Effort assessment in the development of information systems projects

Sažetak

There is a great lack of methods and techniques in the software development processitself, as well as the lack of the appropriate tools that would make it more efficient. Thesignificance of the problem is repeatedly emphasized by the need to ensure a high quality ofsoftware and software−based systems. The main objective of this work is to develop andsystematize the original formal procedure for assessing the development of informationsystems in the early stages of the software life cycle, through metrics of the data model. Wecalculate the metrics of data model by using data that can be read off from a base datamodel, which is represented with an Entity−Relationship (ER) diagram that is defined withfour basic concepts: entities, relationships, attributes of entities or relationships andvalues. The idea is to present the complexity of the process with a function of a number ofthese concepts and a number of attributes for entity types. Assessment techniques representthe basis for planning and successful performance of software projects. Statistical methodwas used in this paper and these assessment processes go under the category of empiricalparametric methods, although they have some characteristics of the expert estimationmethod. A developed assessment process represents a step in the efforts to reach suitablemeasures which we would use to assess the size and complexity of the data model and alsoto estimate the amount of costs and resources necessary for the development of informationsystems. Likewise, certain metrics are developed. By being familiar with the data model, wecan use these metrics to quantify characteristics of an information system as a whole in thelogic design phase. Suggested metrics were tested on specific models and the results areshown here.