Solutions Manual STATISTICS FOR ENGINEERS AND S... LINK

Although the engineers of industry have access to process data, they seldom useadvanced statistical tools to solve process control problems. Why this reluctance? Ibelieve that the reason is in the history of the development of statistical tools, whichwere developed in the era of rigorous mathematical modelling, manual computationand small data sets. This created sophisticated tools. The engineers do not understandthe requirements of these algorithms related, for example, to pre-processing of data. Ifalgorithms are fed with unsuitable data, or parameterized poorly, they produceunreliable results, which may lead an engineer to turn down statistical analysis ingeneral.

Solutions Manual STATISTICS FOR ENGINEERS AND S...

This thesis looks for algorithms that probably do not impress the champions ofstatistics, but serve process engineers. This thesis advocates three properties in analgorithm: supervised operation, robustness and understandability. Supervisedoperation allows and requires the user to explicate the goal of the analysis, whichallows the algorithm to discover results that are relevant to the user. Robust algorithmsallow engineers to analyse raw process data collected from the automation system ofthe plant. The third aspect is understandability: the user must understand how toparameterize the model, what is the principle of the algorithm, and know how tointerpret the results.

The above methodology is illustrated by an analysis of an industrial case: theconcentrator of the Hitura mine. This case illustrates how to define the problem withoff-line laboratory data, and how to search the on-line data for solutions. A majoradvantage of algorithmic study of data is efficiency: the manual approach reported inthe early took approximately six man months; the automated approach of this thesiscreated comparable results in few weeks. 041b061a72