︎

Adam Huth

Designer Across Scales


Digital Product Design
UX/UI | Graphics
  1. Let’s Eat!
  2. Flipout
  3. Fungi on Foot
  4. Hondezvous
  5. Exclusion in Design
  6. Social Media Profile Redesign
  7. Digital Art & Other Fun Stuff

Experimental Design
Provocation | Speculation
  1. Design for Uncomfortable Consumption
  2. Fruit Face
  3. Vitals
  4. STAR & ARTS
  5. MUTO-21
  6. Fungi on Foot
  7. Smiles by Mail

Industrial Design
3D Modeling | Furniture
  1. Alka-Seltzer Academy
  2. Table I
  3. Table II
  4. Bar
  5. Bed 
  6. Electronics Mounts for Rocket

Art
Interventions | Sculpture
  1. Picnic
  2. Projection
  3. Handing Off: Art in Pandemic Times (A Virtual Mail Art Exhibition) 

About Me
  1. A designer across scales with experience in digital product design, speculative design, entrepreneurship, and art, with a unifying passion of creating unique experiences that spark joy, provoke questions, and make the world a more accessible place.
Mark

4. Hondezvous




2021
In collaboration with Nikhil Dutt, Xuerui Song, Charlie Duarte, Isabel Zavian, Ebru Odok, and Honda’s 99P Labs
Data Science; Data Visualization; UI
https://medium.com/data-x-hondezvous
           
Our team was selected by 99P Labs, an innovation lab supported by Honda and The Ohio State University, to develop machine learning methods that can predict a vehicle’s future location and dwell time (how long a car will remain stationary for) enabling a host of services: make a car a delivery location, get services on the move, and enter a personal car into the gig economy to earn passive income.

Markov Chain Model; a simplified visual representation of our model for predicting a vehicle’s next location; the model deciphers patterns between a driver’s first location and their destination with observations, such as the time of day

K-Nearest Neighbors (KNN) Model; the model uses location to train a predictive model (XGBOOST, SVM) that can predict dwell time; clustering predictions and location to gain contextual information of predictions, identifying locations of high and low accuracy

Over the course of five months, we developed two models to predict a vehicle’s next location and associated dwell time. To determine the destination of a given car based on an initial start position in time, we developed a Markov Model. We then creatively combined DBScan, K-NN, and XGboost algorithms to achieve accurate dwell time forecasts. Once the two models were built, they were consolidated into an efficient UI for service providers. Our model achieved an accuracy of nearly 80% on our test set.
Minimum viable user interface for ease of interaction with the models; selecting a particular vehicle displays the next predicted location and dwell time; built with Streamlit.io

Spread of presentation slides, used to pitch the value of the Hondezvous system to Honda’s 99P Labs; a large emphasis was placed on storytelling and showcasing potential applications
Mark