Semi-automated Tool Recommender for Software Development Processes
Electronic Notes in Theoretical Computer Science (ENTCS)
| Hi-index | 0.00 |
The use of commercial-off-the-shelf (COTS) / government-off-the-shelf (GOTS) applications as components in software systems is increasingly prevalent. The critical step of tool selection for an integrated suite is usually based on identifying the tools that best match the functionality requirements needed. Other factors tangential to technical performance are playing a more important role in the tool selection process and making the mapping of customer needs to technical requirements less obvious. This paper suggests a shift from the traditional "best tools" selection approach, where tools are selected for their performance to a more holistic "end-to-end" approach, where customer concerns, business and cost benefits, and technical performance are weighed concurrently. The end-to-end methodology was applied to an integrated suite for the intelligence analysis process and was compared to a theoretical system employing a best tools approach. This showed that the end-to-end approach resulted in significant software related cost reductions.