Findings and Gaps in Fit-Tech Research.
As part of a PhD course I took while at ID, I conducted deep research into the current landscape of knowledge at the intersection of what “fit” means, customization, supporting technologies, and inclusivity.
The outcome of this research was a paper of structured arguments that I titled, “Computational Measurement on Modes-of-Fit for Inclusive Custom-Fit Fabrication and Co-Design Services.” But for the sake of this blog, I would like to share with you the synthesized notes and findings.
Earlier in my research, it became apparent that to research how to support “modes of fit”, the mental model of “customization” seemed most appropriate to follow, as the goal is to try to understand and design for everyone’s unique expectations and requirements around what “fit” means to them. By researching the technologies that currently do, or potentially could enable customization, the following domains (themes) emerged. Within these domains, more specific research was conducted into how “modes of fit” and inclusivity is (or is not) being supported.
These domains include data collection, data analysis, data visualization, co-design, and data actualization.
These domains make up the potential process for the customization of individualized “modes of fit”, seen below:
Data collection is how the designed process for gathering the users interpretation of fit (i.e., 3d scanning and photogrammetry for collecting dimensions of physical fit, and a questionnaire to gather information to make identity and societal fit per the users desires). Data analysis is the design system processing that data and relaying it into a digital twin. With a digital twin developed, designs can be generated. Data visualization allows users a chance to see the potential final designs. And should they want to make any corrections to the design parameters (including the data collected regarding their modes of fit), co-designing gives uses the ability to partake and interact with the design so that it truly aligns to their sensibilities. Finally, certain technologies (both digital and physical) enable the design to be actualized into a usable product for the user.
Most of these domains are being supported by new technologies addressing “fit”. I did deeper research into the intersections and new knowledge being conducted around these fit-technologies, and how they were in alignment of “modes of fit” (or not), and where the general body of new knowledge around this domain was inclusive to the spectrum of fit.
Below are high-level findings and insights from the research and paper written.
1. Digital Twins support predictive modeling for access of custom-fit transitional designs
We need to collect data [data collection phase] to create digital twins, so that designs can emerge from this data to express the individuals data in context to the broader data.
Example: Stitch Fix
2. Data Methodology for Fit Contextualization
Quantitative and qualitative data needs to be used in tandem to understand and evolve Design’s approach to “fit”, because the user's perception of fit contextualizes their evaluation of the physical fit.
Example: Stitch Fix
3. Data Methodology and Custom-fit for the Inclusion of “Outlier” / Marginalized Users
Custom-fit services can provide a more inclusive alternative to traditional fitting services, because it utilizes quantitative and qualitative data differently than the methods and practices designers traditionally use. The field of Human-Centered Design has traditionally been known for its reliance on qualitative data, collected through methods such as ethnographic research and interviews, and their ability to extract meaningful insights from said data, including the parameters of user interaction with the product of service. It also lends itself to findings related to different modes of fit, as Graham Pullin explains, “user-centered design is broadening from traditional physical ergonomics into cultural diversity and individual identity.”
4. Current solution for fit assistance maintain biases of current fitting processes
Current market offerings and innovations in the industry attempt to solve the problem of fit through digital fitting platforms and best-fit recommenders (Sizer, Fit Analytics, Fit Me, Stitch Fix, etc). However, digital fitting platforms and fit recommenders preserve algorithmic biases in biometric data, because they only collect data that fits a “normal” ideal and representation of a body, not collecting any data from user’s with body’s that deviate from this norm.
5. Anatomical data measuring for increased user experience and expectation
Companies should invest in technologies that provide nuanced ways of measuring the human body, because it could provide a more accurate fit for the wearer, therefore elevating the wearing experience for the customer and increasing sales for the company.
Example: Volumental + Nervous System New Balance Midsoles
6. Designing inclusive systems using anatomical data capturing technologies
Though anatomical data-collection could be revolutionary to people traditionally marginalized by the mainstream fashion industry by exclusion of their size and abilities, much research still needs to be done on these data collection technologies used, because most of them available today are still exclusionary to those with disabilities.
7. Sensitivity for interface and interaction involvement [virtual try-ons]
Though custom-fit processes could be revolutionary to people traditionally marginalized by the mainstream fashion industry by creating objects that are made specifically for the individual and their body, much research still needs to be done on the interface and interaction experience needed to create these custom-fit designs, because access and user involvement is not only equitable but true to the nature of customization. A common affordance to current fitting solutions, albeit a best-fit size recommendation or custom-fit service is the need for the user to measure their body, and input this data for the means of transaction for the fitting process. Oftentimes, this data is then visualized in a virtual try-on through a digital avatar or virtual reality. Yet sensitivity should strongly be considered here, as these processes require the user to overcome privacy concerns, and emotional and psychological barriers they may have with regards to their body, such as body image disturbances. With this, gender, cultural influences, age and other influences may impact the experiences a user has with a type of interface and interaction with a fitting service.
Example: Userway
8. Contribution to Anthropometrics
An open source and collective dataset on body measurements and related shape, sizing, and movement needs to be developed, as it would not only lead to novel and nuanced understanding of how bodies change over time, through movement, by gender, race, age, disability, region and more, but to the creation of better fitting products and services given those factors.
9. Accessibility and democratization of the design process
Generative design can help to democratize design, because it has the potential to give users the access, agency and tools to design what they want with consideration for their individualized needs.
10. Digital fabrication advancements support the development of custom-fit designs in medical applications
Medical wearables could benefit from the affordances of customization, because it would not only increase efficiency in the production process and accuracy in physical fit, but could provide wearers the opportunity to instill their identity within the design.
