Limits...
A distribution-free multi-factorial profiler for harvesting information from high-density screenings.

Besseris GJ - PLoS ONE (2013)

Bottom Line: Partial effects are sliced off systematically from the investigated response to form individual contrasts using simple robust measures.Main benefits of the method are: 1) easy to grasp, 2) well-explained test-power properties, 3) distribution-free, 4) sparsity-free, 5) calibration-free, 6) simulation-free, 7) easy to implement, and 8) expanded usability to any type and size of multi-factorial screening designs.The method is elucidated with a benchmarked profiling effort for a water filtration process.

View Article: PubMed Central - PubMed

Affiliation: Department of Mechanical Engineering, Advanced Industrial & Manufacturing Systems Program, Technological Educational Institute of Piraeus, Aegaleo, Greece. besseris@teipir.gr

ABSTRACT
Data screening is an indispensable phase in initiating the scientific discovery process. Fractional factorial designs offer quick and economical options for engineering highly-dense structured datasets. Maximum information content is harvested when a selected fractional factorial scheme is driven to saturation while data gathering is suppressed to no replication. A novel multi-factorial profiler is presented that allows screening of saturated-unreplicated designs by decomposing the examined response to its constituent contributions. Partial effects are sliced off systematically from the investigated response to form individual contrasts using simple robust measures. By isolating each time the disturbance attributed solely to a single controlling factor, the Wilcoxon-Mann-Whitney rank stochastics are employed to assign significance. We demonstrate that the proposed profiler possesses its own self-checking mechanism for detecting a potential influence due to fluctuations attributed to the remaining unexplainable error. Main benefits of the method are: 1) easy to grasp, 2) well-explained test-power properties, 3) distribution-free, 4) sparsity-free, 5) calibration-free, 6) simulation-free, 7) easy to implement, and 8) expanded usability to any type and size of multi-factorial screening designs. The method is elucidated with a benchmarked profiling effort for a water filtration process.

Show MeSH

Related in: MedlinePlus

The quantized plateau for increasing trial-run capacity in  schemes: Graphing the experimental recipe count versus the number of examined effects.
© Copyright Policy
Related In: Results  -  Collection


getmorefigures.php?uid=PMC3756950&req=5

pone-0073275-g001: The quantized plateau for increasing trial-run capacity in schemes: Graphing the experimental recipe count versus the number of examined effects.

Mentions: A few inherent properties of the arrays that reveal the idiosyncrasies of such schemes are now briefly addressed. Firstly, the uptake of controlling factors in a given plan is quantized such that the required recipes are populated in multiples of four [12]. The quantization rule may be expressed as k = n−1 with n = 4· i (i = 1, 2, 3…). This immediately implies that for a given , accommodation of a group of any m investigated effects in the experimental scheme is enabled subject to the restriction: k−3 ≤m≤k. This is further interpreted to mean that the number of trials reaches a plateau until the array progressively fills up with assigned factors up to the maximum capacity of k variables (Fig. 1). If m = k then the array is defined to have achieved a ‘saturated’ state [11]. While driven to saturation, the experimental design is primed now to allow extracting maximum information for the effort expanded. This is because a saturated engages all its available array columns by assigning them in full to the respective investigated factors. As a result, experts scrutinizing different strategies to engineer optimal experimental plans have recognized long ago that ‘saturation’ may be a desirable condition for maximizing the performance of information harvesting with minimum cost [19]–[23]. Likewise, it was discovered that additional gains in cost and time could be anticipated by potentially eliminating data replication at all [24]–[30]. Unreplicated trials simply constitute a single execution of a specified set of recipes as dictated by a given plan. Clearly, the conditions of conducting either unreplicated or saturated trials are distinctly different. However, the specialized methods which have been devised for handling either of the two aforementioned situations have been misconstrued to be employed interchangeably [31]–[32]. We clarify at this point that it is trivial to manipulate a saturated dataset which has been replicated using mainstream statistical tools as much as it is possible to treat straightforwardly an unreplicated dataset that is not saturated. For either case, the researcher has several options to treat the experimental data starting with the standard multi-factorial comparison methods, such as the Analysis of Variance (ANOVA) along with the multi-parameter regression-type solvers abiding to the classical General Linear Model (GLM).


