Measurement of productivity growth during plant startup

Measurement of productivity growth at thirty continuous steel casting plants disclosed universally long startup duration ranging from one to six years. This research finding extends the development of productivity growth measurement, which began with the aircraft learning curve forty years ago, to a significantly sized sample taken from one new machine-intensive technology. The manufacturing progress function, which is the inverse of the learning curve, was the model used for measurement. Log-log graphs of data from these thirty plants, which are located in ten countries, on four continents, provide visual evidence of regular productivity growth during startup. Statistical results from linear regression of its logarithmic transformation show that the manufacturing progress function describes startup very well. This description is improved by three new parameters of the function which have been introduced and defined here. The new parameters are: manufacturing progress, startup duration, and lost capacity. These improved measurements developed in the continuous steel casting technology are generalizable to many other machine-intensive technologies. As a result, there are a number of implications for engineering managers which can lead to better measurement, control, prediction, and acceleration of startup.