Implementing metrics for process improvement

There is increasing interest in the use of metrics to control the software development process, to demonstrate productivity and value, and to identify areas for process improvement. Research work completed to date is based on the implementation of metrics in a 'standard' software development environment, and follows either a top-down or bottom-up approach. With the advent of further European unity, many companies are producing localised products, ie products which are translated and adapted to suit each European country. Metrics systems need to be customised to the processes and environment of each company. This thesis describes a 12-step process for metrics implementation, using an optimum approach which is a combination of top-down and bottom-up approaches, with a set of applicable metrics, covering the software development process, which can be adapted for any development environment. For the case study, a software localisation company, the suggested implementation process is followed, and relevant measures are adapted to suit the different environment, with a particular emphasis on quality metrics. This thesis also demonstrates that a metrics system is itself subject to continuous improvement, and rather than being a once-off implementation, it is an evolutionary process, changing as the software development process comes under control