Detailed study on fusion characteristics of rigid poly(vinyl chloride) nanocomposites: The comparison of using multiple regression analysis and artificial neural network

Fusion behavior of poly(vinyl chloride) (PVC) compounds plays an important role in the development of physical properties of processed material. The fusion characteristics in PVC processing are governed by material variables that affect the fusion with some interactions. In this research, the aim was to characterize the effects of formulation ingredients on fusion characteristics of PVC. Four material parameters, including the contents of nanoclay (NC), azodicarbonamide, calcium stearate, and processing aid, are proposed as affecting variables. The fusion time (FT) as well as fusion factor (FF) are considered fusion indicators and are experimentally determined in some different levels of affecting parameters. The multivariable regression analysis (MRA) and the Artificial Neural Network (ANN) modeling are considered as two analytical methods. The regression analysis result for the FT denotes, in part, significant linear and quadratic effects of NC and also its significant interactions with azodicarbonamide and calcium stearate, whereas that of FF indicates only a linear effect of NC. ANN modeling is performed with a three‐layer (input, hidden, and output) neural network. The results of the comparison of the MRA and ANN predictions with experimental values are reported as the correlation coefficient (R²), mean‐square error, and mean absolute percentage error for both FF and FT parameters. The obtained values clearly denote that the ANN results are more precise and especially more general than those of MRA. However, in the case of FT, improvement of the ANN modeling is much greater than that of FF. J. VINYL ADDIT. TECHNOL., 21:147–155, 2015. © 2014 Society of Plastics Engineers     

D-side: A facility and workforce planning group multi-criteria decision support system for Johnson Space Center

“To understand and protect our home planet, to explore the universe and search for life, and to inspire the next generation of explorers” is NASA's mission. The Systems Management Office at Johnson Space Center (JSC) is searching for methods to effectively manage the Center's resources to meet NASA's mission. D-Side is a group multi-criteria decision support system (GMDSS) developed to support facility decisions at JSC. D-Side uses a series of sequential and structured processes to plot facilities in a three-dimensional (3-D) graph on the basis of each facility's alignment with NASA's mission and goals, the extent to which other facilities are dependent on the facility, and the dollar value of capital investments that have been postponed at the facility relative to the facility's replacement value. A similarity factor rank orders facilities based on their Euclidean distance from Ideal and Nadir points. These similarity factors are then used to allocate capital improvement resources across facilities. We also present a parallel model that can be used to support decisions concerning allocation of human resources investments across workforce units. Finally, we present results from a pilot study where 12 experienced facility managers from NASA used D-Side and the organization's current approach to rank order and allocate funds for capital improvement across 20 facilities. Users evaluated D-Side favorably in terms of ease of use, the quality of the decision-making process, decision quality, and overall value-added. Their evaluations of D-Side were significantly more favorable than their evaluations of the current approach.     

A new algorithm for modeling online search behavior and studying ranking reliability variations

We design an information retrieval algorithm that mimics the stochastic behavior of decision-makers (DMs) when evaluating the alternatives displayed by an online search engine. The algorithm consists of a decision tree that incorporates all the 1024 decision nodes that may arise from the information retrieval process of DMs. We calibrate the behavior of the algorithm to the one observed from online users and run several sets of 1,000,000 queries. Each query lets DMs decide which subset of the ten alternatives composing the initial page of results to click, allowing us to evaluate their behavior as ranking reliability is assumed to decrease when DMs decide not to click on an alternative. We compare the click-through rates (CTRs) obtained when modifying the degree of ranking reliability derived from the alternatives displayed on the first page of search results. We illustrate how the stability of the CTR prevails among the top-ranked alternatives within relatively reliable scenarios while it drops when imposing large initial decrements in reliability. The resulting consequences regarding the importance of relative ranking positions are analyzed, the top three alternatives exhibiting a generally contained decrease in their CTRs that contrasts with the cumulative pattern arising from the fourth position onwards.

     

A robust optimization approach for imprecise data envelopment analysis

Crisp input and output data are fundamentally indispensable in traditional data envelopment analysis (DEA). However, the input and output data in real-world problems are often imprecise or ambiguous. Some researchers have proposed interval DEA (IDEA) and fuzzy DEA (FDEA) to deal with imprecise and ambiguous data in DEA. Nevertheless, many real-life problems use linguistic data that cannot be used as interval data and a large number of input variables in fuzzy logic could result in a significant number of rules that are needed to specify a dynamic model. In this paper, we propose an adaptation of the standard DEA under conditions of uncertainty. The proposed approach is based on a robust optimization model in which the input and output parameters are constrained to be within an uncertainty set with additional constraints based on the worst case solution with respect to the uncertainty set. Our robust DEA (RDEA) model seeks to maximize efficiency (similar to standard DEA) but under the assumption of a worst case efficiency defied by the uncertainty set and it’s supporting constraint. A Monte-Carlo simulation is used to compute the conformity of the rankings in the RDEA model. The contribution of this paper is fourfold: (1) we consider ambiguous, uncertain and imprecise input and output data in DEA; (2) we address the gap in the imprecise DEA literature for problems not suitable or difficult to model with interval or fuzzy representations; (3) we propose a robust optimization model in which the input and output parameters are constrained to be within an uncertainty set with additional constraints based on the worst case solution with respect to the uncertainty set; and (4) we use Monte-Carlo simulation to specify a range of Gamma in which the rankings of the DMUs occur with high probability.     

