Statistical Methods in Rubber Research and Development

As this article has demonstrated, statistical methods have considerable utility in rubber research and development. Such methods are cost effective and perhaps, more important, help the user to formulate questions in such a way as to make research programs more productive. The use of statistical methods in formulation development have been particularly fruitful. For example, without statistical methods, the compounders common task of simultaneously achieving a specific balance of cost, property levels, and processing behavior would be an expensive undertaking at best and hopeless at worst. Statistical methods help the practitioner make sound decisions in the light of often extreme variability in the data. This is especially the case when the data is fatigue, tensile strength, or some other fracture property of rubber. In the light of such broad distributions, hit or miss and one variable at a time strategies are seriously flawed. It is not hard to find a sequential study in which decisions about each trial were driven simply by changes due to random variability which were erroneously attributed to the intentional changes in variable level(s) from the previous trial. Such research studies are very similar to random walks. Many excellent and innovative studies are reported above. However, given the wide utility of statistical methods, these methods appear to be underutilized judging from the unexpectedly small number of papers which were found. Similarly, it is not difficult to find published rubber research which would not have benefitted by the use of statistical experimental design and/or data analysis. Fortunately, however, this appears to be changing. The increasing availability of good affordable software for personal computers makes it possible for good statistics to be done with less dependence on trained statisticians which are not available in many organizations. As this trend continues, the number of studies employing statistical methods will surely grow.