Search

With the growing demand for distributed collaboration for large and complex design in modern engineering, the collaboration inefficiencies of traditional computer-aided design (CAD) tools are increasingly conspicuous. Emerging cloud-based multi-user computer-aided design (MUCAD) platforms bring a new working style for CAD in the form of real-time synchronous collaboration. Little research exists to characterize collaboration in CAD, and specifically the synchronicity of collaboration has yet to be examined. In this study, we analyzed the backend action logs of 101 teams’ design processes from a large-scale virtual robotic design competition, where all designs were modelled in a commercially available MUCAD platform. Metrics of interest were analyzed with regression and mediation analyses to uncover factors that correlated to a team’s success in the competition. Results show that team size is a positive predictor of team performance. Large teams, which tend to see a large amount of time commitment from members, were more likely to perform more CAD actions and achieve high scores from the competition. This suggests that the benefits of collaboration (e.g., economies of division of labor, learning) outweigh the potential downsides (e.g., coordination overhead, free riding) in this context. While controlling for team size, increased synchronous collaboration occurrences were observed to negatively correlate to teams’ performance — a novel finding which we discuss in detail. Thus, we conclude that although large teams benefited from the MUCAD environment, a tendency for synchronous real-time collaboration did not coincide with higher performance. This study provides important evidence in the ongoing design and innovation research fields aiming to better understand collaboration. Future research should investigate the characteristics of effective collaboration strategies in MUCAD environments to develop best practice for the increasing number of design teams moving to such tools.

Previous efforts in the area of collaborative computer-aided design (CAD) suggest that a team of designers working synchronously in a multi-user CAD (MUCAD) environment can produce CAD models faster than a single user. Our research is the among the first to investigate assemblies in MUCAD. Due to the lack of hierarchical feature dependency in assemblies, we propose that CAD teams can optimize assembly through modularization and parallel execution. In our study, 20 participants were tasked with assembling pre-modelled CAD parts of varying complexity in teams of one, two, three or four. We analyze audio recordings, team activity, and survey responses to compare the performance of individuals and virtual collaborative teams during assembly, while working with the same MUCAD platform. This paper features a multimodal approach to analyze team trends in communication, workflow, task allocation and challenges to determine which factors are conducive to the success of a multi-user CAD team and which are detrimental. In our work, the success of a team is measured by its productivity score, which is the number of mates added by a team within a given time frame. We present evidence that teams can complete an assembly in less calendar time than a single user, but single users are more efficient based on person-hours, due to communications and coordination overheads. Surprisingly, paired contributors exhibit an assembly bonus effect. These findings represent a preliminary understanding of collaborative CAD assembly work. Our work supports the claim that collaborative assembly activities have the potential to improve the capabilities of modern product design teams, delivering products faster and at lower cost. We identify areas for future research, and highlight areas of improvement for collaborative CAD platforms and engineering design teams.

Inspired by popular personality type indicators, we develop a framework for classifying individuals by their computer-aided design (CAD) behaviours. We are motivated by the trend of modern CAD software towards cloud platforms and expanded collaborative features. Cloud-CAD platforms enable collaboration by increasing access, and reducing conflicts and barriers to file-sharing. In order to generate insight to support CAD collaboration, we analyze the real-world data from an industry partner’s product development project, consisting of eight professional designers working on a cloud-CAD platform. This data corresponds to more than 1,420,000 actions over a span of eight months. Via hierarchical clustering, we group 79 unique CAD activities into 14 categories of CAD action groups, such as Part Studio, Assembly, Comment, View/Scan and Export. Next, we identify the degree to which each of the eight designers performs activity in these CAD action groups. We demonstrate the usefulness of this framework by highlighting insights revealed by the CAD action group mapping, confirmed via discussion with the industry partner. This CAD-type behaviour framework provides a tool for assessment and reflection on the types of behavioural tendencies present or missing on a team of designers. It can assist CAD educators and trainees in understanding their comprehensive CAD learning trajectory. Future extensions of the framework could leverage artificial intelligence techniques to provide real-time feedback on designer roles.

Online communication and collaboration tools are changing the way teams design products. The tools also generate a rich data source from which to study trends in communication. This paper focuses on how engineering teams utilize Slack, a popular team messaging software platform. We aim to better understand communication and coordination in product design teams via analysis of team social network dynamics, unique patterns of chat-like messaging (emoji usage), and the evolution of communication topics over time. Our study analyzes the online interactions of 32 teams, sent during a 3-month senior undergraduate product design course. These 400,000+ messages represent the team communications from 4 years of teams, with 17–20 students per team. We find that 1) Slack communications resulted in high density network maps, 2) network analysis of teams reveals that leaders have more central positions in the network, 3) strong teams have lower average centrality among members, equivalent to less public channel membership per person, 4) stronger teams use emojis at a higher rate, and 5) emojis are used most by leaders and highly connected members. These findings represent preliminary foundations for best practices in online messaging, which may lead to more effective collaboration in product design.

Technology is transforming the way engineering designers work and interact with others; Synchronous collaborative computer-aided design (CAD) tools allow designers to manipulate the same model at the same time. We present a new method using automated facial emotion detection software and cursor tracking to map designer emotions and corresponding designer activities in synchronous collaborative CAD. We present findings from a dataset of 9 participants that were assigned to two distinct working styles in the same synchronous CAD environment: single participants working by themselves and paired participants working together. In general, our results show that designers working in the paired workflow exhibited more emotion compared to designers who worked alone. A frequency analysis was performed by linking occurrences of each emotional response to their antecedent activities, revealing that user emotions were predictable to some degree by specific antecedent activities of CAD work. We concluded that activities happening in the graphics area were the most frequent antecedent events of emotions for single-users, while for paired participants, activities in the chat section and feature menu were the most frequent antecedent events for joy and fear, respectively. Finally, logistic regression was applied for each combination of event and emotion for each participant in order to further investigate the relationships between the user activities and emotions, and meta-regression was used to aggregate the regression results for the two different working styles. In particular, for single-users, activities in the model tree were found to be positively correlated to joy and negatively correlated to disgust, and navigating the feature menu increased the likelihood of contempt. For participants in pairs, communicating with CAD partner and receiving communications from partner was associated with joy, navigating the feature menu was associated with sadness, anger and disgust were associated with partner’s action in the model tree, and contempt corresponded to the designer’s own activities in the model tree area. The approach and conclusions presented in this paper allow us to better understand designer emotions in fully synchronous CAD, which leads to insight related to designer satisfaction, creativity, performance and other outcomes valued by engineering designers in a virtual collaborative environment.