INTECoM : an integrated conceptual data modelling framework : a thesis presented in partial fulfilment of the requirements for the degree of Doctor of Philosophy in Information Systems at Massey University, Palmerston North, New Zealand

Abstract

Conceptual data models, a fundamental component of information systems development, traditionally play two essential roles, as communication tools and database design blueprints. However, despite their importance to the success of information systems, and a considerable amount of research effort, no definitive method for constructing them has yet been described. Entity-Relationship (E-R) Modelling, accepted as a de facto standard for a number of years, has been increasingly criticised. A number of alternatives have been proposed and some, such as Object-Oriented modelling, have gradually been accepted by the practitioner community. Nevertheless, effective conceptual data modelling continues to be recognised as a difficult activity, both to teach and to practice. This study investigates conceptual data modelling in the context of the relational database development process. However, rather than specifying a new method or exploring the efficacy of existing ones, it focuses on the nature of the activity itself. The construction of a conceptual data model encompasses both the analysis and design stages of systems development. Some fundamental differences in modeller behaviour, required by these activities, are explored. The disparate purposes of a conceptual data model are also investigated and the effectiveness of using the same modelling method in both stages, and for both purposes, is questioned. To explore the suitability of current methods to specific development activities, the procedures inherent in the use of two conceptual data modelling approaches, E-R Modelling and the Natural Language Information Analysis Method (NIAM) are also investigated. The result of this exploration is the recognition that the methods have exclusive strengths and that it is more productive to view them as complementary rather than competing. Consequently, a database design framework, INTECoM, is constructed in which the two methods are integrated and matched to the activities for which they are most suited. The framework is supplemented by a new technique, NaLER, to facilitate communication in the design stage. The soundness and viability of this theoretical framework is examined through its use on a small development and the implications of the adoption of INTECoM on both education and practice are considered.