Engineers often stipulate the existence of user preferences in formulating decision models. Research in psychology and the behavioral sciences indicate that preference structures are much more complex than we often assume and frequently contain inconsistencies. This research merges engineering and psychology research to understand the structure of preferences and derive preference assessment tools that are rigorous from the behavioral sciences viewpoint and useful to product designers. Applications include environmentally sustainable products and product aesthetics.

Keywords: Product semantics, preference inconsistency, Kano method, interactive GAs, conjoint analysis, prefmaps

Product purchase and use heuristics, Identification of crux/sentinel heuristic
by Erin MacDonald, PhD Candidate

I hypothesize that judgments and decisions about products have their own specialized subset of heuristics that allow people to make, for example, "fast and frugal" decisions about products. Some of these heuristics come from judgments of information sources like advertising and store placement. Product heuristics are associated with the product's design. Product heuristics encompass research in emotional design, which studies the emotional connotations that a product's visual form conveys, and product semantics, which postulates that a product's form include meaning. Emotional and semantic research focuses on perception heuristics, and mainly visual heuristics, such as researching car silhouettes that people judge as "sporty." If the wide breadth of heuristics discovered in decision-theory psychology applies also to product heuristics, then product heuristics must come from a variety of sources reaching beyond direct perceptual information to other cognitive processes. I currently focus my research in product heuristics on green products, in response to the fundamental question of my thesis work. Judging the "greenness" of a product remains a daunting task. Gaining control over the judgment information conveyed through product heuristics will decrease reliance on advertising to shape the customers' judgments, and thus avoid customer skepticism of "greenwashing." Green products heuristics are still nascent and fluctuating, so understanding why and how customers use product heuristics will ensure that the heuristics I discover remain applicable beyond short-lived styling trends.

I have tested for the heuristic I term the crux/sentinel attribute relationship by creating a heuristic identification method based on product attribute importance and preference inequalities exhibited across different choice scenarios. The experiment, an online survey about paper towels, demonstrated that people use towel quilting (sentinel) to judge absorbency (crux). While I had hypothesized that people use recycled paper content to judge towel quality, the data did not support this conclusion.

Downloadable materials: ODE poster

Decision perspectives in product design, Consistently imagining delight versus necessity
by Erin MacDonald, PhD Candidate

The integration of differing viewpoints in the design process has been addressed in many forms by other researchers. Judgments and decisions enter into the product design process from a variety of perspectives including: engineering, customer, management, and artistic design. Decisions are made at different levels of expertise. Judgments and the accompanying preference assessments can be quantitative or qualitative. Reports and assessments of preference have different levels of consistency across report forms, type of decision, or particular decider. Judgments and decisions from different perspectives can be integrated in a variety of ways. In this research theme, I study the integration of different decision-making processes and judgments in the design process to offer methods that address the problems and potential advantages found in viewing decisions from not only different perspectives but also from different types of resolution such as quantitative and qualitative, expert and novice, rational and non-rational, and consistent and inconsistent.

I have shown that, across a specific set of related survey questions, people consistently imagine product attributes delighting them, but do not consistently imagine product attributes as necessities (MacDonald, Gonzalez & Papalambros, 2007a). Interestingly, not all experimental subjects exhibited this tendency, indicating that there may be a separate identification metric for this population. When designers solicit customer input on the necessity of product attributes, they must keep this discrepancy in mind, as delight versus necessity of a product attribute can result in assigning a variable versus a parameter to the attribute, and thus effect decision trade-offs in the design.

Downloadable materials: ODE poster

Assessing and Applying Product Shape Preference Within Design
by Jarod Kelly, PhD Candidate

In a growing global economy, consumer products within mature markets that garner intense competition are increasingly selected by users based on aesthetic attributes because the technical attributes across product offerings are well met by producers. Thus, given a choice between several products which all meet certain technical expectations, users may make choices based on the visual aspects of a product. We examine the use of several methods for assessing the shape preference of a user for the design of a cola bottle. By doing this we can explore how shape preference can be quantitatively applied to product design, and how shape preference can tradeoff with more traditional engineering design objectives.

In this research we explore shape preference using both conjoint analysis and interactive genetic algorithms (IGAs). Conjoint analysis is a tool developed within the marketing community which has been applied to understand preference for a wide range of products and features. IGAs have been developed within the computer science community, and use evolutionary techniques allow people to interact with designs. This work has explored the assumption that IGAs can be used to understand preference, experimental results suggest that this assumption is correct. By decomposing a design into a parametric model we can dictate specific parameters that can be treated as variables. In doing so, the shape of the product can be analytically examined. We used the mentioned techniques to examine shape preference and then use engineering models to understand how different design objectives can be viewed as tradeoffs with shape preference.

Vehicle Attributes Difficult to Include in Choice Models
by Bart D. Frischknecht, PhD Candidate

Preferences elicited through stated preference questionnaires using such techniques as choice-based conjoint are limited as to the product attributes that can be considered. Such limitations include not only the number of attributes a respondent considers simultaneously but also the attributes that can be adequately described in a written or visual survey context. Choice-based conjoint alone falls short in capturing consumer preferences for vehicles, where experiential attributes such as ride, handling, occupant accommodation, noise, and vibration, may play an important role in vehicle purchase or product recommendation. Given an enterprise product development framework and a choice model for a subset of identified product attributes, a parametric study can be conducted to indicate the "cost" to improve the vehicle attributes not included in the consumer choice model.

Downloadable materials: ODE poster

Subjective Attributes are Important in the Decision Process
by Tahira Reid, PhD Candidate

It has been well established that functionality and usability are no longer sufficient attributes to assist consumers in making decisions about products. Products are a combination of sensory cues that convey different concepts to consumers. Of interest are visual cues. A visual cue is any physical change to a product or artifact that can be detected by the eyes. A variety of methods exist in marketing to assess consumer preference based primarily on utility and some subjective attributes. What’s missing from most models are the inclusion of a perceptual link that influences consumers with the decision making. Combining methods from industrial design, psychology and engineering, a more holistic approach towards enhancing the product design process can be realized.