What is Decision-Based Design?

HELLO READER! This is a version of a presentation on Decision Based Design (DBD) presented at the April '97 face-to-face meeting. The original ideas were developed at the January '97 face-to-face meeting. What you are about to read are the notes from the April '97 presentation with commentary from meeting attendess highlighted.

Decison-Based Design (DBD) is One Perspective on Design

There are many perspectives through which we can study design. DBD is one of the many perspectives. We believe that it has value as a perspective and are examining this perspetive on design to learn its complete value.

Understanding Decision-Based Design (DBD)

What is the purpose of attempting to understand DBD?

  • To develop consensus on what we mean when we say decision-based design
  • To determine the place of the DBD perspective on design within the current milieu of design perspectives extant within the design theory and methodology
  • To examine this perspective to develop the foundational principles of DBD
  • Commentary: The main point is that decision-making is an important part of design that can't be ignored during the design process and that may yield insight into how to perform the design process more effectively.

    Begin with Engineering Design

    To answer the question, "What is DBD?", we must first answer the question, "What is design?". Since we are approaching this from an engineering background, we prefer to answer the question, "What is engineering design?".

  • Engineering Design is an activity.
  • The results of the engineering design activity include:
  • A product of value. This product may be a tangible artifact or an intangible concept or process.
  • Knowledge associated with the product of value.
  • Knoweledge associatied with the design process by which the activity was carried out.
  • Commentary: Since DBD can be applied in a broader context than just engineering design, we may wish to broaden our focus here.

    What is DBD?

    In DBD we view design as a decision-making process involving:

  • Human values (or just "values" if you subscribe to the philosophy that only humans can have values)
  • Uncertainty
  • RISK
  • Commentary: This simple statement generated a great deal of discussion and highligts the need for work on developing an agreed-upon lexicon of terms for use in any research area. The discussion centered upon the questions, "What is the difference between uncertainty and risk?".

    Uncertainty involves the inability to have all known facts at one's disposal when making a decision. On the other hand, risk is a condition in which there are no known facts. Risk involves predictions.

    One meeting participant suggested that DBD involves ambiguity independent from uncertainty.

    There are many sources providing the values necessary and present in the decision-making proces. We identify them as follows:

  • The market in which the product will be offered, the end-user, society as a whole
  • The enterprise that is peforming the design activity
  • The desiger(s)
  • Commentary: It is interesting to observe that all those who might be sources of the values we must consder during the process, can also be made partners in the design process. The "Design For X" mentality can be reworked in a "Design With X" mentality -- as suggested by advocates for a veiw of design as more negotiation than strict decision-making.

    A DBD Process

    We begin our discussion of the DBD process by observing that design is a decision-making process. The tools used in the design process include representations, modelling, analysis, languages, etc.. The role that decisions and decision-making play during the application of each of these tools are of great interest to the DBD researcher. It is easy to agree with the statement, "Design activities involve decision-making." However, we believe that not all design activities are decision activities.

    We normally characterize a design process as a series of activities, and DBD is no different. The generic decision-based design process includes the following steps:

  • Generation of an option space
  • Exploration or formulation of the option space
  • Computation of expectation of outcomes resulting from the choice of a particular option
  • Establishment of a value system to rank the expectations of all the outcomes possible from selection of options
  • Commentary: The participants gained consensus on the point that design involves decision-making, but did not believe that all activities in a design process are strictly decision activity steps. If we accept that DBD decisions always involve the consideration of values, uncertainty, and risk, then how do we use DBD to generate an option space ? The group agreed that the option space is the set of potential solutions that will meet the needs articulated in the design problem. In considering this space at the begninning of a problem, usually no attempt is made to weed out solutions that are high-risk or that have negative consequences (i.e., may conflict with a value of one of the end-users). That being the case, the creation of this option space seems NOT to be the result of a strict DBD process. Yet, the generation of an option space requires MANY decisions.

    These observations support the view that there may be activities in the design process that are decision activities but do not require the same type of decision-making as that used for a selecting a final design -- which necessarily involves values, uncertainty, and risk assessment.

    It appears then, that there is a fuzzy front end -- a soft boundary -- delineating the beginning of the hard core, value-laden DBD process. Once the option space exists, then the next three steps take place with a concern for values, uncertainty, and risk.

    What is Not DBD?

    Commentary: Our original intentention in this discussion was to contrast the above definitions of the DBD process to other design methods popular in the research community. This way we could answer the question, "What is not DBD?" However, our meeting participants quickly directed us to the more appropriate question, "Where and how does DBD fit in?

    Where Does DBD Fit in?

    Decision-based design is a framework by which the design activity can be understood, structured, managed, and evaluated. A good way to define a science of decision-based design is to determine how this perspective on design fits into the overall body of knowledge on design theory and methodology. We can then compare the DBD approach to others.

    Several perspectives on design have emerged from Open Workshop discussions:

  • Design as decision-making
  • Design as problem-solving
  • Design as negotiation
  • Information view of design
  • Cybernetic view of design
  • Energetic view of design
  • Where do we go From Here?

    This introduction to decision-based design (DBD) should give Open Workshop visitors some background for commenting on the discussion. To do so, please visit our "Open Research Questions" page to enter your comments.

    If you have comments or suggestions, email me at lschmidt@eng.umd.edu