DATA SCIENCE

Domain Experts in Upstream Oil and Gas.

Finding patterns and meaning in data is not a new Science. What is new is the volume of data available and the capability of the tools (software) to analyse that data. 

Steve Adams has been working with variants of these tools for over 20 years and understands when they are most useful, and how to check the veracity of derived relationships and predictions.

Be aware that Machine Learning results can sometimes be very misleading! "Sense"  checking is necessary.

Finding Insights

Looking at Large Datasets in many different ways allows relationships between the data to be identified. Some of these relationships may be obvious, but others are certainly not. Finding dependencies between different data leads to insights in what is happening within your dataset. TPL will use machine learning to help find dependencies, but we find that the best insights come with an understanding of why the identified dependencies exist.

Identifying Opportunties

Insights alone may improve understanding, but these insights must be leveraged into decision-making to have an impact on performance. TPL has the upstream Domain Expertise to translate Insights into business Opportunities.

We can also suggest optimal strategies to implement identified models into upstream technical processes.

Scientific Method

Systematic observation, measurement, and experiment followed by the formulation of hypotheses, their testing and refinement.

This work is not statistics, nor computing, although knowledge and expertise in both these disciplines aids any investigation. 

Tools

Data Science as a discipline is technology agnostic i.e. the technology used is not "Data Science". The technology used is only a set of tools to assist in carrying out the scientific investigation.

TPL uses tools that are appropriate for each task. Some of these tools are: MS-Excel, MS-PowerBI, OpenRefine, R-Project, Orange3, Python, SageMath, Scikit-Learn.