Papers by Keyword: Factorial Design

Abstract: Recent findings on the production of quantum dots from various carbon sources shed light on their advantages such as sustainability, low toxicity and cost, and one-step synthesis over their heavy-metal counterpart. This paper focused on developing and analyzing the production of carbon quantum dots from glycerol via hydrothermal carbonization and conjugated with Tetraethylenepentamine (TEPA). A 2³ full factorial experimental design was applied considering factors: the compositional ratio of TEPA (A), time of exposure (B), and temperature of reaction (C). Statistical analyses revealed experimental factors A and B; and interactions of AB and AC had statistically significant effects on the response variable, quantum yield (QY). Factor C as the main effect was not significant but was included in the statistical model to maintain hierarchy and integrity. Coded and actual statistical models were presented here.

Abstract: The usage of groundwater as drinking water source in many parts of Kelantan encourages the research and development of various cost-effective alternative adsorbent material for turbidity reduction and drinking water purification. The preparation, characterization, and use of a magnetic biocarbon adsorbent composite (MBAC) is introduced in this study as an option to treat turbid groundwater. In contrast to commercial activated carbon (CAC), peak shifts and peaks denoting Fe-O bending were observed in the FTIR spectrum of MBAC. The adsorption process for turbidity reduction by MBAC and CAC was investigated. A factorial design matrix consisting of four parameters were tabulated, namely, adsorbent dosage (0.02, 0.04, and 0.06 g), agitation time (15, 30, and 60 min), agitation rate (150, 200, and 250 rpm), and two adsorbent particle size ranges (M: 300 < x ≤ 500, and Q: ≤ 45 μm). The predictive model was validated with 0.04 g MBAC of ≤ 45 μm in particle size, agitated at 150 rpm, for 48 min, that attained 98.46% turbidity removal efficiency with a final NTU reading of 0.40. Conversely, CAC removed 88.19% for a final NTU reading of 3.07. Overall, the iron oxide impregnated biocarbon composite showed better turbidity reduction capability compared to CAC. The findings of this work support the potential application of MBAC as an alternative adsorbent for the treatment of groundwater sourced drinking water.

Abstract: The awareness on sustainability of the environment among the researchers leads to the exploration of natural fiber composite materials. Hybridization of synthetic fiber and natural fiber is one of the potential strategies to enhance the mechanical properties as well as the degradability of such composite materials. However, less information concerning the optimization of tribological properties of this hybrid composite is available in literature. The aim of this study is to propose a statistical model to predict and optimize wear and coefficient of friction of kenaf/carbon reinforced epoxy composite. The value of parameters; load and sliding velocity ranges from 10 to 30 N and 20.9 to 52.3 m/s, respectively, are used to assess wear and coefficient of friction (COF) of different stacking sequences using the Analysis of Variance (ANOVA). The tribological test was conducted using a pin-on-disc tribometer. Multifactorial design analysis was employed to optimize the test control variables. It was found that, the optimized factors that affects the coefficient of friction and wear is at load 30 N and sliding velocity of 52.36 m/s. The proposed statistical models for wear and COF have 99.5% and 97.6% reliability, respectively. The generated equation models are bounded within the wear test control factors and ranges. The outcome from this study will be very useful for main parameter prediction for an optimized wear and COF.

Abstract: This paper aims to discuss the influence of mix composition of cement mortar on fresh and rheological properties of cement mortar. Two different natural fibres, basalt (BA) and sisal (SL) are selected based on fresh and rheological behaviour for its usability in a cementitious mixture. The workability and rheological behaviour are evaluated by flow table test, cone penetrometer and slump test of the mixture. A full factorial design method was used to investigate the effects of four mix components: dosage of cement content (B), percentage of fly-ash (FA) by mass of cement, dosage of basalt fibre (BA) and dosage of superplasticizer (SP) along with a water/binder ratio of 0.41. A mathematical model which predicts the main effect and interactions of these components for each of the measured properties are derived using the factorial design. The proposed mixtures consist of two levels of binder content as 550 kg/m3 and 650 kg/m3, FA as 5% and 20% by mass of cement, BA as 1 kg/m3 to 3 kg/m3 and SP as 2 kg/m3 to 4 kg/m3. By reducing the number of test batches needed, the mathematical models produced with this method can expedite optimizing the mixture proportions of cement mortar to achieve desired fresh and rheological properties.

Abstract: Reproducibility in respect to welded structures realization is one of the main requirements for a wide variety of industrial applications. One of the international tendencies regarding the use of the steel is the replacing, in critical areas, of structural steels with high performance steel, e.g. with HSLA steels. The paper presents the results of a factorial designed experimental program focused on determining mathematical correlations between the GMAW process parameters for T joints of 4mm thick steel plates of structural (S235JR+AR according to SR EN 10025-2) and hot-rolled, high-strength low-alloy (HSLA) steel plates (S420MC according to EN 10025-4), respectively. A comparison between the obtained mathematical correlations that connect the welding parameters and the main mechanical characteristics is presented. The correlations can be used for applying the optimal combination of welding process parameters for realizing the T-joints of welded products.

