheuristics use an acceptance criterion to whether accept the newly generated solution instead of the incumbent solution to escape from the local optimal. The most used acceptance criterion in the literature is the Metropolis criterion or simulated annealing-like acceptance criterion that decides whether accept the new solution by calculating its acceptance probability. In this study, we propose a fuzzy rule-based acceptance criterion that fuzzies the inputs of the metaheuristic within a fuzzy inference system to create the decision output about the acceptance. The proposed new acceptance criterion is compared with the well-known probabilistic approach in the experimental study with traveling salesman problem, multidimensional knapsack problem, single machine weighted earliness/tardiness problem, linear regression problem, and two continuous optimization problems. The statistical analyses reveal that the number of acceptance solutions while using the proposed acceptance criterion is less than the probabilistic one has but the metaheuristic performance increases with the proposed criterion. Additional analyses are made to explain why the fuzzy criterion convergences better to the global optimum than the probabilistic criterion.

Forecasting Natural Rubber Price in Malaysia by 2030

Natural rubber (NR) has recently become one of Malaysia's most important economic sectors. Despite, the price of Standard Malaysia Rubber 20 changes frequently. That is why it is important to develop a NR price forecasting model. Because there was a significant time lag between making output decisions and the actual output of the commodity in the market. The aim of this study is to determine the time series pattern for natural rubber price in Malaysia within 1995 until 2020 and to forecast the natural rubber price in Malaysia for 10 years ahead. The data used is from year 1995 until 2020 that were obtained from Malaysian Rubber Board (MRB). This study also used univariate forecasting like Naïve with Trend, Double Exponential Smoothing, Holt’s Winter and Autoregressive Integrated Moving Average (ARIMA). Then, the measurement error is used to determine the best method to forecast the future data. The measurement error that used in this study are Mean Absolute Error, Mean Squared Error, Root Mean Square Error, Mean Absolute Percentage Error and The Theil Inequality Coefficient. Result: The natural rubber price in Malaysia showed a trend pattern. Then, ARIMA is used to determine the forecast of natural rubber price for next 10 years since it has the lowest measurement error. Conclusion: There are volatility in the price of natural rubber in Malaysia over the next 10 years.

The impact of hedging on risk-averse agents’ output decisions

In this study, we develop a forward-looking hedging factor variant of the output Euler model for an economy featuring uncertainty and risk-averse agents. Our novel approach goes beyond the traditional prospective viewpoint characterized by output growth expectations. Using a hedging factor constructed as the net-long trading measure obtained from quarterly equity futures data, covering the period 1993 to 2020, we identify different channels through which uncertainty about macroeconomic fundamentals affects economic output. The results reveal that the hedging factor influences uncertainty via the market volatility channel. This study provides the first piece of empirical evidence indicating that risk-averse agents place meaningful and significant weight on the hedging factor in economic decisions. The findings, which are robust to various measures of output growth, indicate that the hedging factor is a relevant state variable that influences future output decisions under conditions of macroeconomic uncertainty.

Energy and Spectrum Efficient Blind Equalization With Unknown Constellation for Air-to-Ground Multipath UAV Communications

In unmanned aerial vehicle (UAV) communications, frequency-selective fading can severely deteriorate the quality of transmitted signal by generating undesired and disordered constellation diagrams due to scatters in the air-to-ground (ATG) mutipath channels. In this paper, we propose a low-overhead blind equalization method to combat frequency-selective fading in air-ground multipath UAV channels. Specifically, a pre-equalization method is proposed based on a constant modulus algorithm to restore the contour of the constellation diagram. Moreover, the similarity measure function and the difference measure function are derived using template matching to identify the constellation of M-ary quadrature amplitude modulation. Furthermore, we propose a weighted constant cross algorithm (WXA) to reduce the residual mean square error and construct a cross-shaped modulus value, by utilizing the statistical information of the identified normalized standard constellation diagrams and the equalizer output decision symbols' weighting value. The proposed method requires less information and no training sequences and pilots, therefore, if achieves energy and spectrum efficient ATG multipath UAV communications. Simulation results show that the proposed WXA algorithm can reduce the residual mean square error convergence value between -22dB and -25dB, making it very useful for the equalization of the frequency-selective fading channel in typical UAV communication scenarios.

