AVL RFQ pre-fills a call for tender form (containing thousands of requirements) with relevant answers from previously submitted tenders thus reducing time and effort of engineers.
In this project we developed a prototype that supports engineers in technical assessment of a product specification. The developed prototype
- analyses a product specification, immediately searches for applicable norms, relevant legislations, and similar specifications using textmining;
- ii) extracts requirements from unstructured text using natural language processing technologies, and
- iii) digests user input during the process (note-taking, marking extracted requirements as feasible etc.) into technical documentation.
As a result the technical analysis process becomes more efficient via the proactive search. The automatic documentation functionality improves quality and availability of documentation. The prototype also ensures that every requirement needs to be explicitly marked and thus increases the explicit accountability for decisions of requirements engineers. The prototype was designed based on a work process analysis carried, which identified searching, reading, analyzing, reporting and quoting as key steps in the work process; and based on co-design with AVL engineers and decision makers. Close interaction with stakeholders at all levels was key for making the prototype innovative, and for achieving end user acceptance and relevant business value. A paper on the project results will be submitted to the International Journal of Knowledge Management.