Organizing principles for the design of classification and coding software

Abstract

Abstract This paper describes several organizing principles useful in designing classification and coding (C&C) software in a relational database management system (DBMS) environment. Specifically, these software design principles include the usage of Operational Constructors, Meta Models, Dynamic Binding, and Associative Naming to aid the design of full-purpose classification and coding software that can take as input a classification model of part families. These principles aid the software development effort in the following ways. Operational Constructors are universal programming concepts that allow the programmer to take advantage of the graphical user interface (GUI) features of the DBMS, such as list boxes, scroll boxes, subforms, and the like. Meta models represent the structure of the classification model. By implementing any classification model as a meta model, it can easily be instantiated into the software without the need for custom development. Dynamic Binding provides a general mechanism to query the database for part information by dynamically stitching user selections to a general structured query language (SQL) query. Finally, Associative Naming allows the user to retrieve data in a general manner through an innovative table-naming scheme and traversal process. A case study is presented showing application of these principles in support of a C&C system for a purchasing application used for procurement of raw materials, electronic components, and consumable supplies.