In my previous post, I discussed the recovery time frame across key industrial and energy markets. In case you missed it, find it here:

Some recent realizations:

  • The industrial and energy markets are very dynamic and thereby, frequent forecasting is required to calibrate, re-tune, and respond.
  • At this juncture, due to the double whammy (COVID and oil price swings), we are not sure if it is an economic depression or a recession. The reason I say that is because we won’t know for sure until six quarters or more have passed. We know the U.S. GDP shrank by 4.8% in Q1, but will it continue its decline? While the phased opening of cities will drive growth spurts, the overall economy will be under continued stress.
  • As the new normal has sunk in for many organizations, we have observed a rough few weeks with many quarterly investor releases showing revenue contraction. Operating companies (in oil and gas upstream), on an average, have cut back on capital expenditures (by at least 47%) and operational expenditures (by at least 12%). Similar trends ensue across chemicals, petrochemicals, mining, power, etc. Few industries that are bucking this trend are medical devices, Lifesciences, and food and beverage.

The new dawn, post-COVID-19, continues to challenge the status quo. At Frost & Sullivan, we pride ourselves on our obsessive focus on growth. Internally, at times we think of ourselves as ethical growth hackers. To understand the growth opportunities, it is important to be cognizant of trends that drive development. Here, I have outlined five key trends that we believe will change the industrial and energy markets, as we know them, forever.


No alt text provided for this image

Many of the diversified industrial conglomerates have an asset-heavy transactional business, with revenue cycles either tracing the market average or in a good year growing 2-3x the market growth rate (due to new product releases, major project orders, etc.). Further, another common characteristic is their significant installed base. Leveraging digital technologies (sensing, edge, cloud, analytics), organizations are pivoting from the stock and flow business model to a platform-centric digital outcomes delivery business. It is expected that primitive products (For context, let us call them Industry 1.0 products) will be embedded with digital technologies to make them digitally advanced products (Industry 4.0 enabled products). Engineering this shift will facilitate organizations to realize the new revenue streams:

  • Digitalization advisory services
  • Networking and system integration services
  • Premium $ for digital assets (at CAPEX stage), that comes connected out of the box
  • Subscription revenue for managing asset/process performance throughout its lifecycle.
  • Enterprise apps for connected assets


No alt text provided for this image

Inspired by a large oil and gas operating company, this chart helps convey the autonomy trend. All along, our industrial and energy markets have embraced automation (executes functions pre-set by humans) as a key platform to drive productivity, safety, and profitability. But what’s beyond automation? A case in point here is: Meat plants are closing down in the U.S. due to COVID-19 exposure which hurts the downstream and upstream supply chain. Imagine a scenario, wherein the meat plant is autonomous (where machines will learn, self-optimize based on dynamic conditions) end-to-end. This will allow factory personnel to run the plant by wire/remotely. The same logic can be applied to the onshore oilfield, offshore unmanned platforms, specialty chemical processes, etc. We are a bit far from this vision, today, but some technologies exist today to help with a partial state of this future vision. Another analogy that comes to mind is for organizations to build master algorithms for factories and equipment to run in a ‘lights out’ scenario. As COVID-19 has taught us to do more with less, I expect pioneering firms to push the autonomy envelope further and faster.


Tech stacks are incredibly complex today. Heterogeneity, a multiplicity of system types, varying vintage, and communication modalities slow down data flow. Industries have to make a concerted effort to move away from latency laden ISA-95 five-layer architecture and embrace simplified three-layer architecture. The three-layer architecture will include:

  • Assets and devices: This base layer includes production assets and instruments.
  • Edge: These are open compute platforms that have immense potential to scale. Lightweight microservices can be used to build highly scalable applications and deployed in a standardized manner.
  • Meta-platforms: Cloud service providers (AWS, Azure, Google Cloud, etc.) ensure real-time orchestration between the facility and the enterprise.

