TABLE OF CONTENTS Statistical methods enable the industrial engineer to make better decisions in the context of the variability inherent to engineering processes. This chapter introduces fundamental ideas of statistical thinking: quantifying and explaining variability in sampled data and appropriately accommodating this variability when drawing conclusions. The key statistical concepts presented are obtaining and graphically displaying sampled process data, selecting an appropriate probability model for the data, and using the model to draw conclusions of interest. The chapter then discusses and illustrates three broad classes of statistical models particularly useful to industrial engineers, concluding with an overview of additional relevant techniques.
Often, engineers believe that they can precisely predict (or control) a process if they know (or can control) the variables entering that process. This belief produces the familiar mathematical model y = f( x), where f( ) is a known function relating a vector x of input variables to y, a process response of interest. Real-world processes, however, vary for unknown or uncontrollable reasons. This extra variation is difficult to account for in the short term, but usually exhibits enough regularity in the long run to be estimated with some confidence. Statistical...
