- Hello, I'm very excited to have the opportunity to share a few practical learnings on data science for oil and gas industry. Data science is becoming one of the hottest fields in industrial sector given its enormous potential of helping businesses increase efficiency and safety, reduce operational costs and discover new opportunities for the use of data. The objective of this video lecture is to introduce a few key concepts about data science and how you can apply in your field. So let's get it started. Data science is a field that combines mathematics, statistics and computer science. A data scientist is not necessarily a mathematician nor a software engineer. But they have a strong understanding of varied concepts covered by these roles. In the oil and gas industry, a data scientist is usually expected to have a good grasp of geoscience or engineering. So they can derive requirements and solutions that align with a business. As computer power has increased and storage has become cheaper, it has made it possible for mathematicians, statisticians, and computer scientists to implement new models that rely heavily on processing massive amounts of data to generate actionable results. This has reignite a strong interest in research and development around data mining, machine learning, computer vision, artificial intelligence and other technologies. The results of these developments have been staggering. And avalanching a strong era of smart data driven algorithms and learning machines. These are exciting time for the oil and gas industry since advent of these technologies will enable the discovery of new opportunities generating more efficient workflows, increase safety and significantly reduce operational cost. The following diagram illustrates a snapshot of what is typical of serving the oil and gas industry. Mathematicians and statisticians are typically reserved for research and development of medical and pluralistic models to support engineers and geoscientists. At the same time, computer scientists and IT specialists are usually hired to provide technical support and building house software solutions for engineers and geoscientists. Lastly, engineers and geoscientists might have a good grasp of computational scientific topics. And defining this professional face the task of setting the basis for starting a data and analytics team. Given the wide disciplinary characteristics, a data scientist will need to cover and the complexities of the oil and gas industry, it is quite challenging to find an individual that can fill all these roles. For instance, some data scientists may be stronger with math and statistics, but weaker at coding. Others may be the domain experts, but they lack a strong mathematical background. Rather than looking for unicorns, it is more efficient to gather teams of specialists with complementary background that can work together to solve a wide range of challenging problems. A data scientist is primarily focused on developing first prototypes by using their achieving approaches. These prototypes are personalized solutions that cannot be easily satisfied by an off-the-shelf solutions. As part of execution, data scientists will solve problems by using an arsenal of algorithms in computational science, uncertainty quantification, optimization, ah-cah-ting machine learning and AI. However, if an algorithm doesn't exist, they will create a new one. Despite the versatility of a data scientist, it is important to know that they cannot deliver on their own a timely product that can scale for the business needs. Once a data scientist has solidate a prototype software engineers can pick it up where they left off. Software engineers are capable of architecting and implementing a scalable platform, With the help of a data scientist, they can transform the prototype to production-ready code that is optimized for GPUs and distributed cloud computing. They can also deploy the code into a big data platform or integrate it into IoT devices such an sensors, drones, or robots. Unfortunately, the potential of this collaboration is usually overlooked by many industries, including oil and gas. On the bottom of this diagram you will find two traditional supporting roles. Namely data analysts and data engineers. A data analyst can be considered a more junior data scientist that primarily focus on data exploration and cleansing, building basic models um-buh-lee lay then. On the other hand, data engineers can be considered junior software engineers that have the ability to convert the prototypes from data scientists or data analysts into IoT-oriented code that can be still tested and integrated into the production platform.