Timeline
September 2021 - January 2022

The Problem
In daily mine operations, operators have to go and check every piece of mine equipment to determine it’s current status. This is a time consuming procedure and leaves room for human error (forgetting to check a piece of equipment, mixing up equipment, etc,). At the time there was no platform available to view all the current data related to the mine operation equipment or for planning the mine operation for the weeks ahead.

Mission
To create a web application for the mine operations team that enabled them to view the current status of mine operation equipment, check historical and forecasted mine operation plans, see daily, weekly, and yearly analytical trends, as well as be able to suggest states for mine equipment to aid in the mining operation.

Front End & API
The UI for the project was built using React, VisX, and mob-x-state-tree. The API (C# and AWS infrastructure) aggregated the data from the back end and delivered it to the front end of the application to enable itd vast functionality, which included:
1) A comprehensive chart visualization that
- Allowed users to drill down to specific metrics for specific weeks of mine operations.
- View trends in data for the current season as well as the historical season, and compare them.
2) A dashboard with important mine operation metrics for daily and weekly operations.
3) A weekly planning component which displayed the mine equipment configuration for the week based on the machine learning model combined with additional human input.
4 A daily planning component which gave users the ability to check the equipment configurations for the current day and highlight any discrepancies between the mine equipments suggested/intended status and the actual status of the equipment.
5 The ability to manually change the suggested status of mine operation equipment.

Back End
The backend of the application used historical and IoT data from the mine operation to train machine learning models with the purpose of forecasting optimized mine operations. This data is communicated to users through the applicaiton where they could make changes to the suggestions made by the machine learning model, leading to optimal mine performance.
 
Project Results
The mine operation was able to track data trends, optimize mine performance, and monitor mine equipment all from one platform.

Tech
This was a full stack project that utilized AWS, AWS CDK, TypeScript, C#, React, VisX, Docker, Python, and CircleCI.