Artificial intelligence—morphological approach as a new technological forecasting technique

Technological forecasting is located in technological space for a certain engineering domain and is the first step in the research and development of a new technology. There are many techniques already developed for technological forecasting like extrapolation of trends, heuristics forecasting by expert e.g. Delphi method, etc. The morphological method (MM) is used to describe a technique for identifying, indexing, counting and parametrizing the collection of all possible devices to achieve a specified functional capability. This is not a forecasting per se, but it is a useful organizing tool, a source of insights, and a starting point for further analysis by other methods, Mila?i? (1976). In the paper we expand the capabilities of MM using artificial intelligence (Al) principles. The AI production system has a global data base, a set of production rules and a control system. Special attention is paid to control strategy-developing searching techniques in order to distinguish the explicit local knowledge about how to proceed toward a goal from any state from the implicit global knowledge of the complete solution. Some practical examples are also shown.     

产品技术进化潜力预测研究

  产品技术预测是企业长久保持竞争优势的有效方法,发明问题解决理论（TRIZ）的技术进化理论能够对技术预测提供理论指导．在介绍TRIZ的技术进化模式、技术进化路线的基础上,提出如何搜索产品技术进化潜力及构建进化潜力雷达图的方法,并对产品技术进化潜力预测技术作了系统的总结．基于产品技术进化潜力预测技术开发了专用软件——产品技术进化潜力预测系统（EPMS）,可帮助企业快速、准确地把握产品的技术走向,并应用软件EPMS对蝶阀密封技术的进化潜力进行了预测．     

A reliable decision support system for fresh food supply chain management

The paper proposes a decision support system (DSS) for the supply chain of packaged fresh and highly perishable products. The DSS combines a unique tool for sales forecasting with order planning which includes an individual model selection system equipped with ARIMA, ARIMAX and transfer function forecasting model families, the latter two accounting for the impact of prices. Forecasting model parameters are chosen via two alternative tuning algorithms: a two-step statistical analysis, and a sequential parameter optimisation framework for automatic parameter tuning. The DSS selects the model to apply according to user-defined performance criteria. Then, it considers sales forecasting as a proxy of expected demand and uses it as input for a multi-objective optimisation algorithm that defines a set of non-dominated order proposals with respect to outdating, shortage, freshness of products and residual stock. A set of real data and a benchmark – based on the methods already in use – are employed to evaluate the performance of the proposed DSS. The analysis of different configurations shows that the DSS is suitable for the problem under investigation; in particular, the DSS ensures acceptable forecasting errors and proper computational effort, providing order plans with associated satisfactory performances.     

最优组合预测方法在家用汽车需求预测中的应用

为了提高预测的准确性,引入了组合预测模型,将几个单一预测模型有机地结合起来,综合各个预测模型的优点,对未来几年内家用轿车需求进行预测.通过使组合预测误差平方和最小,确定各个单一预测方法的权重系数,得出更为准确的预测结果.计算结果表明该方法具有较好的实用性.     

Deep learning with long short-term memory networks and random forests for demand forecasting in multi-channel retail

This paper proposes a novel forecasting method that combines the deep learning method – long short-term memory (LSTM) networks and random forest (RF). The proposed method can model complex relationships of both temporal and regression type which gives it an edge in accuracy over other forecasting methods. We evaluated the new method on a real-world multivariate dataset from a multi-channel retailer. We benchmark the forecasting performance of the new proposition against neural networks, multiple regression, ARIMAX, LSTM networks, and RF. We employed forecasting performance metrics to measure bias, accuracy, and variance, and the empirical evidence suggests that the new proposition is (statistically) significantly better. Furthermore, our method ranks the explanatory variables in terms of their relative importance. The empirical evaluations are replicated for longer forecasting horizons, and online and offline channels and the same conclusions hold; thus, advocating for the robustness of our forecasting proposition as well as the suitability in multi-channel retail demand forecasting.     

