1. INTRODUCTION
1.5. RESEARCH STRATEGY
1.5.1. DESIGN SCIENCE APPROACH
Design science45 studies artificial artefacts and it involves a search process to discover an effective solution to a problem. Problem solving uses available means to achieve desired goals while satisfying existing laws in the environment. Designing46 is a task and process within and across traditional disciplines. The purpose of designing is to take a specification setting needs (an idea for a new or revised process or product, and a set of statements of performance and constraints) and transform it into full instructions for manufacturing a product and/or implementing a process.
The design process includes frequently repeated subprocesses of problem solving. Iterative working modes are nearly inevitable, because the products, processes or services to be designed are usually difficult to understand fully at once.
Cross has advocated design sciences and the understanding of them. To diminish confusion and controversy over the nature of valid design research, Cross places some criteria for proper design science research47:
• Purposive (based on an identified issue or problem worthy and capable of investigation)
• Inquisitive (seeking to acquire new knowledge)
• Informed (conducted from an awareness of previous, related research)
• Methodical (planned and carried out in a disciplined manner)
• Communicable (generating and reporting testable results which are accessible by others) According to Cross, the list describes good research in any discipline. Cross48 continues that design research has to concern the development, articulation and communication of design knowledge. He has given three sources of design knowledge: people, processes and products. Based on these sources of design knowledge, Cross has developed taxonomy of design research.
44 Hevner, March, Park and Ram: Design Science in Information Systems Research (2004) MIS Quarterly, Vol. 28, No. 1, p. 75
45 Simon: The Sciences of the Artificial (1996) pp. 4-6, 111-138
46 Eder: Developments in Education for Engineering Design: Some Results of 15 Years of Workshop Design-Konstruktion Activity in the Context of Design Research (1994)
Journal of Engineering Design, Vol. 5, No. 2, pp. 135-136
47 Cross: Design as a Discipline (2002)
The Inter-Disciplinary Design Quandary Conference, 13th February 2002, De Montfort University
48 Cross: Design Research: A Disciplined Conversation (1999) Design Issues, Vol. 15, No. 2, pp. 5-9
Cross’ taxonomy of design research is the following:
• Design epistemology (study of designerly ways of knowing)
• Design praxiology (study of the practices and processes of design)
• Design phenomenology (study of the form and configuration of artefacts)
According to Cross, design knowledge resides firstly in people: in designers especially, but to some extent in everyone. Designing is a natural human trait. Many of the most valued achievements of humankind are works of design, including anonymous, vernacular design as well as the "high design"
of professionals. Therefore, the human ability to design is an immediate subject of design research. It is strongly related to considerations of how people learn to design and how to teach designing.
Design knowledge resides secondly in its processes. A major part of design research focuses on methodology: the study of the processes of design and the development and application of design techniques. Computers have stimulated a wealth of research into design processes, enabling new practices in industries such as concurrent engineering.
Thirdly, it cannot be forgotten that design knowledge resides in the products themselves. Knowledge is displayed in the forms, materials and finishes which embody design attributes. A lot of everyday design work comes from using precedents or previous exemplars, not because of designers’ laziness but because the exemplars actually contain knowledge of what the product should be.
Finally, Cross states that it would be foolish to disregard or overlook this informal product knowledge simply because it has not been made explicit yet. It is a task of design research to make implicit product knowledge explicit. Furthermore, he states that there is a need for a distinction between works of practice and works of research. He says that normal works of practice cannot be regarded as works of research. Cross declares that the whole point of doing research is to extract reliable knowledge from either the natural or the artificial world, and make knowledge available to others.
Spiral Research Model in the Design Science Approach
Design science research49 is inevitably iterative. The search for the best, or optimal, design is often intractable for realistic solutions. Heuristic search strategies produce feasible, good designs for implementations in their environment. Design is essentially a search process to discover a suitable solution to a problem. Simon50 proposes that problem solving can be viewed as utilizing available means to reach desired ends while satisfying laws existing in the environment. He describes the nature of the design process as a cycle of generating alternatives and testing the generated alternatives, as illustrated in Figure 8.
Figure 8. Design as a Search Process
49 Hevner, March, Park and Ram: Design Science in Information Systems Research (2004) MIS Quarterly, Vol. 28, No. 1, pp. 88-89
50 Simon: The Sciences of the Artificial (1996) pp. 128-130
Generation of Solution Alternatives Testing Generated Alternatives against Requirements and Constraints
Hevner et al: Design Science in Information Systems Research, MIS Quarterly 28:1 (2004), p. 89 (adapted)
Simon states that abstraction and representation of appropriate means, ends and laws are crucial in design science research. These factors are problem and environment-dependent, invariably involving creativity and innovation. The means are the set of actions and resources available for constructing a solution. The ends represent goals and place constraints on the solution. The laws are uncontrollable forces in their environment. Effective design requires knowledge of both the application domain (e.g.
requirements and constraints) and the solution area (e.g. technical and organisational).
