AC 2007-2287: DISTINGUISHING AMONG PROCESSES OF PROBLEM SOLVING, DESIGN, AND RESEARCH TO IMPROVE PROJECT PERFORMANCE

Professionals in all disciplines are continually engaged in problem solving, design, and research. Because steps in these processes appear similar, many faculty conceptualize a single, universal model for all three processes. However, for students who are just learning these processes, a universal model may not be the best way to build performance skills. This work was undertaken to help novices understand unique characteristics of each process and the circumstances under which each process is most effective and efficient. This paper examines two tools that were created to build this understanding: (i) a matrix analyzing the similarities and differences among the processes and (ii) a graphical presentation highlighting key skills that are hypothesized for each process. Effectiveness of the two tools was evaluated in a freshman design course where teams of five students work on a six-week design mini-project. Data collected included notes by the instructor, observations by peer coaches who observed an activity, and written feedback provided by student teams. In the activity, teams were asked to use the tools to distinguish between problem-solving and design activities that they had performed earlier in the semester. Next, the students were asked to classify a number of simple scenarios. Finally, feedback was solicited about the greatest strengths and areas of improvement for each of the tools as well as insights gained through this class activity. Findings were validated by separate focus groups with design faculty and with students enrolled in a capstone design course. Both students and faculty envisioned the two tools to be a natural extension of project work, prompting new insights about the role of problem solving, design, and research in engineering practice. Introduction One of the most valued skills of an engineer is the ability to solve problems. However, the definition of “problem solving” varies widely depending on the context or community in which it is used. Many faculty tend to favor a definition that is all encompassing – where any task, no matter how large or small, with an unknown solution denotes a problem. For mature problem solvers, such a definition is powerful and meaningful 1 . While valuable insights can be derived from a universal model, there may be drawbacks to doing so from the standpoint of novice problem solvers. Universal models tend to focus on methods rather than intermediate results. For those in a learning role, it is often difficult to translate abstract steps of a universal methodology into concrete, relevant actions. There may also be unique, value-added learning skills associated with problem solving, design, and research that tend to be diluted in a universal model. Lack of attention to limiting learning skills, in turn, may hinder development of expertise in a process area. Understanding key differences between problem solving, design, and research allows one to select the process that best supports a desired outcome, gives clearer vision of one’s location when executing the process, and provides guidance for making transitions between processes. The objective of this paper is not to re-define problem solving. Instead, working definitions that explicitly distinguish between problem solving, design, and research are presented to students who are then asked how this framework could benefit their project work. In addition to the qualitative feedback, comprehension of the three definitions was measured by asking students to classify common engineering challenges as primarily problem solving, design, or research. Students were also prompted to think about skills that limited their performance in problem solving, design, and research. This was initiated by asking students to process a Venn diagram prepared by the authors that highlighted learning skills likely to be associated with each process. Skills that are common to all three processes as well as those that intersect two of the processes were also hypothesized. This paper first introduces working definitions for problem solving, design, and research along with observations how these processes are commonly taught. Next two tools are outlined: (i) a table analyzing the similarities and differences between the processes in terms of common attributes and (ii) a figure highlighting key process-specific skills from the viewpoint of the authors. The bulk of the paper is a case study where the tools were examined in a freshman design course as well as a capstone design course. At the end of the paper, we present insights and edits based on student and faculty feedback. Background Definition of Processes: WordNet from Princeton University defines Problem Solving as “the area of cognitive psychology that studies the processes involved in solving problems, or, the thought process involved in solving a problem.” 2 One of the more popular definitions comes from Newell and Simon 3 which was summarized by Woods: “A situation where a person desires to resolve the gap between a goal state and an initial state. Some blockage in the gap prevents the person from immediately seeing a course of action. If there is no blockage, then the situation is an exercise, not a problem.” 4 Woods later refined this definition stating that: “Problem Solving is a process whereby a ‘best’ value determined for some objective or unknown, subject to a specific set of constraints and criteria. The problems that we focus on to solve are ones where there is no immediately apparent procedure, idea, or route to follow; if one has an idea of how to solve ‘the problem,’ then this problem is simply an exercise. What we call a problem is a real challenge; it is a situation where we really have to struggle to define it, figure out what it means, and resolve it.” 5 While these definitions offer valuable insights to veteran problem solvers, they lack detail necessary for a novice to improve their skills. In this paper, we choose to associate “problem solving” with smaller day-to-day challenges, recognizing that these “problems” are often parts of a much larger “Problem.” In an academic context, problem solving is often viewed as synonymous with homework assignments. However, the authors consider many homework assignments to be more “transfer exercises” or “analytic problem solving” 6 which involve calculations leading to one correct answer. “Creative problem solving”, in contrast, is much more open-ended and revolves around a situation—not a calculation 1 . Creative problem solving involves resolution of a discrepancy between one’s expectations and the reality of one’s situation. Where “analytic” problem solving tends to focus on strictly cognitive issues 7 , “creative” problem solving includes significant social and affective dimensions. The authors definition of problem solving is synonymous with creative problem solving. The term “design” is used fairly consistently across multiple disciplines 8 . Design commonly involves a third-party customer/user, and the innovative devising of a product or process, eg. hardware, software, or production system, that satisfies a need. Many people confuse design with fabrication. While manufacturing is often a large component of design, design involves as much planning and analysis as it does physical prototyping. Research is often a major component in evaluating faculty performance. Research starts with a gap in current knowledge and seeks to fill this gap with theory and data that is accepted by a wider community 9 . The knowledge gap does not focus on personal knowledge, but rather the knowledge of a research community. Research therefore is the discovery and dissemination of empirical knowledge that is not currently known by a community of experts. Knowledge that is new to one person, but not to others, may be better characterized as “project learning”. An example of project learning is when students “research” the literature, and write a “research paper.” Teaching the Processes: Introductory as well as advanced engineering texts 6,10 cover problem solving and design in separate sections. However, these sections are often interconnected and there is often loose use of terminology. When teaching creative problem solving special emphasis is usually given to problem definition. Sometimes the real problem is not as it first appears. Problem definition, brainstorming, data gathering, picking a best solution, implementation of the solution, and anticipating possible outcomes of implementing the solution are cited as the critical steps by Oakes et al. 6 . Wankat and Oreovicz 10 identify analysis, synthesis, generalization, simplification, creativity, and decision-making as central elements of problem solving. Many authors present guides on teaching engineering design 6,10,11 . The steps in the design process are usually given to be some variation of problem definition, gathering information, generation of ideas/alternatives, modeling, feasibility analysis, evaluation, decide on one alternative, communication, and implementation/production. All authors emphasize that design should be taught so that students can experience the steps as a process. Atman et al. 11 conclude that there are several general characteristics of a successful design process: (1) the use of a prescribed methodology that allows for flexibility and opportunistic design, (2) the effective use of transitions among design steps; and (3) the development of good conceptual models, including effective scoping of the problem. Iteration is frequently mentioned as critical to design, and important to emphasize with novice engineers who may believe that linear thinking leading to one correct endpoint is a desirable course of action. Methods of (basic) research are usually reserved for graduate study. When research principles are introduced to undergraduates, they are frequently presented in the context of “the scientific method”. For graduate students, the research process 9,12 m