Concurrent multiresponse non-linear screening: Robust profiling of webpage performance

Besseris, George J.

doi:10.1016/j.ejor.2014.06.021

Cited by 3 publications

(8 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Once the number of replicates was settled, the response dataset for each characteristic was condensed using ranking operations [14][15][16][17][18]. Both monitored characteristics followed the "smaller-is-better" optimization direction based on Taguchi categorization [13].…”

Section: Discussionmentioning

confidence: 99%

Machinability evaluation and screening of leaded and lead-free brasses using a non-linear robust multifactorial profiler

Toulfatzis

Pantazopoulos²,

Besseris

et al. 2016

Int J Adv Manuf Technol

Self Cite

View full text Add to dashboard Cite

The aim of this work was the evaluation of the machinability of leaded brass namely CuZn39Pb3 (CW614N) in comparison to three lead-free brasses alloys namely CuZn42 (CW510L), CuZn38As (CW511L), and CuZn36 (C27450). The machinability of the studied alloys was investigated, based on chip morphology and power consumption, as quality characteristics. Microstructure examination and hardness testing was employed for the characterization of the selected alloys. Design of experiments (DOE) was employed in a multi-non-parametric study in order to identify the critical-to-machinability parameters and obtain their optimum values for high performance machining. The attempted joint screening using a four-level orthogonal array revealed that the depth of cut and the alloy type were the two statistically predominant effects. The chip morphology and the power consumption in a balanced concurrent optimization effort were optimal when the depth of cut was set at 0.5 mm for the alloy type CW614N chip morphology optimization. The optimal chip morphology response was split in equal chances to produce needle or arc chips. The predicted power consumption was confined in a range of values with an upper boundary at 65 W. The findings of the statistical evaluation were experimentally confirmed, validating the DOE approach. Keywords Machinable brass . Lead-free brasses . Machinability . Design of experiments (DOE) Nomenclature ANOVA Analysis of variance ER Error contribution HV final The chip average hardness (Vickers hardness) HV initial The initial surface hardness (Vickers hardness) MANOVA Multivariate analysis of variance M CM G Grand median value of CM data M P G Grand median value of P data mCM DC o Median value of CM for the optimal setting of DC effect mCM M o Median value of CM for the optimal setting of M effect mP DC o Median value of P for the optimal setting of DC effect mP M o Median value of P for the optimal setting of M effect MR Master response (ranked values of ssRS) OA Orthogonal array R 2 Coefficient of determination rCM i (i = 1,2) Ranked values of the response of the chip Morphology replicates rP i (i = 1,2) Ranked values of the response of the power consumption replicates srCM rSM 1 + rSM 2 srP rP 1 + rP 2 rCM Ranked values of srCM rP Ranked values of srP ssRS rCM 2 + rP 2 WH

show abstract

Section: Discussionmentioning

confidence: 99%

Machinability evaluation and screening of leaded and lead-free brasses using a non-linear robust multifactorial profiler

Toulfatzis

Pantazopoulos²,

Besseris

et al. 2016

Int J Adv Manuf Technol

Self Cite

View full text Add to dashboard Cite

show abstract

“…Hence, no inference is possible at this stage which awards no statistical significance to the contrasts. Ostensibly, the analysis is rendered inconclusive due to the known incompatibility of non-linear unreplicated-saturated OAs with ordinary multi-variable converters [9,30,37]. A practical (subjective) outlet to arrive to an approximate solution is to follow Taguchi's recommendation to pool and disaffect some of the weaker effects.…”

Section: Discussionmentioning

confidence: 99%

“…However, the introduction of GRNN to defuzzify the OA-dataset is not intended to furnish the terminal profiling outcome of the examined effects. This is not feasible for unreplicated-saturated OA designs due to the additional messiness which is ushered inherently by the random recipe-partitioning requirement which is imposed by the intelligent solver [9,37]. Instead, the GRNN is utilized to create "smart" samples.…”

Section: Homogenizing Oa Effective-slope Data With An Intelligent Solvermentioning

confidence: 99%

“…The initial L9(3 4 ) OA-dataset (Appendix A) has been cast into a non-linear saturated-unreplicated form where the output role is replaced by the 'slopes' vector. Variance homogeneity cannot be verified at this stage due to the saturated-unreplicated condition imposed on the OA [8,37]. Thus, the input-output OA-data arrangement is transferred to the GRNN converter for intelligent processing.…”

Section: Concurrent Screening and Optimizationmentioning

confidence: 99%

“…A flexible yet exclusive rubric of multifactorial recipes accommodates the trial patterns for increasing hordes of investigated effects. Tabulating apt recipe guides that interweave the precise combinations of the investigated factor-settings have been standardized in the augmenting families of orthogonal arrays (OAs) [8,37].…”

Section: Taguchi Oa Arrangement For Controlling Factors and Controllamentioning

confidence: 99%

See 2 more Smart Citations

Screening dense and noisy DOX-datasets with NN-blending and “dizzy” swarm intelligence: Profiling a water quality process

Besseris

2017

Swarm and Evolutionary Computation

Self Cite

View full text Add to dashboard Cite

A novel nature-inspired method is presented in this work for resolving product/process development or improvement with design of experiments (DOX). The technique is suitable for difficult Taguchi-type multifactorial screening and optimization studies that need to simultaneously contain the double hassle of controllable and uncontrollable noise intrusions. The three-part sequential processing routine requires: 1) a regressive data-compression preprocessing, 2) a smart-sample generation using general-regression neural networks (GRNN), and 3) a screening power prediction using 'reverse' swarm intelligence (SI). The approach is primed to confront potential non-linearity and data messiness in the examined effects. The Taguchi-type orthogonal-array (OA) sampler is tuned for retrieving information in controllable (outer OA) and uncontrollable (inner OA) noises. The OA-saturation condition is elicited for maximum data exploitation. GRNN-fuzzification consolidates into a single contribution the uncertainty from all possible sources. The resulting 'smart' sample is defuzzified by a robust-and-agile data reduction. Screeningsolution meta-power is controlled with a new SI-variant. The independent swarm groups, as many as the studied effects, are tracked toward preassigned targets, i.e. their ability to return to their host beehives. The technique is illustrated on a complex purification process where published multifactorial data had been collected for a critical wastewater paradigm and thus may be used to test the benchmark solution. However, environmental water-qualimetrics are profoundly dominated by messy data as justified in this work. We elucidate on several issues that regular Taguchi methods may be benefited by the proposed GRNN/SI processing while emphasizing the consequence of overlooking the underlying assumptions that govern standard comparison models. The new swarm itelligence method offered a practical way to estimate a first-time "soft" power measure for the inner/outer OA optimization case that was impossible with ordinary statistical multi-factorrial treatments. Key performance advantages in efficiency, robustness and convenience are highlighted against alternative approaches.

show abstract

Process parameter optimization for lifetime improvement experiments considering warranty and customer satisfaction

Shao

et al. 2022

Reliability Engineering & System Safety

View full text Add to dashboard Cite

Concurrent multiresponse non-linear screening: Robust profiling of webpage performance

Cited by 3 publications

References 39 publications

Machinability evaluation and screening of leaded and lead-free brasses using a non-linear robust multifactorial profiler

Machinability evaluation and screening of leaded and lead-free brasses using a non-linear robust multifactorial profiler

Screening dense and noisy DOX-datasets with NN-blending and “dizzy” swarm intelligence: Profiling a water quality process

Process parameter optimization for lifetime improvement experiments considering warranty and customer satisfaction

Contact Info

Product

Resources

About