A study of translating customer statements to need statements in new product development

Abstract

Automated translation of need statements would be very instrumental in the efficient and accurate extraction of customer needs. With the focus for product development shifting to customer active paradigm (CAP) rather than the manufacturing active paradigm, this would in turn allow for more accurate specification to be accurately defined in product development. The project proposes to facilitate the translation by creating a framework, through the use of natural language processing and available rules such as the distinction of product attributes and the need attributes to provide for a highly automated translation of a customer need statement. The results are then analyzed to determine if the customer’s statements sentiments are positively skewed or negatively skewed using sentiment analysis. The final translated statements are then compared with manual translating methods based on current available rules and relevant guidelines. Results show that the automated translation method facilitates fast and systematic translation of large amounts of raw customer data. However, the current automated translation method is shown to be less accurate than a human based manual translation method. The methodology provides some ground work to allow for further development in customer opinion mining that allows for further processing to provide for more comprehensive need statements. This paves the way for trend projections and specification analysis. Due to short development cycles and rapid transition from cradle to grave for product life cycles, automatic translation of need statements will allow developers to link the shortcomings of their products much faster than previously possible.