On‐line reliability estimation for individual components using statistical degradation signal models

With very few exceptions, most contemporary reliability engineering methods are geared towards estimating a population characteristic(s) of a system, subsystem or component. The information so extracted is extremely valuable for manufacturers and others that deal with product in relatively large volumes. In contrast, end users are typically more interested in the behavior of a ‘particular’ component used in their system to arrive at optimal component replacement or maintenance strategies leading to improved system utilization, while reducing risk and maintenance costs. The traditional approach to addressing this need is to monitor the component through degradation signals and ‘classifying’ the state of a component into discrete classes, say ‘good’, ‘bad’ and ‘in‐between’ categories. In the event, one can develop effective degradation signal forecasting models and precisely define component failure in the degradation signal space, then, one can move beyond the classification approach to a more vigorous reliability estimation and forecasting scheme for the individual unit. This paper demonstrates the feasibility of such an approach using ‘general’ polynomial regression models for degradation signal modeling. The proposed methods allow first‐order autocorrelation in the residuals as well as weighted regression. Parametric bootstrap techniques are used for calculating confidence intervals for the estimated reliability. The proposed method is evaluated on a cutting tool monitoring problem. In particular, the method is used to monitor high‐speed steel drill‐bits used for drilling holes in stainless‐steel metal plates. A second study involves modeling and forecasting fatigue‐crack‐growth data from the literature. The task involved estimating and forecasting the reliability of plates expected to fail due to fatigue‐crack‐growth. Both studies reveal very promising results. Copyright © 2002 John Wiley & Sons, Ltd.     

Combined particle swarm optimization and heuristic fuzzy inference systems for a smart home one-step-ahead load forecasting

This article presents a hybrid approach that combines particle swarm optimization (PSO) and heuristic fuzzy inference system (HFIS) for smart home one-step-ahead load forecasting. Smart home load forecasting is an important issue in the development of smart grids. Generally, the electricity consumption of a household is inherently nonlinear and dynamic and heavily dependent on the habitual nature of power demand, activities of daily living and on holidays or weekends, so it is often difficult to construct an adequate forecasting model for this type of load. To address this problem, a hybrid model, consisting of two phases, is proposed in this article. In the first phase, the popular PSO algorithm is used to determine the locations of fuzzy membership functions. Then, the proposed HFIS technique is used to develop the one-step-ahead load forecasting model in the second phase. Because of the robust nature of the proposed HFIS technique, which does not need to retrain or re-estimate model parameters, it is very suitable for smart home load forecasting. The proposed method was verified using two different households’ load data. Simulation results indicate that the proposed method produces better forecasting accuracy than existing methods.     

Using grey models for forecasting China’s growth trends in renewable energy consumption

This study primarily investigated the forecasting of the growth trend in renewable energy consumption in China. Only 22 samples were acquired for this study because renewable energy is an emerging technology. Because historical data regarding renewable energy were limited in sample size and the data were not normally distributed, forecasting methods used for analyzing large amounts of data were unsuitable for this study. Grey system theory is applied to system models involving incomplete information, unclear behavioral patterns, and unclear operating mechanisms. In addition, it can be used to perform comprehensive analyses, observe developments and changes in systems, and conduct long-term forecasts. The most prominent feature of this theory is that a minimum of only four data sets are required for establishing a model and that making stringent assumptions regarding the distribution of the sample population is not required. However, to address the limitations of previous studies on grey forecasting and to enhance the forecasting accuracy, this study adopted the grey model (1, 1) [GM(1, 1)] and the nonlinear grey Bernoulli model (1, 1) [(NGBM)] for theoretical derivation and verification. Subsequently, the two models were compared with a regression analysis model to determine the models’ predictive accuracy and goodness of fit. According to the indexes of mean absolute error, mean square error, and mean absolute percentage error, NGBM(1, 1) exhibited the most accurate forecasts, followed by GM(1, 1) and regression analysis model. The results indicated that the modified NGBM(1, 1) grey forecasting models demonstrated superior predictive abilities among the compared models.     

Forecasting the state of ground control and measurement facilities

Methods are proposed for forecasting the technical state of ground control and measurement facilities, which are modifications of standard methods: guaranteed forecasting and a method based on statistical simulation. Results are presented from practical tests on those methods, which demonstrate convergence and good forecasting. Translated from Izmeritel'naya Tekhnika, No. 9, pp. 3–6, September, 1996.     

Selecting exponential smoothing constants: an application of pattern search

Exponential smoothing is one means of preparing short-term sales forecasts on a routine basis. To use exponential smoothing, however, one must decide the proper values for the smoothing constants in the forecasting model. One method for selecting the smoothing constants involves conducting a grid search to evaluate a wide range of possible values. An alternative method, called pattern search, is presented in this paper and is compared to the grid search procedure. Two criteria are used in comparing these procedures: (1) the standard deviation of the forecast error, and (2) the computing time necessary for a solution, The results of this comparison indicate that, when the exponential smoothing model includes trend and seasonal adjustments, pattern search requires far less computer time than the grid search procedure to produce smoothing constant values for which the forecast error values are comparable.     

