Evaluation of Component-Based Development Methods

Component-based development (CBD) has received a lot of attention in software engineering literature over the last few years. Awareness has been raised that CBD is the way to go in software development, especially in the domain of e-business where the benefits of reusing components, i.e., faster time-to-market and quality, are essential. The question now is how to realize the full potential of CBD? Did we achieve reuse yet? In order to answer these questions, we evaluate and compare five popular methods for component-based development, including Catalysis, the Rational Unified Process, and Select Perspective, on their maturity and fitness-for-use in the context of e-business engineering. The evaluation is done based on our own reference framework for e-business development and a list of objective criteria. The methods each emphasize certain aspects of CBD, but as yet none of them offers a complete solution.

Participatory Development of Enterprise Process Models

In this chapter we present practical experience from using a technique we call Modeling Conferences, a method for participatory construction and development of enterprise process models. Process models are an important way to support communication, coordination, and knowledge development within an organization. The Modeling Conference method focuses on broad participation from all actors in the organization, is grounded in a social constructivist perspective, and has its theoretical basis in the method of search conferences and process modeling. In an engineering consultancy firm, the Modeling Conference method has been used to develop process models for main common work tasks that have been implemented on an intranet. Independent evaluations show that participation through the Modeling Conferences led to significantly more ownership to the process models, and that the actors have developed new collective knowledge.

Two Meta-Model for Object-Role Modeling

For conceptual information analysis, the object-role modeling (ORM) approach is arguably more suitable than entity-relationship modeling and the class modeling technique within the Unified Modeling Language. Although ORM has been used for three decades and now has industrial modeling tool support, it has no official, standard meta-model. Because of its extensive capability for expressing business rules, ORM is currently being considered as a possible standard for business rules expression within the Object Management Group (OMG), and for use in ontology standards. To support these initiatives and the interchange of ORM model data between different software tools, this chapter discusses recent research by the authors to pave the way for a standard ORM meta-model that employs a standard interchange format. Two different ways of meta-modeling ORM features are presented, design trade-offs are examined, and extensions to ORM are proposed. As proof of concept, a working prototype that is compliant with the OMG’s Meta-Object Facility is also discussed.

Constraints on Conceptual Join Paths

To ensure that a software system accurately reflects the business domain that it models, the system needs to enforce the business rules (constraints and derivation rules) that apply to that domain. From a conceptual modeling perspective, many application domains involve constraints over one or more conceptual schema paths that include one or more conceptual joins (where the same conceptual object plays roles in two relationships). Popular information modeling approaches typically provide only weak support for such conceptual join constraints. This chapter contrasts how these join constraints are catered for in object-role modeling (ORM), the Unified Modeling Language (UML), the Object-oriented Systems Model (OSM), and some popular versions of entity-relationship (ER) modeling. Three main problems for rich support for join constraints are identified: disambiguation of schema paths, disambiguation of join types, and mapping of join constraints to implementation code. To address these problems, some notational, metamodel, and mapping extensions are proposed.

Assessing Enterprise Modeling Languages Using a Generic Quality Framework

Statoil, one of Norway’s largest organizations, recently wanted to standardize an enterprise modeling language for process modeling for sense-making and communication. To perform the evaluation, a generic framework for assessing the quality of models and modeling languages was specialized to the needs of the company. Five different modeling languages were evaluated according to the specialized criteria. Two languages were, through this, found as candidate languages, and further criteria related to tool and process support for using the languages in actual modeling were used to decide the language to choose for future standardization. This work illustrates the practical utility of the overall framework, where language quality features are looked upon as means to enable the creation of models of high quality. It also illustrates the need for specializing this kind of general framework based on the requirements of the specific organization.

An Approach for Evolution-Driven Method Engineering

This chapter considers the evolutionary perspective of method engineering. It presents an approach for method engineering supporting evolution of an existing method, model, or meta-model into a new one satisfying a different engineering objective. The authors hope that this approach could be helpful for a project-specific method construction. This approach proposes several different strategies to evolve from the initial paradigm model to a new one and provides guidelines supporting these strategies. The approach has been evaluated in the Franco-Japanese research project around the Lyee methodology. A new model called Lyee User Requirements Model has been obtained as an abstraction of the Lyee Software Requirements Model. The chapter illustrates this evolution case.

Analyzing and Comparing Ontologism with Meta-Models

High-level ontologies provide a model of reality and are of increasing popularity for the evaluation of modeling methods. Most of the common modeling methods have been studied using ontologies such as the BWW representation model and Chisholm’s ontology. However, at this stage only limited guidance is provided for the selection and evaluation of the appropriate ontology. The aim of this chapter is to propose meta-models for analyzing, comparing, and engineering ontologies. It discusses a methodology using extracts of meta-models for two well-known ontologies that had been used previously in Systems Analysis and Design research. The approach provides a theoretical analysis technique for evaluating these ontologies according to their equivalence, depth of structure, and comprehensiveness of scope.

Validating an Evaluation Framework for Requirements Engineering Tools

Automated support for the requirements engineering (RE) process is a recognized research area. However, the mainstream practice still relies on word processors and drawing tools rather than the requirements engineering tools (RETs). The aim of this chapter is to validate an evaluation framework for RETs. The validation process concerns an RET acquisition process for concrete organizational needs. An observation of maintaining requirements specification shows the important organizational and environmental characteristics for a proper automated support of RE process. The contribution of this work is twofold: first, the validation of the evaluation framework for RETs according to environmental needs in a specific environment, and second the identification of environmental needs, which emerge from the requirements specification maintenance process.

Comprehension of Hierarchical ER Diagrams Compared to Flat ER Diagrams

HERD is a semi-algorithmic method for creating hierarchical ER diagrams from bottom up. The method is based on packaging operations that are applied in several steps on a given flat ER diagram. The result is a hierarchy of simple and interrelated diagrams (namely ER structures) with external relationships to other such diagrams. This chapter describes the application of HERD method using an example from a hospital domain, and an experiment in which we compare the comprehension of HERD diagrams with flat ER diagrams.

Goal Modeling in Requirements Engineering

In the field of requirements engineering, goal modeling approaches have received much attention in recent years by researchers and practitioners alike. This chapter identifies the uses of these approaches in different contexts of requirements analysis phases. It examines goal modeling in terms of five methodological orientations and defines a framework for their analysis. Using this framework it provides an analysis of goal modeling approaches in a systematic and consistent manner. The aim of this analysis is to understand the best fit for purpose of different goal modeling approaches and to highlight open issues that provide a foundation for further research in this important area of requirements engineering methodology.