Do you have enough of the right data? Can you tell?
Yes, you probably do
Current data historians contain plenty of high quality data. Log files, SCADA systems and sensors existed long before the term “Industrial IOT” became popular.
However, it’s in silos
Non-collected and non-collated data exists in oil and gas upstream workflows as it works in silos. Within these silos a number of vendor proprietary data stores exist based on the oilfield services company contracted to perform part of the oil and gas upstream workflow.
It can’t be used “as-is”
The time required to deal with data wrangling, ETL, preprocessing and even anonymization of data without any in-house tools, may take up to 80% of the overall analytics project effort. For analytics teams not familiar with oil and gas will take even longer.
Having data stores is not enough
Data stores are not designed for computationally heavy data analytics work. Instead, they are optimized for storage and basic 2-D information lookup for human consumption.
How can AlphaX help?
Data Engineering through CDAT©
Determining whether the data you have is viable and can be used to look at a particular problem is an arduous task. Without the right infrastructure and processes to do so, simply iterating through the data engineering steps can waste weeks of time. Our Customer Data Acceptance Test (CDAT©) process can take your core dump of data and do a first pass at understanding it and educating you on the possibilities of the data. This methodology can go one step further and give you a customized report of all the statistics you need on the data and visualizations to understand potential correlations. The specs for CDAT vary for the different analytics applications related to oil and gas.
Data Anonymization Software
All log files require unique forms of data anonymization in order to begin processing by an outside data scientist. Unfortunately, there are no available solutions based on current taxonomies tailored to quickly eradicate or mask identifiers, such as well names, field names, latitude, longitude, even service company names which are pervasive in these logs. This process can be multi-step and every piece of information removed or changed needs to be accounted for. Further, domain expertise combined with statistics may allow a data scientist to use attributes to reverse engineer to break the privacy of the dataset. It may be possible for a data scientist to figure out certain things, like region or geography based on the other information in the dataset. Finally, it may be important to transform data entirely before providing it for outside use. These issues and many others are handled in our Data Anonymization software, which is incorporated in our advanced analytics methodology (RADAM©)
Custom Data Analysis and Management Solutions
From oil and gas operators to independents to data service companies there are a never ending set of opportunities to enhance data by collecting new information, build richer data sets by incorporating third party data, and examine the overall health and utility of the data that you are currently collecting. Our experts can look at the current process you have in place and suggest better methods to improve your overall efforts toward data management, data analysis and ultimately, advanced data analytics.