Developing New Optimization Models for Alternative Fuel Allocations
๐Ÿงช

Developing New Optimization Models for Alternative Fuel Allocations

๐Ÿ—’๏ธ Background

The Argonne National Laboratory is a U.S. Department of Energy multidisciplinary science and engineering research center, where talented researchers work on answering science with a lens on human sustainability. Argonne frequently collaborates with universities. This project is part of a collaborative research project with the Industrial Engineering Department at Northwestern University. The project was led by our Managing Partner, Saif Bhatti.

image
image

๐Ÿ› ๏ธ Challenge

Develop an MVP connecting alternative fuel data, consumer driving data, fuel station locations and GHG emissions into county-level analysis of optimal fuel allocations to minimise overall emissions.

๐Ÿ’ก Solution

The solution involved the construction of a data pipeline taking in county-level automobile registrations, fuel GHG emission factors, as well as vehicle miles travel statistics. The first stage was to understand the current fuel allocations at a county-level.

By establishing a basis for current fuel allocations, the next stage was examining the current fuel station's location. Drawing driving-distance circles of 6 miles around each EV and E85 station, each county is assigned an EV and E85 viability score. To allow for residential EV charging, following literature-driven trends, suburban and urban counties were given a percentage boost dependant on population density.

๐Ÿ“Š Results

The developed solution successfully determines optimal fuel allocation for automobiles in the relevant states in the United States. The entire solution visualized with Streamlit, allowing users to choose driving-distances to fuel stations, as well as overall GHG emissions reduction goal. The end-to-end pipeline is written in Python and uses Continuous Integration / Continuous Deployment to integrate new changes into the existing deployment, which is hosted on an AWS machine. This project is part of research with the Industrial Engineering Department at Northwestern University, and a full paper is currently in review for publication.

Let's Connect

Interested in collaborating? Get in touch.

ยฉ๏ธ Go Fish Analytics 2021 - All Rights Reserved - Built with โค๏ธ in Austin, TX