Skip to content
Geetesh Veeravalli

Case 03 / ASU LightWorks · Nov 2024 — present

Tempe, Arizona

A digital twin for a million acres.

Data Scientist · data modeling and digital-twin lead

Data modeling for a proposed biomass supply system across more than a million acres of northern Arizona forest: an interactive digital twin, the suitability maps, and the delivered-cost model that prices the whole chain for the funders.

1M+acres in the proposed biomass system
$2MUSFS pilot proposal the modeling supports
18months in the proposed pilot
Elevation contours of the Flagstaff region, derived from USGS data
A USGS elevation illustration of the Flagstaff region, not a screenshot of the digital twin.

Problem

A proposed biomass supply system across more than a million acres needs a way to connect geography with logistics and delivered costs. The work supports planning for the Four Forest Restoration Initiative in northern Arizona.

Role

I lead the data modeling and digital-twin work on the 18-month, $2M USFS pilot proposal, alongside the grant budget workbook and delivered-cost model that price the system for the funders.

Approach

The twin is an interactive Leaflet.js map of the system: suitability overlays for where biomass can be sourced, coded nodes for the facilities, and corridor distances between them. The delivered-cost model turns that geography into a price per unit at the gate. In a related project with Dr. Rimjhim Aggarwal I built an optimization model for an independent biomass system operator that minimizes transport and subsidy costs across stakeholder groups.

Engineering focus

The digital twin and cost model support a proposed supply system. Suitability, facility locations, corridor distances, and delivered costs are the central planning inputs. The $2M figure describes the USFS pilot proposal, not awarded funding.

Outcome

The funders can look at the proposed system instead of reading about it: where biomass can be sourced, where the facilities would sit, how far the material travels between them, and what it costs at the gate. The same model priced the budget in the proposal, so the map and the number are one argument rather than two documents that have to agree.

What this demonstrates

I can turn a geography into something people can look at and something they can price, and hand both to the people who have to decide whether to fund it.

Leaflet.jsPythonGeospatial analysisOptimization

Next case / Floodlight, Inc.

Climate risk, priced for a portfolio.