Estimating Construction Material Indices with ARIMA and Optimized NARNETs

Construction Industry operates relying on various key economic indicators. One of these indicators is material prices. On the other hand, cost is a key concern in all operations of the construction industry. In the uncertain conditions, reliable cost forecasts become an important source of information. Material cost is one of the key components of the overall cost of construction. In addition, cost overrun is a common problem in the construction industry, where nine out of ten construction projects face cost overrun. In order to carry out a successful cost management strategy and prevent cost overruns, it is very important to find reliable methods for the estimation of construction material prices. Material prices have a time dependent nature. In order to increase the foreseeability of the costs of construction materials, this study focuses on estimation of construction material indices through time series analysis. Two different types of analysis are implemented for estimation of the future values of construction material indices. The first method implemented was Autoregressive Integrated Moving Average (ARIMA), which is known to be successful in estimation of time series having a linear nature. The second method implemented was Non-Linear Autoregressive Neural Network (NARNET) which is known to be successful in modeling and estimating of series with non-linear components. The results have shown that depending on the nature of the series, both these methods can successfully and accurately estimate the future values of the indices. In addition, we found out that Optimal NARNET architectures which provide better accuracy in estimation of the series can be identified/discovered as result of grid search on NARNET hyperparameters.     

Approximation by Stancu–Chlodowsky polynomials

In this paper, we investigate convergence and approximation properties of a Chlodowsky type generalization of Stancu polynomials (we called Stancu–Chlodowsky polynomials).The rates of convergence of this generalization are obtained by means of modulus of continuity and by using the $K$-functional of Peetre. We also present and prove theorems on weighted approximation and the order of approximation of continuous functions by these operators on all positive semi-axis.     

Natural frequencies of a flat viaduct road part simply supported on two ends

Viaduct roads have wide application in big cities with high traffic loads, in order to decrease traffic density and to connect subways to highways. Viaduct roads are constructed using steel structures instead of concrete ones in areas of earthquake risks. The low weight of steel structures however causes problems such as vibration and noise. There is increasing demand especially in populated areas to suppress vibration and noise on highway roads for reducing noise-related environmental pollution. In this study, bending vibrations of rectangular plate viaduct roads, which are supported by six fixed elements of rectangular cross-sectional elements are considered. Natural frequencies are obtained using the Rayleigh-Ritz technique, finite elements analysis, experimentally and neural networks (NN).     

Instance-Based Regression by Partitioning Feature Projections

A new instance-based learning method is presented for regression problems with high-dimensional data. As an instance-based approach, the conventional method, KNN, is very popular for classification. Although KNN performs well on classification tasks, it does not perform as well on regression problems. We have developed a new instance-based method, called Regression by Partitioning Feature Projections (RPFP) which is designed to meet the requirement for a lazy method that achieves high levels of accuracy on regression problems. RPFP gives better performance than well-known eager approaches found in machine learning and statistics such as MARS, rule-based regression, and regression tree induction systems. The most important property of RPFP is that it is a projection-based approach that can handle interactions. We show that it outperforms existing eager or lazy approaches on many domains when there are many missing values in the training data.     

The Popov criterion and the inverse Nyquist method†

Kosonbrock's inverse Nyquist array (I.N.A.) theory for linear multivariable control Bystema with constant feedback elements is extended, to include systems lip to m nonlinear feedback elements, where the system has m inputs and m outputs. This extension is achieved by considering the Popov criterion for the most general case and through two further theorems. It shows that, as in the ease of Rosenbrock's I.N.A. theory, when certain auxiliary conditions are met with the help of suitable controllers, the design of multivariable controllers containing many non-linear feedbacks, can be based on the m frequency response loci corresponding to the diagonal entries of the open-loop inverse transfer function matrix. This leads to a simple design technique identical to the I.N.A. design method, suitable for use with a computer-aided design facility which permits a designer to use his intuitive understanding of transfer functions based on classical theory. The I.N.A. theory has been extended by Rosenbrock to cover systems having non-linear, time-varying feedback elements very recently.     

Trend Assessment by the Innovative-Şen Method

Hydro-meteorological time series may include trend components mostly due to climate change since about three to four decades. Trend detection and identification in a better and refined manner are among the major current research topics in water resources domain. Even though different methodologies can be found for trend detection in literature, two well-known procedures are the Mann-Kendall (MK) trend test and recently innovative-?en trend method, which provides different aspects of the trend. The theoretical basis and application of these two methods are completely different. The MK test gives a holistic monotonic trend without any categorization of the time series into a set of clusters, but the innovative-?en method is based on cluster and provides categorical trend behavior in a given time series. The main purpose of this paper is to provide important differences between these two approaches and their possible similarities. The applications of the two approaches are given for hydro-meteorological records including relative humidity, temperature, precipitation and runoff from Ergene drainage basin in the north-western part of Turkey. It is observed that although MK trend test does not show significant trend almost in all the cases, the innovative-?en approach yields trend categorizations as “very low”, “low”, “medium” “high” and “very high”, which should be taken into consideration in future flood (“very high”) and drought (“very low”) studies.