An official website of the United States government
This study offers insight into the processes of expert designers at the Jet Propulsion Laboratory (JPL) and how they make use of heuristics in the design process. A methodology for the extraction, classification, and characterization of heuristics is presented. Ten expert participants were interviewed to identify design heuristics used during early stage space mission design at JPL. In total, 101 heuristics were obtained, classified, and characterized. Through the use of post-interview surveys, participants characterized heuristics based on attributes including source/origin, applicability based on concept maturity, frequency of use, reliability, and tendency to evolve. These findings are presented, and statistically analyzed to show correlations between the participant perceptions of frequency of use, reliability, and evolution of a heuristic. Survey results and analysis aim to identify valid attributes for assessing the applicability and value of multiple heuristics for design practice in early space mission formulation.
more »
« less
Framework for the Evolution of Heuristics in Advanced Manufacturing
Abstract This study works toward addressing a knowledge gap in understanding how heuristics are developed, retrieved, employed, and modified by designers. Having a better awareness of one’s own set of heuristics can be beneficial for relaying to other team members, improving a team’s training processes, and aiding others on their path to design expertise. The ability to understand and justify the use of a heuristic should lead to more effective decision-making in systems design. To do this, the heuristics and their characteristics must be extracted using a repeatable scientific research methodology. This study describes a unique extraction and characterization process compared to prior literature. It includes some of the first work towards documenting heuristics for both designers and operators in a hybrid manufacturing setting. Eight participants performed a series of two design journals, two interviews, and one survey. Heuristics were extracted and refined between each method and then verified by participants in the survey. The surveys produced novel statistically significant findings in regard to heuristic characterizations, impacting how participants view how often a heuristic is used, the reliability of the heuristic, and the evolution of the heuristic. Lastly, an alternate perspective of heuristics as an error management bias is highlighted and discussed.
more »
« less
- Award ID(s):
- 2207448
- PAR ID:
- 10418249
- Date Published:
- Journal Name:
- Journal of Mechanical Design
- Volume:
- 145
- Issue:
- 1
- ISSN:
- 1050-0472
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract Design heuristics are traditionally used as qualitative principles to guide the design process, but they have also been used to improve the efficiency of design optimization. Using design heuristics as soft constraints or search operators has been shown for some problems to reduce the number of function evaluations needed to achieve a certain level of convergence. However, in other cases, enforcing heuristics can reduce diversity and slow down convergence. This paper studies the question of when and how a given set of design heuristics represented in different forms (soft constraints, repair operators, and biased sampling) can be utilized in an automated way to improve efficiency for a given design problem. An approach is presented for identifying promising heuristics for a given problem by estimating the overall impact of a heuristic based on an exploratory screening study. Two impact indices are formulated: weighted influence index and hypervolume difference index. Using this approach, the promising heuristics for four design problems are identified and the efficacy of selectively enforcing only these promising heuristics over both enforcement of all available heuristics and not enforcing any heuristics is benchmarked. In all problems, it is found that enforcing only the promising heuristics as repair operators enables finding good designs faster than by enforcing all available heuristics or not enforcing any heuristics. Enforcing heuristics as soft constraints or biased sampling functions results in improvements in efficiency for some of the problems. Based on these results, guidelines for designers to leverage heuristics effectively in design optimization are presented.more » « less
-
Abstract Additive manufacturing (AM) can produce designs in a manner that greatly differs from the methods used in the older, more familiar technologies of traditional manufacturing (TM). As an example, AM's layer-by-layer approach to manufacturing designs can lead to the production of intricate geometries and make use of multiple materials, made possible without added manufacturing cost and time due to AM's “free complexity.” Despite this contrasting method for manufacturing designs, designers often forgo the new design considerations for AM (AM design heuristics). Instead, they rely on their familiarity with the design considerations for TM (TM design heuristics) regardless of the intended manufacturing process. For designs that are intended to be manufactured using AM, this usage of TM design considerations is wasteful as it leads to unnecessary material usage, increased manufacturing time, and can result in designs that are poorly manufactured. To remedy this problem, there is a need to intervene early in the design process to help address any concerns regarding the use of AM design heuristics. This work aims to address this opportunity through a preliminary exploration of the design heuristics that students naturally leverage when creating designs in the context of TM and AM. In this study, 117 students in an upper-level engineering design course were given an open-ended design challenge and later tasked with self-evaluating their designs for their manufacturability with TM and AM. This evaluation of the students' designs was later repeated by relevant experts, who would identify the common design heuristics that students are most likely to use in their designs. Future studies will build on these findings by cementing early-stage design support tools that emphasize the significant heuristics found herein. For example, this work found that the design heuristic “incorporating complexity” was the most significant indicator of designs most suited for AM and should therefore be highly encouraged/emphasized when guiding designers in the use of AM. In doing so, it will be possible for early-stage design support tools to maximally improve designs that are intended to be manufactured for AM.more » « less
-
In designing complex systems, systems engineers strive to turn an existing situation into a situation that is most preferred. A rational decision maker would choose the alternative that maximizes the expected utility of the existing situation, but there are significant computational challenges to this approach. To overcome these challenges, most decision makers revert to heuristics. In this paper, we propose a conceptual framework for heuristics in design. A preliminary empirical study of designers for a robotics design problem was conducted to observe how participants apply heuristics. Participants having at least 2 years of experience designing robots were recruited to partake in a robotics design session in which participant were given 45 minutes to work on a design problem. A preliminary heuristics extraction method was developed, and the identified heuristics were studied to find trends within the overall set. These trends were the basis of a taxonomy of heuristics consisting of three initial classification methods: design phase, field of study, and action intent. The heuristics and classifications are presented, along with the challenges from extracting heuristics and recommendations for future work to further research design heuristics and to improve the method for extraction.more » « less
-
null (Ed.)Heuristics are essential for addressing the complexities of engineering design processes. The goodness of heuristics is context-dependent. Appropriately tailored heuristics can enable designers to find good solutions efficiently, and inappropriate heuristics can result in cognitive biases and inferior design outcomes. While there have been several efforts at understanding which heuristics are used by designers, there is a lack of normative understanding about when different heuristics are suitable. Towards addressing this gap, this paper presents a reinforcement learning-based approach to evaluate the goodness of heuristics for three sub-problems commonly faced by designers while carrying out design under resource constraints: (i) learning the mapping between the design space and the performance space, (ii) sequential information acquisition in design, and (iii) decision to stop information acquisition. Using a multi-armed bandit formulation and simulation studies, we learn the heuristics that are suitable for these sub-problems under different resource constraints and problem complexities. The results of our simulation study indicate that the proposed reinforcement learning-based approach can be effective for determining the quality of heuristics for different sub-problems, and how the effectiveness of the heuristics changes as a function of the designer's preference (e.g., performance versus cost), the complexity of the problem, and the resources available.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1115/1.4055622