An extended compromise ratio method for fuzzy group multi-attribute decision making with SWOT analysis

The technique for order preference by similarity to ideal solution (TOPSIS) is a well-known multi-attribute decision making (MADM) method that is used to identify the most attractive alternative solution among a finite set of alternatives based on the simultaneous minimization of the distance from an ideal solution (IS) and the maximization of the distance from the nadir solution (NS). We propose an alternative compromise ratio method (CRM) using an efficient and powerful distance measure for solving the group MADM problems. In the proposed CRM, similar to TOPSIS, the chosen alternative should be simultaneously as close as possible to the IS and as far away as possible from the NS. The conventional MADM problems require well-defined and precise data; however, the values associated with the parameters in the real-world are often imprecise, vague, uncertain or incomplete. Fuzzy sets provide a powerful tool for dealing with the ambiguous data. We capture the decision makers’ (DMs’) judgments with linguistic variables and represent their importance weights with fuzzy sets. The fuzzy group MADM (FGMADM) method proposed in this study improves the usability of the CRM. We integrate the FGMADM method into a strengths, weaknesses, opportunities and threats (SWOT) analysis framework to show the applicability of the proposed method in a solar panel manufacturing firm in Canada.     

Modeling of bentonite/sepiolite plastic concrete compressive strength using artificial neural network and support vector machine

Plastic concrete is an engineering material, which is commonly used for construction of cut-off walls to prevent water seepage under the dam. This paper aims to explore two machine learning algorithms including artificial neural network (ANN) and support vector machine (SVM) to predict the compressive strength of bentonite/sepiolite plastic concretes. For this purpose, two unique sets of 72 data for compressive strength of bentonite and sepiolite plastic concrete samples (totally 144 data) were prepared by conducting an experimental study. The results confirm the ability of ANN and SVM models in prediction processes. Also, Sensitivity analysis of the best obtained model indicated that cement and silty clay have the maximum and minimum influences on the compressive strength, respectively. In addition, investigation of the effect of measurement error of input variables showed that change in the sand content (amount) and curing time will have the maximum and minimum effects on the output mean absolute percent error (MAPE) of model, respectively. Finally, the influence of different variables on the plastic concrete compressive strength values was evaluated by conducting parametric studies.     

A tuned hybrid intelligent fruit fly optimization algorithm for fuzzy rule generation and classification

Fuzzy rule-based systems (FRBSs) are well-known soft computing methods commonly used to tackle classification problems characterized by uncertainties and imprecisions. We propose a hybrid intelligent fruit fly optimization algorithm (FOA) to generate and classify fuzzy rules and select the best rules in a fuzzy if–then rule system. We combine a FOA and a heuristic algorithm in a hybrid intelligent algorithm. The FOA is used to create, evaluate and update triangular fuzzy rule-based and orthogonal fuzzy rule-based systems. The heuristic algorithm is used to calculate the certainty grade of the rules. The parameters in the proposed hybrid algorithm are tuned using the Taguchi method. An experiment with 27 benchmark datasets and a tenfold cross-validation strategy is designed and carried out to compare the proposed hybrid algorithm with nine different FRBSs. The results show that the hybrid algorithm proposed in this study is significantly more accurate than the nine competing FRBSs.

     

Experimental and macro finite element modeling studies on conformability behavior of woven nylon 66 composite reinforcement

Abstract

In this study, 3D hemispherical forming experiments are conducted to analyze the conformability behavior of nylon 66 plain woven reinforcements with different weft densities. To make insight into the forming process, a macro finite element model is also proposed for 3D forming of a specific woven reinforcement by using the defined ‘Fabric’ material. The results show that the proposed FEM modelling is highly correlated with experimental findings in terms of forming energy values with small and insignificant errors values, which confirm well model validity. It shows that the nylon 66 composite woven reinforcement with a lower weft density exhibits a lower forming energy (toughness) and hence a higher conformability over a hemispherical surface. On the other hand, modeling outputs clearly indicate that more wrinkling intensity appear for woven reinforcement with a higher weft density.     

A chance-constrained portfolio selection model with random-rough variables

Traditional portfolio selection (PS) models are based on the restrictive assumption that the investors have precise information necessary for decision-making. However, the information available in the financial markets is often uncertain. This uncertainty is primarily the result of unquantifiable, incomplete, imprecise, or vague information. The uncertainty associated with the returns in PS problems can be addressed using random-rough (Ra-Ro) variables. We propose a new PS model where the returns are stochastic variables with rough information. More precisely, we formulate a Ra-Ro mathematical programming model where the returns are represented by Ra-Ro variables and the expected future total return maximized against a given fractile probability level. The resulting change-constrained (CC) formulation of the PS optimization problem is a non-linear programming problem. The proposed solution method transforms the CC model in an equivalent deterministic quadratic programming problem using interval parameters based on optimistic and pessimistic trust levels. As an application of the proposed method and to show its flexibility, we consider a probability maximizing version of the PS problem where the goal is to maximize the probability that the total return is higher than a given reference value. Finally, a numerical example is provided to further elucidate how the solution method works.