Simon continues that design science research often starts with a simplified problem presenting only a subset of relevant means, ends and laws, or by decomposing a problem into simpler subproblems.
The research progresses iteratively as the scope of the design problem clarifies. When the means, ends, and laws are refined and become more realistic, the design artefact becomes more relevant and valuable. Spiral research models are common in design science, because research work is not easy to plan in detail beforehand. It can even difficult to decide which data to collect. The researcher must be prepared to change plans when the investigation deepens the understanding of the issue and the shift to the iterative approach51 is necessary. Iteration, returning to an earlier stage of the research process, makes the research process resemble a spiral rather a linear succession of decisions.
The development spiral52 can be regarded to consist of a repetitive cycle: (1) action proposal, (2) evaluation, (3) reflection, (4) abstraction, and (5) planning changes, as illustrated in Figure 9. The action proposal is expected to improve in each development round of the spiral. The development spiral continues until an acceptable result is found, or the resources are exhausted.
Figure 9. Development Spiral
Resulting Knowledge of Design Science
Using Van Aken’s terminology, the resulting knowledge53 of design science concerns with three designs: (1) object-designs, (2) realisation-designs, and (3) process-designs. The object-design knowledge comprises the design of the intervention or the artefact. The realisation-design knowledge consists of the ability to plan the implementation of the intervention or the building of the artefact. The process-design knowledge includes the professional’s own plans for the problem solving. In other words, it involves methods to be used to design solutions to problems.
51 Arteology > Models for Research Process
Routio, Pentti (The University of Art and Design Helsinki) http://www2.uiah.fi/projects/metodi/144.htm#tyo (25.05.2005)
52 Arteology > Action Research
Routio, Pentti (The University of Art and Design Helsinki) http://www2.uiah.fi/projects/metodi/120.htm#intrinsc (31.05.2005)
53 Van Aken: Management Research Based on the Paradigm of the Design Sciences: The Quest for Field-Tested and Grounded Technological Rules (2004)
Journal of Management Studies, Vol. 41, No. 2, pp. 226-228
Empiria
5 1
Planning Changes
Abstraction
Reflection
Action Proposal
Evaluation 3
4
2
Routio: Arteology, http://www2.uiah.fi/projects/metodi/120.htm#intrinsc (31.05.2005) (adapted) Theory
Prescriptions (technological norms in Olkkonen’s terminology) are an important category within each three types of design knowledge. The logic of a prescription follows the idea “if you want to achieve Y in situation Z, perform action X”. There are algorithmic prescriptions, which operate like a formula.
They are typically in a quantitative format and they can be proven through deterministic or statistical generalisation based on observations. However, many prescriptions in design science are heuristic by nature. They can rather be formulated as “if you want to achieve Y in situation Z, then something like action X will help”. “Something like action X”, presents a design exemplar. A design exemplar is a general prescription, which has to be translated to the specific problem at hand. To solve a problem, one has to design a specific variant of that design exemplar.
According to Olkkonen54, technological norms are used in business science to describe normative clauses, recommendations, in order to improve decision-making in organisations. They promote scientification of human practices and constitute knowledge in given situations. Olkkonen claims that technological norms form the ideological core of the research philosophy in normative research.
Successful Research in the Design Science Approach
The fundamental principle of design science research is that knowledge and understanding of a design problem and its solution are acquired in building and applying the artefact. Seven guidelines, presented in Table 4, have been developed to instruct on how to carry out proper design science research55. These guidelines are supposed to apply in constructive research. Researchers’ creative skills and judgment determine when, where, and how to apply these guidelines. Reviewers, editors, and readers will later decide how nicely the research satisfies the intent of the guidelines.
The used terminology is according to Information Systems.
Table 4. Design Science Research Guidelines
Guideline Description
1. Design as an artefact Design-science research must produce a viable artefact in the form of a construct, a model, a method, or an instantiation.
2. Problem relevance The objective of design-science research is to develop technology-based solutions to important and relevant business problems.
3. Design evaluation The utility, quality, and efficacy of a design artefact must be rigorously demonstrated via well-executed evaluation methods.
4. Research contributions Effective design-science research must provide clear and verifiable contributions in the areas of the design artefact, design foundations, and/or design methodologies.
5. Research rigor Design-science research relies upon the application of rigorous methods in both the construction and evaluation of the design artefact.
6. Design as a search process The search for an effective artefact requires utilising available means to reach desired ends while satisfying laws in the problem environment.
7. Communication of research Design-science research must be presented effectively both to technology-oriented and management-technology-oriented audiences.
54 Olkkonen: Johdatus teollisuustalouden tutkimustyöhön (1993) p. 56-58
55 Hevner, March, Park and Ram: Design Science in Information Systems Research (2004) MIS Quarterly, Vol. 28, No. 1, pp. 82-83
Hevner et al.: Design Science in Information Systems Research, MIS Quarterly 28:1 (2004), p. 83