02 September 2020
We chatted to Dr Sia Doshvarpassand about what its like working with artificial intelligence as a Data Scientist and Reliability Engineer
Dr Sia Doshvarpassand is the Lead Data Scientist at VROC. Sia has a unique background that includes both engineering as well as data science, and he is now applying that experience to further develop the VROC Artificial Intelligence platform as well as deliver exceptional results for our clients.
We thought it was time to sit down with Sia and get an insight into how this blend of data science, asset reliability and artificial intelligence works together, and what he sees is the future for data scientists and asset realibility engineers.
As a mechanical engineer with more than a decade working in Oil and Gas, what made you switch focus to Data Science?
It would have certainly come across contradictory and to some extent weird if ten years back you would introduce yourself like: “Hi, I am mechanical engineer, I work for an IT firm as a data science consultant and AI solution architect”. But not anymore.
The impact of industrial data has become huge, thanks to cheap availability and significant reliability of sensors, IIoT and broad and remote sensing and communications. The resource and energy industries focus on benefiting from their data has been long overdue. Sectors like retail, finance and marketing have been using AI and advanced analytics for quite sometime.
After working a decade in oil and gas downstream design and integrity departments, I began to realise, that using design codes and practice set based on statistical and mathematical modelling of certain case studies is no longer best practice. The design process has been significantly affected by reliability studies during recent years. In fact, the reliability as the key exercise during asset after-design life and during operation has had a significant impact on optimising the design process. So not only the reliability and more importantly the asset condition monitoring will increase the asset life cycle but also will optimise the design process through focusing on hotspot and risk points.
As part of a transformation journey I began a couple of years back, AI and some of its subsections such as machine vision, machine/deep learning became mandatory to my evolution and transition from an engineer to an analyst. Despite being present in academic environment for a quite some time, which generally drives the candidates’ mindset towards theoretical contributions, I never intended to use my two masters and PhD achievements in order to expand my career contributing to the mathematics behind the AI. However, it was very exciting for me to solve problems using AI as a tool.
How do you see your Engineering and Data Science experience benefiting each other?
There is a misconception out there that data scientists are able to work on any type of data and extract insights out of them. This can be true if as a data scientist you spend quite sometime understanding the physics and foundation behind the data you are working on to make sure you are not looking at the data at face. Working on industrial (sensor) data has always been fairly straightforward, as well as exciting for me according to my background. However, the impact of interaction with client’s SME and being benefited by their experience and subject matter expertise to validate AI models results was a key success factor in my multiple projects in VROC.
What attracted you to work at VROC?
After working on multiple projects as an AI solution architect and asset analyst, a mentor introduced me to VROC. As a data scientist who spends a major portion of his time on developing solutions through hard coding, VROC suddenly came across as a scalable and powerful tool which removed a significant amount of time spent on developing solutions. I internally participate in developing AI solutions while using my engineering insights and capabilities alongside VROC as a powerful tool in a client facing position, I help clients to get the full benefit of using AI for the long-term monitoring of their assets’ health.
What do you see as the future of Data Science and Artificial Intelligence?
Data science as a career will be eventually divided into two less pronounced categories. The evolution of customised and industry-specific AI tools such as VROC predictive maintenance platform will require hybrid breed of engineers and data scientist to look after AI models and client asset health, account management, client training and conducting PoCs. The second group are hybrid breed of people with DevOps, data engineering, mathematics and statistics skills contributing to optimising solution and developing off-the-shelf tools.
If you had one motto to live life by what would it be?
Life is short. Make positive impact no matter how small.
If you would like more information on how VROC's platform can help you with your asset reliability and maintenance get in-touch with our team, we can set up a demo or get you started with your own POC.
Social MediaShare this article