I read the above case study of one of the largest energy and chemical storage companies in the world in a CIOApplications article. Very inspiring to see how 100+ years is shaking off its older past and rejuvenating it for a solid future. I believe, the edge and its ecosystem will be a space to watch out for. Low code/no-code application development using flow charts will further support the growth of this market. The profit pools in this new tech stack will also change as edge is poised to become a profit pool of choice, as it will aggregate hardware, intelligence/insights, and quality of service. Meta platforms will see monetization based on the quantum of data flowing through its cloud.


An aging workforce, growing technology complexity, the pressure to reduce OPEX, above-site exception-based monitoring, and continuous process optimization are some of the core drivers for operating companies to embrace remote services. We expect an exponential shift towards managed services/long term service/remote service agreements for critical assets. Solutions like iCenter (a solution from Baker Hughes) standout in this regard. I continue to be impressed by its remote operations capabilities and domain expertise for highly critical assets like gas turbines, reciprocating compressors, high volume pumps, etc. Even though, less than 10% of an asset fleet is critical – they take up more than 50% of the OPEX spend.

No alt text provided for this image

Another trend that will see some dusting off is a digital twin. There are two kinds of digital twins: Asset twin and process twin. There are three sub-types of asset twins:

  • Primitive digital twin: This is just a digital representation of the physical asset. The virtual model leverages design elements and represents it in a 3D view.
  • Sensorized digital twin: Sensor data feed on the physical asset can be replicated and viewed on the virtual/digital asset. This is used by maintenance personnel to understand asset behavior but would not be able to execute recommended actions, as the data feeds are uni-directional (from the physical asset à digital asset). Some level of inventory information is added onto the virtual asset.
  • Dynamic digital twin: This model the dynamic behavior (process and assets). A true reflection of the physical asset performance can be viewed on the digital asset. Further, high-fidelity models allow operators to have an immersive experience. This can be leveraged for dynamic process simulation, maintenance activities, training, virtual commissioning and asset walkthrough. If it is an equipment-centric digital twin, the model would show maintenance history, remaining useful life, forward-looking parameters, potential maintenance issues. This is a rich digital image of the physical asset with all key parameters streaming in real-time.

Process twins allow companies to simulate changes and run it, before actually exercising these changes in the actual facility. A high-fidelity digital replica of the process can even allow solution providers to provide assistance services to operators, on a real-time basis. A couple of solutions that stand out in this area are Yokogawa’s Mirror Plant, Co-pilot and Emerson’s Mimic.


We are all familiar with subtractive manufacturing (CNC, VMC’s, etc.). Two key sub-trends open up opportunities in this space:

  • Advancements in AI and heavy-duty graphical processing capability
  • Context-based/customized manufacturing (n=1) using subtractive techniques is expensive and time-consuming. A classic example is how Baker Hughes delivered drillbit-as-a-service (using AI, additive manufacturing), based on the geological pattern of the well being drilled. Using generative manufacturing technologies, Baker Hughes was able to dramatically shrink the delivery cycle for customized drill bit. Another fascinating use case is how DENSO applied the same process to develop a lightweight engine control unit. Please read it here –

Instead of starting on a clean sheet of paper, AI can help with multiple design possibilities that meet the requirements (context, condition). Further, the chosen design can be simulated under multiple conditions to achieve optimal performance.

The future is shaping up to be very interesting and fascinating! Strap in, as the journey has started on full steam, with or without you.

About Ram Ramasamy

Ram Ramasamy is an industry expert with 14+ years of manufacturing operations management and strategy consulting experience. He is passionate about creating growth opportunities for clients and tracks horizontal markets such as Digital Industrial Platforms, Industrial IoT, Analytics (Artificial Intelligence, Machine Learning), Drones, Services 2.0 and Ecosystem Partnerships.

Ram Ramasamy

Ram Ramasamy is an industry expert with 14+ years of manufacturing operations management and strategy consulting experience. He is passionate about creating growth opportunities for clients and tracks horizontal markets such as Digital Industrial Platforms, Industrial IoT, Analytics (Artificial Intelligence, Machine Learning), Drones, Services 2.0 and Ecosystem Partnerships.

Share This