One step-ahead ANFIS time series model for forecasting electricity loads

In electric industry, electricity loads forecasting has become more and more important, because demand quantity is a major determinant in electricity supply strategy. Furthermore, accurate regional loads forecasting is one of principal factors for electric industry to improve the management performance. Recently, time series analysis and statistical methods have been developed for electricity loads forecasting. However, there are two drawbacks in the past forecasting models: (1) conventional statistical methods, such as regression models are unable to deal with the nonlinear relationships well, because of electricity loads are known to be nonlinear; and (2) the rules generated from conventional statistical methods (i.e., ARIMA), and artificial intelligence technologies (i.e., support vector machines (SVM) and artificial neural networks (ANN)) are not easily comprehensive for policy-maker. Based on these reasons above, this paper proposes a new model, which incorporates one step-ahead concept into adaptive-network-based fuzzy inference system (ANFIS) to build a fusion ANFIS model and enhances forecasting for electricity loads by adaptive forecasting equation. The fuzzy if-then rules produced from fusion ANFIS model, which can be understood for human recognition, and the adaptive network in fusion ANFIS model can deal with the nonlinear relationships. This study optimizes the proposed model by adaptive network and adaptive forecasting equation to improve electricity loads forecasting accuracy. To evaluate forecasting performances, six different models are used as comparison models. The experimental results indicate that the proposed model is superior to the listing models in terms of mean absolute percentage errors (MAPE).     

Comparison of Missing Data Imputation Methods in Time Series Forecasting

Time series forecasting has become an important aspect of data analysis and has many real-world applications. However, undesirable missing values are often encountered, which may adversely affect many forecasting tasks. In this study, we evaluate and compare the effects of imputation methods for estimating missing values in a time series. Our approach does not include a simulation to generate pseudo-missing data, but instead perform imputation on actual missing data and measure the performance of the forecasting model created therefrom. In an experiment, therefore, several time series forecasting models are trained using different training datasets prepared using each imputation method. Subsequently, the performance of the imputation methods is evaluated by comparing the accuracy of the forecasting models. The results obtained from a total of four experimental cases show that the -nearest neighbor technique is the most effective in reconstructing missing data and contributes positively to time series forecasting compared with other imputation methods.     

Comparative assessment of forecasts

A novel analysis method for deciding upon, and improving, the comparative goodness of competing forecast methods is described. The approach is based upon the centred forecast observation diagram of change, which has been extended to include measures of analysis and supplemented with the recently developed measures of comparative error described in this contribution. The method has application in any situation where there is a need to undertake comparison of competing forecasting methods or where comparative assessments of a given variable are to be undertaken. The method is simple to understand and apply and has proven to be extremely useful in improving focus on deficiencies in forecasting methods.     

Reconstructing the parameters of geofields taking measurement noise into account. Part I. Techniques for smoothing,filtration, and regularization

Various methods of interpolation (approximation, smoothing, and filtering types) which take measurement noise into account (piecewise polynomial approximation, smoothing splines, kriging, filters, etc.) are analyzed. It is found that regularization methods are used in all of them. The advantages and disadvantages of these methods are pointed out and prospects for the use of program packages by nonspecialists in mathematics and information technology are examined. Recommendations regarding the proper choice of estimation techniques, quality criteria, and parameters for monitoring them are made in the case where geofields are to be approximated and modelled.     

以TRIZ进化理论为基础的产品技术预测支持系统研究

产品技术预测是使企业保持长久竞争优势的有效方法之一,而TRIZ的技术进化理论恰恰是众多产品技术预测理论中最具优势和生命力的。以TRIZ的技术进化理论为基础,运用先进的计算机软件编程技术--基于Windows2000环境, 以Visual Basic 6.0为开发语言,Access作为后台数据库,采用DAO数据库访问技术,构建了基于TRIZ进化理论的产品技术预测支持系统。该系统由8个模块组成,分别是系统维护、待预测产品数据库、产品技术成熟度预测、技术进化路线分析、技术进化知识库、工程报表、编辑和学习帮助模块,通过系统的应用可以实现产品技术成熟度的预测分析,为产品提供可能的技术进化模式或进化路线,将以前手工完成的工作软件化,从而使企业快速、准确地把握产品的技术走向,实现产品和技术的创新。     

An integrated method for interdisciplinary topic identification and prediction: a case study on information science and library science

