The term "data science" has become more tech-savvy industries. But for data science specialist Pramath Parashar, it is also shaping the way mining companies operate, report, and optimize their decisions.
Parashar is a Data Science Specialist working in one of the world’s leading resource companies, where he applies data science to solve complex industrial problems. Since joining after his postgraduate studies at Kent State University, he has led the design of automation-driven solutions that deliver measurable business value in high-impact operational settings.
An example of delivering business value is the Annual Report Automation Engine, a Python-based tool developed by him that processes financial data and generates formatted tables for official use. The tool has reduced manual reporting time by 85%, leading to more than $50,000 in annual cost savings.
Apart from financial data, mining operations also generate massive volumes of unstructured data, legacy PDFs, environmental reports, and technical documents. To tackle this, Parashar built a SharePoint Sync System that automatically processes, validates, and tags over 4,800 PDF files. The automation doesn't just save time; it ensures that the metadata is accurate 98% of the time, making compliance checks and information retrieval far more efficient.
He's developed tools not just for data analysts or business managers, but also for field engineers, keeping in mind the cross-team operations. The CTD Processor App, for example, is a desktop tool that calibrates environmental sensor data (like conductivity, temperature, and depth). It is built with a PyQt interface, enabling standardization of legacy field data.
Parashar has also prioritized building scalable automation. He developed logic for batch-based operations in Power Automate to manage 4000+ SharePoint entries efficiently, avoiding system lags and crashes, something Power Automate wasn't natively designed to do.
The actions yielded results. Over 4,800 files were processed with 98% metadata accuracy, a 70% improvement in document retrieval speed across SharePoint libraries, and there was also improved standardization and efficiency in field-based data calibration through the CTD Processor App.
All these results came with their considerations. He had to tackle disjointed data by building a unified syncing system with intelligent data-matching logic. Introduced validation checks and UNIQUE ID mapping to eliminate duplicate entries and built user-friendly, GUI-based tools to allow field engineers and admin teams to use data automation without coding knowledge.
Parashar's insights go beyond his official responsibilities. He's published five papers covering topics including market modeling, automatic plant watering, distributed storage systems, visual statistical inference vs linear model testing and the CGRG algorithm in association rule mining, has four more currently under review, ranging from corrosion detection to 3D image processing. While the subjects vary, the common thread is clear: solving complex problems through practical data science.
When asked about how data science will benefit the mining sector, Parashar points to its unique blend of physical asset tracking, environmental compliance, and financial analysis. "My experience shows that data science is most impactful when it solves long-standing manual pain points, from metadata inconsistency to legacy data digitization," he says.
Looking at the current trends, he sees AI-native infrastructure and hybrid systems, where Python, Power Platform, and REST APIs coexist, defining the next frontier. "The future lies not in replacing people but in augmenting every decision-maker with reliable, real-time, structured intelligence," he emphasised.
In a sector often seen as conservative, Parashar's work shows that organizations that invest in intelligent automation, data traceability, and user-centric tools will lead the change toward sustainable, efficient, and compliant mining ecosystems.