Frank Harmsen

Meta-modelling based assembly techniques for situational method engineering

Abstract Method engineering for information system development is the discipline to construct new advanced development methods from parts of existing methods, called method fragments. To achieve this objective, we need to clarify how to model the existing methods and how to assemble method fragments into new project-specific methods, so-called situational methods. Especially, to produce meaningful methods, we should impose some constraints or rules on method assembly processes. In this paper, we propose a framework for hierarchical method modelling (meta-modelling) from three orthogonal dimensions: perspectives, abstraction and granularity. According to each dimension, methods and/or method fragments are hierarchically modelled and classified. Furthermore, we present a method assembly mechanism and its formalization as a set of rules. These rules are both syntactic and semantic constraints and presented in first order predicate logic so that they can play an important role in the assembly process of syntactically and semantically meaningful methods from existing method fragments. The benefit of our technique is illustrated by an example of method assembly, namely the integration of the Object Model and Harels Statechart into Objectcharts.

Assembly Techniques for Method Engineering

As projects for developing information systems are getting larger and more complicated, we need to have more advanced development methods suitable for every development situation. Method engineering is the discipline to construct new methods from parts of existing methods, called method fragments. To achieve this objective, we need to clarify how to model the existing methods and how to assemble method fragments into new project-specific methods, so-called situational methods. Especially, to produce meaningful methods, we should impose some constraints or rules on method assembly processes. In this paper, we propose a framework for hierarchical method modelling (meta-modelling) from three orthogonal dimensions: perspectives, abstraction and granularity. According to each dimension, methods and/or method fragments are hierarchically modelled and classified. Furthermore, we present a method assembly mechanism and its formalization as a set of rules. These rules are presented in first order predicate logic and play an important role in the assembly process of meaningful methods from existing method fragments. The benefit of our technique is illustrated by an example of method assembly, namely the integration of the Object Model and Harels Statechart into Objectcharts.

Design and implementation of a method base management system for a situational CASE environment

Situational method engineering focuses on configuration of system development methods (SDMs) tuned to the situation of a project at hand. Situational methods are assembled from parts of existing SDMs, so called method fragments, that are selected to match the project situation. The complex task of selecting appropriate method fragments and assembling them into a method requires effective automated support. The paper describes the architecture of a tool prototype offering such support. We present the structure of its central repository, a method base containing method fragments. The functions to store, select and assemble these method fragments are offered by a stratified method base management system tool component, which is described as well.

A Method Engineering Language for the Description of Systems Development Methods

We propose a Method Engineering Language, called MEL, as a formal representation language for the description of method fragments, i.e. the development processes, and the products and deliverables of a systems development method. The language allows representing the structures of method fragments, the applicable consistency rules, and a variety of method assembly operators, all of which the semantics are formally defined. The MEL language is illustrated by a simple example of a Sequence Diagram of UML.

Intranet Facilitated Knowledge Management: A Theory and Tool for Defining Situational Methods

The methodical support of large systems development efforts is still problematic, due to the cumbersome access to paper manuals, the lack of modification capabilities, and the inflexible standard project structures. This paper presents a case in which a tool based on intranet technology is used to facilitate the management of systems development knowledge acquired in an empirical research project on situational method definition. A fundamental theory for situational method definition, called the S4 model, has been developed and validated. The S4 model expresses the argumented dependencies of the Situation factors that determine the Succes of a project, to which is contributed by Scenario aspects, that yield a Situational method. This systems development knowledge and experience was captured and recorded according to the ASSUME (Acquisition, Structuring, Storage, Utilization and Maintenance of Expertise) approach for knowledge management. The knowledge acquisition exercise took place in three large organizations for information systems development involving 43 experienced project managers. The collected results of the empirical work were filled in the Mouseion tool, an intranet application with hypermedia facilities (WWW, ftp, newsgroups and search engines) for situational method support to systems development.

Success and Failure in Offshore Development Projects

This chapter describes an in-depth analysis of successful and unsuccessful offshore custom software development (CSD) projects. Offshore projects tend to be unsuccessful, because physical, time, cultural, organizational, and stakeholder distances negatively influence communication and knowledge exchange between onshore and offshore project team members. The success rate of 19 offshore CSD projects was characterized with regard to scope, quality, time, and costs, by interviewing onshore and offshore project managers. Unsuccessful projects had a complex organization and team members who did not work together in previous projects. All reasons for success and failure were categorized and compared. A characteristic that successful projects had in common was the availability of informal mutual adjustment, which means facilitating the informal communication between the team members in the right way. A major characteristic that unsuccessful projects had in common was improper planning, which has a large influence on the team results in an offshore CSD project. The implementation of standards was neither mentioned by the successful projects as a major reason for success, nor by the unsuccessful projects as a major reason for failure. This research advises that in order to be successful, a project manager of a new offshore CSD project should not spend too much time on standards, but on planning and informal mutual adjustment.

A Decision Procedure for the Where and How to Outsource

Many organizations are currently deciding whether to insource or outsource their IS function or parts thereof, but are unsure as to what kind of organizational structure to arrange for sourcing and where to locate the sourced activities. To assist in this matter, several IT consultancy firms are providing sourcing consultancy to their clients, resulting in a large body of practical knowledge that is stored in the experience of consultants. Systematic scientific knowledge is largely based upon qualitative case studies. This chapter presents the results of a policy capturing study that attempts to quantify the sourcing advice practice. Twenty-nine management consultants with experience in sourcing advice were presented with examples of situations in which sourcing advice was to be given. Using Factor Analysis and Multilevel Regression the researchers obtained insight into how variables such as costs, flexibility, time-to-market, and quality influence decisions with regard to both sourcing location as well as sourcing relationship. The research method that was used proved to be helpful in the explicitation of knowledge of consultants but needs further refinement.

IT Performance Management in Five Organizations

This research provides insight in how organizations can achieve and optimize the benefits of IT performance management. Added value from investments is achieved through a focus on IT performance management that enables the removal of non value-adding activities and processes (and thereby organizational efficiency) and the enhancement of organizational effectiveness. In our work we have used a framework that contains IT performance management characteristics that are mapped to organizations, to determine their IT performance management maturity. Although the organizations subjected to our study operate in the same sector, our results show a high diversity in applied IT performance management approaches and the different maturity levels between organizations.