Given that many frontiers and hotspots of science and technology are emerging from interdisciplines, the accurate identification and forecasting of interdisciplinary topics has become increasingly significant. Existing methods of interdisciplinary topic identification have their respective application fields, and each identification result can help researchers acquire partial characteristics of interdisciplinary topics. This paper offers an integrated method for identifying and predicting interdisciplinary topics from scientific literature. It integrates various methods, including co-occurrence networks analysis, high-TI terms analysis and burst detection, and offers an overall perspective into interdisciplinary topic identification. The results of the different methods are mutually confirmed and complemented, further overviewing the characteristics of the interdisciplinary field and highlighting the importance or potential of interdisciplinary topics. In this study, Information Science and Library Science is selected as a case study. The research has clearly shown that more accurate and comprehensive results can be achieved for interdisciplinary topic identification and prediction by employing this integrated method. Further, the integration of different methods has promising potential for application in knowledge discovery and scientific measurement in the future.     

车内噪声声品质的支持向量机预测

对多元线性回归、神经网络和支持向量机的三个预测模型进行了研究。以车内噪声为例,建立了基于以上三种方法的车内噪声声品质预测模型,并采用留一法交叉检验作比较,所构建的支持向量机模型预测精度高于其他两种方法。实验结果同时也表明,支持向量计算法具有较强的稳健性和良好的泛化能力,能够用于车内噪声声品质的预测。     

A Hybrid Neural Network and Box-Jenkins Models for Time Series Forecasting

Time series forecasting plays a significant role in numerous applications, including but not limited to, industrial planning, water consumption, medical domains, exchange rates and consumer price index. The main problem is insufficient forecasting accuracy. The present study proposes a hybrid forecasting methods to address this need. The proposed method includes three models. The first model is based on the autoregressive integrated moving average (ARIMA) statistical model; the second model is a back propagation neural network (BPNN) with adaptive slope and momentum parameters; and the third model is a hybridization between ARIMA and BPNN (ARIMA/BPNN) and artificial neural networks and ARIMA (ARIMA/ANN) to gain the benefits of linear and nonlinear modeling. The forecasting models proposed in this study are used to predict the indices of the consumer price index (CPI), and predict the expected number of cancer patients in the Ibb Province in Yemen. Statistical standard measures used to evaluate the proposed method include (i) mean square error, (ii) mean absolute error, (iii) root mean square error, and (iv) mean absolute percentage error. Based on the computational results, the improvement rate of forecasting the CPI dataset was 5%, 71%, and 4% for ARIMA/BPNN model, ARIMA/ANN model, and BPNN model respectively; while the result for cancer patients’ dataset was 7%, 200%, and 19% for ARIMA/BPNN model, ARIMA/ANN model, and BPNN model respectively. Therefore, it is obvious that the proposed method reduced the randomness degree, and the alterations affected the time series with data non-linearity. The ARIMA/ANN model outperformed each of its components when it was applied separately in terms of increasing the accuracy of forecasting and decreasing the overall errors of forecasting.

     

Optik braucht Vakuum

Optics needs Vacuum — Challenges in Vacuum Optics Many applications in high technology use optical signals, optical information or laser radiation in vacuum as a part of manufacturing or characterization processes. These optical signals often have to be transferred into or out of a vacuum system. This article shows basic requirements and specifications for assemblies that connect optical elements and vacuum components hermetically. Four different, common joining methods are presented that fulfill both optical and vacuum requirements. The ability of these joining methods to preserve the properties of the individual components is discussed. After pointing out strengths and weaknesses of the joining methods, typical specifications profiles for each method are derived. It is concluded, that vacuum optics holds proper solutions for every application and its specific requirements.     

Identifying risks and emergent risks across sociotechnical systems: the NETworked hazard analysis and risk management system (NET-HARMS)

Accidents are a systems phenomenon and multiple methods are available to enable retrospective analysis of accidents through this lens. However, the same cannot be said for the methods available for forecasting risk and accidents. This paper describes a new systems-based risk assessment method, the NETworked hazard analysis and risk management system (NET-HARMS), that was designed to support practitioners in identifying (1) risks across overall work systems, and (2) emergent risks that are created when risks across the system interact with one another. An overview of NET-HARMS is provided and demonstrated through a case study application. An initial test of the method is provided by comparing case study outcomes (i.e. predicted risks) with accident data (i.e. actual risks) from the domain in question. Findings show that NET-HARMS is capable of forecasting systemic and emergent risks and that it could identify almost all risks that featured in the accidents in the comparison data-set.