Engineering design at the edge of rationality
Gurnani, Ashwin Prabhu
MetadataShow full item record
In recent times, the engineering design community has accepted and established that product design is primarily a decision-making process. Decisions are made at different stages of the design process with varying amounts of information. The study of engineering design as a series of decisions is termed as Decision-Based Design (DBD) . The body of research pertaining to DBD, design decision modeling and decision support development is well established and ever-growing in both academia and industry. The fundamental assumption in all developed decision support tools and methodologies is that the decision maker makes rational choices. A rational choice is one where the decision maker selects the option that maximizes the yield on the preferred outcome. Decisions that do not subscribe to this accepted belief are termed as mistakes or irrational choices and discarded. The focus of this dissertation is investigating the assumption that engineers are capable and expected to make rational design decisions. This dissertation approaches the study of rationality in design decision making through two research issues. The first research issue pertains to achieving an understanding of how rationality and mistakes are defined. Since design decisions are made by engineers, the study of rationality in engineering requires some understanding of the human mind. Research in behavioral economics, psychology, and cognitive science involves understanding the human mind. This dissertation leans on the research in these fields and accepts the suggested notion of Bounded Rationality . Bounded rationality refers to the inability of human beings to accurately choose "rational" options as prescribed by decisions support models such as expected utility. Some of the causes of bounded rationality are the inability to process large amounts of information, lack of precise problem definitions, inability to distinguish between alternatives from a large set, developing simple mental heuristics for complex decision problems etc. These factors also apply to engineers making design decisions. The research in this dissertation uses the basis of bounded rationality to model engineering decisions and studies the impact of propagating these decisions within engineering design paradigms. The focus of the second research issue is on leveraging an understanding of rationality and mistakes in engineering design to develop approaches to improve existing design practices. Towards this end, this dissertation presents two approaches. The first approach incorporates decisions that exhibit bounded rationality within a decentralized design paradigm to achieve design solutions that are an improvement over traditionally obtained designs. The second approach seeks to reduce errors resulting from inaccurate information communication between designers through the use of Really Simple Syndication (RSS) feeds within a web- based environment. This allows for accurate, immediate and autonomous communication of latest design information. Through the process of this dissertation, the aspiration is to better understand the design decision making process, to identify the causes for erroneous decisions and to leverage this understanding to improve upon design practices. Decision making in product design is highly interdisciplinary involving large number of people from different educational, cultural and technical backgrounds. It is no surprise then that a dissertation on decision making, such as this one, spans several different research areas.