Exponentially Weighted Moving Average (EWMA) Prediction in the Software Development Process

For some years, Statistical Process Controls (SPC) techniques such as traditional Shewhart control charts add value to monitor and to control the Software Development Process (SDP) efficiently. Nonetheless, the application of Shewhart control charts to the SDP involves a considerable problem, since the availability of a sufficiently large set of observations is essential when constructing traditional control charts. Especially at the start-up of each SDP phase such a set cannot be provided. To remedy this problem, Q control charts widely used when monitoring short-run manufacturing processes have been introduced successfully. This paper focuses on the predictive property of Exponentially Weighted Moving Average (EWMA) Q control charts and investigates whether the predictive property is attractive for monitoring and controlling the SDP. Results of initial experiments are also reported.