Abstract: In all industrial fields, the product requirements are more and more demanding. HSLA steels are designed to provide higher atmospheric corrosion resistance and improved mechanical properties than structural steels. The paper presents the results of an experimental program based on factorial design, applied to predict the mechanical properties of butt-welded joints of S420MC and S460MC hot-rolled, high-strength low-alloy (HSLA) steel plates with 2mm, 4mm and 8mm thickness. Gas Metal Arc Welding (GMAW) was used and correlations between the main process parameters and the related mechanical properties of the welded joints were found. Obtained mathematical correlations can be exploited to provide optimal combination of welding parameters to fit the quality requirements of the end-users for envisaged welded product.

Abstract: This research is part of the mortar’s plastic shrinkage study. Contents in this article is related to how the Expansive Additive (EX), Shrinkage Reducing Agent (SRA), and Fly Ash (FA) help to reduce and control the shrinkage and to compare the effectiveness of these substances used in the normal curing, i.e., at 30 °C and in the steam curing process by using the factorial design with 3 factors and to be divided into 2 levels. Factors to be studied are the amount of EX, SRA and FA replacement. The test of plastic shrinkage was conducted in accordance with the ASTM C1579-06 standard, placing the strain gauge 0.5 centimeters beneath the surface in the middle of the mold, recording the shrinkage rate starting from the initial setting time for 24 hours. The results showed that, in normal curing, the EX influences the expansion while, in steam curing, the EX and SRA significantly influences the expansion. To add the FA in high volume along with the EX significantly effects the expansion for both the normal and steam curing. Furthermore, the study model and equation for plastic shrinkage of mortar are presented in the form of factor proportion to be considered from the factorial design study basis.

Abstract: Optimization problems often involve a large number of design variables, and the exact influence of each of these variables upon the objective function can become rather complex; there may exist local optima for the objective function, but for the typical heat-integrated distillation sequence, the matter of interest is solely the global optimum. Therefore, it is necessary to create a stochastic algorithm method which can synthesize distillation systems with multiple components. The encoding process employs and integer number series which allows the system flow sheet structure to be portrayed and then managed. Within this portrayal, the broad synthesis problem takes the form of an implicit MILP (mixed-integer linear programming) problem. This study considers the attributes of six well-known optimization algorithms: Harmony Search algorithm (HS), Artificial Bee Colony (ABC), Bat Algorithm (BA), Crow Search Optimization (CSO), Grew Wolf Optimization (GWO) and Monarch Butterfly Optimization (MBO). The optimal variables which influence the harmony search algorithm can be determined through full factorial design analysis. These variables can then be employed in the search to discover the optimal heat-integrated distillation sequence. The study investigates the attributes of the optimal configuration solution, in terms of harmony size (HS), required number of iterations, harmony memory considering the rate (HMCR), and pitch adjustment rate (PAR). The study then adopts the HS algorithm which is duly improved in order to address the problem. In comparison with alternative techniques, HS is more effective and more robust than other approaches.

Abstract: Electrospun shellac nanofibers might be potentially used for wound dressing application due to its natural origin and excellent protective properties. In this study, a full factorial design with three replicated center points was performed in order to investigate the main and interaction effects of shellac content (35-40% w/w), applied voltage (9-27 kV) and flow rate (0.4-1.2 mL/hr) on the morphology of shellac nanofibers. A total of 11 experiments were conducted. The response variables were the diameter of nanofibers, the distribution of diameter and the amount of beads. The results showed that the concentration of shellac was the most significant impact on shellac nanofiber diameter, while applied voltage, interaction between shellac content and voltage, and feed rate were minor factors, respectively. Shellac content and applied voltage had negative relationships with bead amount. When reducing the concentration of shellac and voltage, the amount of beads was increased. However, the influence of these parameters on diameter distribution seemed to be not significant. Based on response surface plot, nanofibers with thinner diameter (~493 nm) and less number of beads (~0.47) could be obtained at the optimum conditions; the shellac content of 38.5% w/w, the voltage of 21 kV and the feed rate of 0.4 mL/hr.

Abstract: Biomaterial components (PMMA/HA/Sericin) are composite biomaterial that have complemented each other, which HA is able to improve the mechanical properties, biocompatible and osteoconductivity. PMMA has the ability biocompatible with human tissue and easy manipulated, while sericin can improve cell viability. Analysis of the composite composition is important to do more in depth, but the design method in the analysis of research also has significant impact in generating research output and prediction output. Analysis of the design method to determine the method of designing the optimal research was needed. This research used 3 factors: Poly methyl methacrylate (PMMA) powder (with concentration of 40% to 60%), Hydroxyapatite (HA) powder (with concentration of 10% to 25%) and Sericin (with concentration of 0.16% to 0.32%). This research is done to identify the effect of individual components of the composite and analyze the design method to obtain the best of diametral tensile strength (DTS) with a small error. The result of two methods was found that the 2^k factorial design method give higher DTS (PHS401032 = 25.49MPa) than that of Taguchi method (PHS601532 = 24.82MPa). The error of 2^k Factorial design method between validation and model prediction is 0.27%, which smaller than Taguchi method (7.45%).