QRB/501
Standard deviation is an important tool that is used today by almost every corporation. That corporation can be NYSE, stores such as Wal-Mart, investment firms even software developing companies. One would wonder how a software company uses deviation to be successful. Software companies uses the deviation value of the (D-Value) to measure for software data involved during software date development. The D-Value gives an alternative to software metrics based upon per number of lines of code; example error rate (number of error rates per thousand lines), and documentation rate (number of pages of module design documentation per thousands of lines of code). Using D-Value the data software modules are much more fairly evaluated than these conventional metrics. Today software companies use D-Value rather than metric value because D-Value is used to find items which actually relate to software quality, and to find the quality of each value measured by D-value becomes better as the documentation rate D-value increases. The purpose of the study in this article was to see how would the program run in both D-Value and metric value under a large scale business system with documentation items with standardized forms, under medium scale electronic switching system with documentation items with not standardize forms, large scale simulation software under language FORTRAN, and large scale banking software. In the results they have concluded that their assumption about using D-Value is closely relatable to the results found in the study. In the study they have also found using D-Value have also increased the software quality and was also closely related to software error rate.
Reference
Miyazaki, Yukio, and Noritoshi Murakami. Proceedings. Tokyo: Software Development Planning Group, 1987. Pages 83-91.