A distribution-free multi-factorial profiler for harvesting information from high-density screenings.

Besseris GJ - PLoS ONE (2013)

The quantized plateau for increasing trial-run capacity in  schemes: Graphing the experimental recipe count versus the number of examined effects.
© Copyright Policy
Related In: Results  -  Collection

Show All Figures
getmorefigures.php?uid=PMC3756950&req=5

pone-0073275-g001: The quantized plateau for increasing trial-run capacity in schemes: Graphing the experimental recipe count versus the number of examined effects.
Mentions: A few inherent properties of the arrays that reveal the idiosyncrasies of such schemes are now briefly addressed. Firstly, the uptake of controlling factors in a given plan is quantized such that the required recipes are populated in multiples of four [12]. The quantization rule may be expressed as k = n−1 with n = 4· i (i = 1, 2, 3…). This immediately implies that for a given , accommodation of a group of any m investigated effects in the experimental scheme is enabled subject to the restriction: k−3 ≤m≤k. This is further interpreted to mean that the number of trials reaches a plateau until the array progressively fills up with assigned factors up to the maximum capacity of k variables (Fig. 1). If m = k then the array is defined to have achieved a ‘saturated’ state [11]. While driven to saturation, the experimental design is primed now to allow extracting maximum information for the effort expanded. This is because a saturated engages all its available array columns by assigning them in full to the respective investigated factors. As a result, experts scrutinizing different strategies to engineer optimal experimental plans have recognized long ago that ‘saturation’ may be a desirable condition for maximizing the performance of information harvesting with minimum cost [19]–[23]. Likewise, it was discovered that additional gains in cost and time could be anticipated by potentially eliminating data replication at all [24]–[30]. Unreplicated trials simply constitute a single execution of a specified set of recipes as dictated by a given plan. Clearly, the conditions of conducting either unreplicated or saturated trials are distinctly different. However, the specialized methods which have been devised for handling either of the two aforementioned situations have been misconstrued to be employed interchangeably [31]–[32]. We clarify at this point that it is trivial to manipulate a saturated dataset which has been replicated using mainstream statistical tools as much as it is possible to treat straightforwardly an unreplicated dataset that is not saturated. For either case, the researcher has several options to treat the experimental data starting with the standard multi-factorial comparison methods, such as the Analysis of Variance (ANOVA) along with the multi-parameter regression-type solvers abiding to the classical General Linear Model (GLM).

Bottom Line: Partial effects are sliced off systematically from the investigated response to form individual contrasts using simple robust measures.Main benefits of the method are: 1) easy to grasp, 2) well-explained test-power properties, 3) distribution-free, 4) sparsity-free, 5) calibration-free, 6) simulation-free, 7) easy to implement, and 8) expanded usability to any type and size of multi-factorial screening designs.The method is elucidated with a benchmarked profiling effort for a water filtration process.

View Article: PubMed Central - PubMed

Affiliation: Department of Mechanical Engineering, Advanced Industrial & Manufacturing Systems Program, Technological Educational Institute of Piraeus, Aegaleo, Greece. besseris@teipir.gr

ABSTRACT
Data screening is an indispensable phase in initiating the scientific discovery process. Fractional factorial designs offer quick and economical options for engineering highly-dense structured datasets. Maximum information content is harvested when a selected fractional factorial scheme is driven to saturation while data gathering is suppressed to no replication. A novel multi-factorial profiler is presented that allows screening of saturated-unreplicated designs by decomposing the examined response to its constituent contributions. Partial effects are sliced off systematically from the investigated response to form individual contrasts using simple robust measures. By isolating each time the disturbance attributed solely to a single controlling factor, the Wilcoxon-Mann-Whitney rank stochastics are employed to assign significance. We demonstrate that the proposed profiler possesses its own self-checking mechanism for detecting a potential influence due to fluctuations attributed to the remaining unexplainable error. Main benefits of the method are: 1) easy to grasp, 2) well-explained test-power properties, 3) distribution-free, 4) sparsity-free, 5) calibration-free, 6) simulation-free, 7) easy to implement, and 8) expanded usability to any type and size of multi-factorial screening designs. The method is elucidated with a benchmarked profiling effort for a water filtration process.

Show MeSH
Related in: MedlinePlus