Paint & Coatings Industry - March 2008 - (Page 58)

deSigns of the Times: Slacking Off! Table 2 | A slack, factorial design for three ingredients Experiment 1 2 3 4 5 6 7 8 9 Factorial Design Standard Order A –1 +1 –1 +1 –1.4 +1.4 0 0 0 B –1 –1 +1 +1 0 0 –1.4 +1.4 0 Factorial Design With Levels A 0.30 0.40 0.30 0.40 0.35 0.35 0.30 0.40 0.35 B 0.30 0.30 0.40 0.40 0.30 0.40 0.35 0.35 0.35 C 0.40 0.30 0.30 0.20 0.35 0.25 0.35 0.25 0.30 Property 60 50 70 140 43 93 63 93 65 Figure 1 | The slack, factorial design space. Figure 2 | The slack, factorial response surface. Visit ads.pcimag.com PCI03084Ariz.indd 1 GRINDING MEDIA FOR ALL MILLING APPLICATIONS Any density - any diameter - any composition Zirconium Oxide, Glass, Steel, Steel Shot, Tungsten Carbide, Alumina, Silicon Carbide, Silicon Nitride, Ottawa Sand, Agate, Titanium Oxide, Flint Pebbles, Rare Earth ZrO/CeO, Nickel, Plastics, etc. Precision Glass Balls for ball bearings, check valves, level gauges and low temperature critical applications are also available Small orders welcome. www.glenmills.com 2/7/08 11:03:47 AM slack, factorial design. Hopefully the formulation with C as low as 0.2 will still work. If not, no data will be obtained for Experiments 4, 6 and 8, and no response surface analysis would be possible. The star points normally lie outside the design range at –1.4 and +1.4 for each ingredient. However, the star point levels were adjusted to –1 and +1 since the design was not to be greater than the range of 0.3 to 0.4. The design space and design points are shown in Figure 1. Assuming the researcher had data for all of the experiments, when the quadratic regression equation is analyzed for the factorial design using the experiments of Table 2 the equation would have the following form. Property = b0 + bAA + bBB + bABAB + bAAA2 + bBBB2 (1) Large stocks - Competitive Prices - Attentive service. Please phone for full information and prices. Glen Mills Inc. 220 Delawanna Ave., Clifton, NJ 07014 Tel: (973) 777-0777 Fax: (973) 777-0070 Visit ads.pcimag.com 58 MARCH 2008 | W W W . P C I M A G . C O M If all the coefficients were significant, one would interpret them by saying, for example, that when A is varied the Property is changed by aAA units and aAAA2 units. Similarly, when B is varied the Property is changed by a BB units and a BBB2 units. In addition of course, when either A or B is varied, one would also see a change in the contributions of the interaction term AB. What is seemingly lacking is what contribution C makes, since it is also being varied. An inexperienced researcher might ignore C’s contribution or assume it is non-existent. The effect of C is buried within the b coefficients, but more of this later. The response surface that is generated from the slack, factorial design data of Table 2 is given in Figure 2. The graph http://www.azic.com http://www.azic.com http://www.azic.com http://ads.pcimag.com http://www.glenmills.com http://www.glenmills.com http://ads.pcimag.com http://WWW.PCIMAG.COM

Table of Contents Feed for the Digital Edition of Paint & Coatings Industry - March 2008

Viewpoint Industry News Calendar of Events Company News Mergers and Acquisitions Names in the News Marketplace Quarterly Stock Watch Reaching Strict VOC Requirements With Outstanding Durability for Industrial Maintenance and Marine Coatings A Unique Fungicide for Wood-Filled Plastic deSigns of the Times: Slacking Off! From Rocks to Nanos Shear-Free Dispersion Process Fluorinated Oxetane Oligomers as Versatile Intermediates for Polymer Modification The Effects of TPO Composition on Adhesion and Proposed Chemical Mechanism Supplier Showcase Color Management Materials Watch Products Classifieds Advertiser Index

Paint & Coatings Industry - March 2008

For optimal viewing of this digital publication, please enable JavaScript and then refresh the page. If you would like to try to load the digital publication without using Flash Player detection, please click here.