Analytics and Digital Transformation Front and Center in Boston

My colleagues and I had the privilege of attending Tata Consultancy Services’ (TCS) Analyst Day event held in Boston on September 21, 2017. There were several interesting and informative presentations covering topics such as the concept of Business 4.0 and how TCS is deploying Digital, deep domain expertise and its deep and vast portfolio of services and solutions to provide its customers with exponential value through mass customization, leveraging ecosystems to help its customers embrace and manage risk while maximizing business outcomes.

As one of the ARC Analysts focused on upstream and midstream oil & gas it was great to learn more about how TCS is leveraging ‘cognitive automation’ through its solutions such as Ignio, an product that provides some very powerful horsepower through its self-learning capability, empowered by machine learning and artificial intelligence (AI), that can move customers from predictive maintenance to prescriptive maintenance, thereby extending the life (and availability) of an asset such as a pump or compressor and also optimizing that asset’s performance and the process for which it is being utilized. I know first-hand that TCS is successfully helping customers in Australia with pump optimization and increasing pump availability as well as, more importantly, helping to increase the customer’s gas processing operations by saving over 100-man days per year, reducing pump downtime and increasing production. Harrick Vin, Global Head of Digitate, explained that he envisions advanced analytics platforms such as Ignio as being technology being augmented by people and one that is capable of learning over time.

ARC often counsels companies in the oil & gas markets that they should focus their operations on what they do best – finding, extracting, producing, refining and/or marketing oil & gas and seek out strong technology partners such as TCS that possess not only deep domain expertise but also the ability to handle complex systems integration and optimization which is vital to maximizing the true power of a digital transformation.


  1. In my personal opinion, machine learning (ML) may not be a silver bullet for every application. There are many applications where other forms of analytics is more suitable. ML requires a learning period. For instance, to detect and distinguish between different kinds of pump failures ML has to see several failures of each kind. This will take a long time since pumps run reliably for years. Moreover, the plant suffers downtime during these failures which is disruptive and costly. Right? Instead using model-based analytics there is no learning period. All the subject matter knowledge is built into the readymade app which just need quick configuration and commissioning

    ML could work for quickly repeating events for sort learning period where these events are not costly or disruptive. Right?