Did OPEC change its behaviour after the November 2014 meeting?

We analyse the behaviour of OPEC as a group by formulating a theoretical model that encompasses the perfect competition model and various forms of the imperfect competition model. By confronting the theoretical model with quarterly data for the period from 1992 to 2013 within the context of a cointegrated vector autoregressive (CVAR) model, we find support for the imperfect competition hypothesis regarding the output decisions of OPEC. We also find that a dynamic equilibrium correction model with imperfect competition is stable in-sample. However, a forecasting exercise for the period from 2014 to 2018 reveals that the model breaks down following the November 2014 meeting at which OPEC decided to keep its supply unchanged despite the huge oil price drop in advance. The model systematically underpredicts OPEC production over the forecast period and by as much as 2.5 million barrels per day at the end of 2016. During 2018, however, the model forecasts OPEC production quite well. Our findings suggest that the behaviour of OPEC did indeed change significantly after the November 2014 meeting.

A multimodel deep learning algorithm to detect North Atlantic right whale up-calls.

We present a new method of detecting North Atlantic Right Whale (NARW) upcalls using a Multimodel Deep Learning (MMDL) algorithm. A MMDL detector is a classifier that embodies Convolutional Neural Networks (CNNs) and Stacked Auto Encoders (SAEs) and a fusion classifier to evaluate their output for a final decision. The MMDL detector aims for diversity in the choice of the classifier so that its architecture is learned to fit the data. Spectrograms and scalograms of signals from passive acoustic sensors are used to train the MMDL detector. Guided by previous applications, we trained CNNs with spectrograms and SAEs with scalograms. Outputs from individual models were evaluated by the fusion classifier. The results obtained from the MMDL algorithm were compared to those obtained from conventional machine learning algorithms trained with handcrafted features. It showed the superiority of the MMDL algorithm in terms of the upcall detection rate, non-upcall detection rate, and false alarm rate. The autonomy of the MMDL detector has immediate application to the effective monitoring and protection of one of the most endangered species in the world where accurate call detection of a low-density species is critical, especially in environments of high acoustic-masking.

Multi-source transfer learning network to complement knowledge for intelligent diagnosis of machines with unseen faults

Most of the current successes of deep transfer learning-based fault diagnosis require two assumptions: 1) the health state set of source machines should overlap that of target machines; 2) the number of target machine samples is balanced across health states. However, such assumptions are unrealistic in engineering scenarios, where target machines suffer from fault types that are not seen in source machines and the target machines are mostly in a healthy state with only occasional faults. As a result, the diagnostic knowledge from source machines may not cover all fault types of target machines nor address imbalanced target samples. Therefore, we propose a framework, called a multi-source transfer learning network (MSTLN), to aggregate and transfer diagnostic knowledge from multiple source machines by combining multiple partial distribution adaptation sub-networks (PDA-Subnets) and a multi-source diagnostic knowledge fusion module. The former weights target samples by counter-balancing factors to jointly adapt partial distributions of source and target pairs, and the latter releases negative effects due to discrepancy among multiple source machines and further fuses diagnostic decisions output from multiple PDA-Subnets. Two case studies demonstrate that MSTLN can reduce the misdiagnosis rate and obtain better transfer performance for imbalanced target samples than other conventional methods.

Emission trading systems and the optimal technology mix

Cap and trade mechanisms enjoy increasing importance in environmental legislations worldwide. One of the important aspects of designing cap and trade mechanisms is the possibility of authorities to grant emission permits for free. Unlike analyzed in the seminal contributions on cap and trade systems, in reality free allocations are not made lump sum, but are updated contingent on firms’ actions, i.e., contingent on production decisions and contingent on facilities entering the market or retiring (see IEA, https://www.oecd-ilibrary.org/content/publication/b7d0842b-en, 2020). As it has already been shown in the literature, such updating yields distorted production decisions of firms (see e.g., Böhringer and Lange in Eur Econ Rev 49(8):2041–2055, 2005, Mackenzie et al. in Environ Resour Econ 39(3):265–282, 2008, or Damon et al. in Rev Environ Econ Policy 13(1):23–42, 2019). The impact of updating on firms’ investment and retiring decisions and the resulting technology mix has received much less attention up to now, however. It is the purpose of the present article to shed light on this aspect and to study the impact of a cap and trade mechanism not only on firms’ output decisions, but also on their investment incentives in different technologies and to analyze the optimal design of emission trading systems in such an environment.

