Data-Driven Midsole: Performance-Oriented Midsole Design Using Computational Multi-Objective Optimization

Runjia Tian*,
Yujie Wang*,
and Onur Yüce Gün, PhD.,
with supports from New Balance R&D Innovation Studio and MIT.

Full Paper accepted in the **ACADIA 2020** (Association for Computer Aided Design in Architecture) Conference: Distributed Proximities
DOI:10.52842/conf.acadia.2020.2.188.
Presented in **Paper Session 10: Environmental Responsiveness and Performance Optimization**
link: http://bit.ly/datamidsole (recording password: m@rshm@ll0w)

Keywords

Multi-Objective Optimization, Machine Learning, Data-Driven Design, Lattice System, Performance Simulation

Abstract

With the advancement of additive manufacturing, computational approaches are gaining popularity in midsole design. We develop an experimental understanding of midsole as a field and develop designs that are informed by running data. We streamline two data types, namely underfoot pressure, and surface deformation to generate designs. Unlike the common approach in which certain types of lattices get distributed across the midsole according to an averaged pressure data, we use ARAMIS data, reflecting the distinct surface deformation characteristics, as our primary design driver. We analyze both pressure and deformation data temporally, and temporal data patterns help us generate and explore a design space within which we search for optimal designs. First, we define multiple zones across the midsole space using ARAMIS data clustering. Then we develop ways to blend and distribute auxetic and iso-surface lattices across the midsole. We hybridize these two structures and blend data-determined zones to enhance visual continuity while applying FEA simulations to ensure structural integrity. This multi-objective optimization approach helps enhance the midsole's structural performance and visual coherence while introducing a novel approach to 3D printed footwear design.

Contributions

In this project, we use ARAMIS data, reflecting the distinct surface deformation characteristics as our primary design driver (instead of averaged pressure data). We proposed a data-driven workflow for midsole design. Our approach demonstrated how to use Deep Temporal Clustering for analysis of the sequential pattern of running experience and segmentation of midsole areas. We developed a multi-objective computational system for midsole design optimization. And we employed computational simulations to make informed decisions to iterate our designs. We developed a method for blending lattice units with different topological configurations.

While computational optimization tools become central in decision-making processes, human interventions become even more crucial for setting and running such systems in efficient and intuitive ways. Re-witnessing the challenges of blending visual and performative design goals in this project. Subjective decision making, such as visual choices cannot easily be quantified. Yet still, such decisions remain significant for design exploration.

Context

Additive Manufacturing in Footwear Industry
Additive manufacturing, also known as 3D printing, has been broadly applied to product prototyping and fabrication. With the advancement of data science and volumetric modeling toolkits, there are opportunities in the footwear industry to employ computationally generated lattice systems in place of conventional components and materials in order to enhance performance with new design possibilities.
Computational Midsole Design
Located between the upper and the outsole, the midsole performs as the main cushioning, stability, and pronation control component. Midsole materials, sub-components, and overall composition determine the quality of the running experience, and thus many running shoes end up featuring a broad range of midsole components and materials. Data-driven computational design offers opportunity for personalization.

Expand performance metrics

Pressure data dominance (value to material distribution)
There have been many explorations of underfoot pressure data as a primary form factor. In recent years, New Balance partnered with Nervous System to develop 3D-printed midsoles for performance running shoes in which the 3D-printed lattice geometry was determined by pressure data.
Expand performance metrics (Temporal pattern + ARAMIS)
The project presented below challenged this conventional footwear data usage by proposing a new design agenda that treats ARAMIS data as the primary factor and pressure data as the secondary factor. We see an opportunity to add onto this data-driven approach through taking the temporal implications of running into consideration: various parts of the midsole come into effect during various stages of human movement, and this can be used to inform lattice distribution and structural units design.