ENGS 89/90 Reports

Year of Graduation

2025

Project Advisor

Charles Hackett

Instructor

Solomon Diamond

Document Type

Report

Publication Date

2025

Abstract

Accurate, hyperlocal weather data is critical for effective winter weather management. However, the Dartmouth campus’s current solutions often lack the precision for optimal ice treatment. This results in excessive and uneven salting practices, leading to increased worker and student injuries, financial and laborious burdens, and environmental damage caused by runoff. To tackle these challenges, our team has developed a comprehensive system. The system integrates hyperlocal weather station data, a peer-reviewed ice prevention optimization formula, and a machine learning model to generate intuitive salt usage recommendations and precise ice formation forecasts. The deliverables include an integrated system featuring functional weather data collection and storage, an ice formation prediction model pipeline, an algorithm for ice treatment recommendations, and a framework for future weather management applications. We aim to improve ice treatment outcomes, reduce financial and environmental impacts, and provide a foundation for future weather-related innovations at Dartmouth. Throughout the project, we have followed an iterative process of planning, developing, gathering user feedback, and problem-solving, improving our product’s scientific reliability and ease of use. The future scope for this project may include using the hyperlocal weather data pipeline to monitor heat stroke risks in collaboration with the Environmental Health and Safety Department, validating the ice formation prediction model through field testing, and improving user interaction through partnership with Dartmouth Facilities Operations and Management.

Download

Restricted

Available to Dartmouth community via local IP address.

Share

COinS