The Flexible Possibilistic-Robust Mathematical Programming Approach for the Resilient Supply Chain Network: An Operational Plan

Nowadays, the management of soil resources for global food security is one of the strategic purposes of the countries. In this regard, creating soil nutrition’s supply chain network for the fertility of agricultural fields is one of the most critical priorities of the government. In the above strategic network, we are faced with uncertain parameters that show a complex behavior under time, geography, and market changes. This parametric uncertainty is aggravated by the occurrence of disruptions such as sanctions and natural disasters, with devastating effects on strategic, tactical, and operational decisions. In this paper, by adopting a set of reactive and preventive strategies, the resilient supply chain network of dependent petrochemical industries is presented, consistent with the structure of its production cycle in Iran. To cope with the uncertainty, a robust programming approach was extended into flexible-possibilistic programming. The final model controls the uncertainty and risk-aversion of output decisions and effectively confronts the adverse effects of disruptions. Finally, using a real industrial study, the effectiveness and applicability of the final model are explained. Furthermore, by providing efficient solutions, mean cost and minimum standard deviations were achieved.

Development of The Multi-Level Fusion based Security System

The article develops the multi-level fusion based intellig ent security system to be applied in home automation. The system contains some detection modules, a supervised computer, an elevator and an experimental platform. The detection modules have active detection modules and passive detection modules that are designed in our research team. The control unit of all detection modules (active and passive) is HOLTEK microchip. Mobile robots are active detection modules, and are classified more and more teams acco rding to the detection functions. Each mobile robot of the team robot system transmits the real-time event signals to the supervi sed computer and the other mobile robots via wireless RF interface, and moves to any floor of the experimental platform using the elevator. The passive detection modules contain wire/wireless detection modules. If the event occurrence, the passive detection modules can decide the event to be true or false in the level one using weighted average algorithm, and transmits the position of the event to the supervised computer and mobile robots. Mobile robots can detect the event to be true or false using Dempster-Shafter (D-S) evidence theory, and transmits the relation location of the event to the supervised computer, too. The supervised computer decides the event to be true according to the feedback signals using D-S evidence theory, and controls the other mobile robots moving to the event location for double check, and recognizes the final decision output using D-S evidence theo ry in the level two. Finally, we present some experimental results using the active and passive detection modules to detect the fire sourc e on the experimental platform

Developing Bayesian networks in managing the risk of Legionella colonisation of groundwater aeration systems

An Australian water utility has developed a Legionella High Level Risk Assessment (LHLRA) which provides a semi-qualitative assessment of the risk of Legionella proliferation and human exposure in engineered water systems using a combination of empirical observation and expert knowledge. Expanding on this LHLRA, we propose two iterative Bayesian network (BN) models to reduce uncertainty and allow for a probabilistic representation of the mechanistic interaction of the variables, built using data from 25 groundwater treatment plants. The risk of Legionella exposure in groundwater aeration units was quantified as a function of five critical areas including hydraulic conditions, nutrient availability and growth, water quality, system design (and maintenance), and location and access. First, the mechanistic relationship of the variables was conceptually mapped into a fishbone diagram, parameterised deterministically using an expert elicited weighted scoring system and translated into BN. The “sensitivity to findings” analysis of the BN indicated that system design was the most influential variable while elemental accumulation thresholds were the least influential variable for Legionella exposure. The diagnostic inference was used in high and low-risk scenarios to demonstrate the capabilities of the BNs to examine probable causes for diverse conditions. Subsequently, the causal relationship of Legionella growth and human exposure were improved through a conceptual bowtie representation. Finally, an improved model developed the predictors of Legionella growth and the risk of human exposure through the interaction of operational, water quality monitoring, operational parameters, and asset conditions. The use of BNs modelling based on risk estimation and improved functional decision outputs offer a complementary and more transparent alternative approach to quantitative analysis of uncertainties than the current LHLRA.