An Empirical Evaluation of a Reuse Based Approach for ERP Customization

One concern in ERP customization is the matching of ERP features with the organizations' requirements. An approach to achieve this more efficiently than with the traditional similarity measures, consists in reusing generic domain knowledge. However, except for a case study, this claim has never been substantiated before. This paper reports the first of a series of controlled experiments aimed at evaluating the efficiency of reusing generic domain knowledge in the matching activity. The experiment showed that starting at the generic domain level is always more efficient than starting directly at the specific level, and this whether the domain knowledge is provided or constructed on the fly. Two questions remain however: (a) which is the most efficient formalism to specify generic domain knowledge? (b) Is the lesson learned valid for any domain, or is it only specific to some domains such as charity associations considered in this experiment?