The Hard Problem of Automation


24/7 is synonymously used for something that is always on, so is 360-degree view for an all-encompassing perspective, and 365 to indicate that service is open every day, although in the last case I wonder what happens on February 29 during leap years? And then there is this question of whether 1900 was a leap year at all – but that was way before digital times. If you Google details of what is called the Gregorian calendar you will come upon an interesting fact that the October revolution in Russia actually happened in November!

In a prior article on Hyper Automation, we had laid the basis for enabling connected systems and products using Make, Manage, Move, Market and Maintain.  In this 24th article we will pitch our case on solving the hard issue of automation. Herewith we present our perspective on how we plan to build an all-encompassing ecosystem consisting of Digital Threads and Digital Twins to enable the mélange of the physical and digital realms by automating the new and helping create unique smart and connected products and services.


Automation is the hard merger of technology, engineering and digital, and the people, processes and platforms needed to enable and manage them. It’s not just about seeing robots in action as they manufacture stuff. Laying out the process right from its inception to smart manufacturing and coordinating it with procurement and supply chain logistics, maintenance, support, and service, and what happens to the product after it leaves the assembly line is a complex problem. Parallel activities of marketing, sales and customer support also needs to be considered with all the safety and certification standards that need to be adhered to both inside the domain of the enterprise and outside.

The challenges of automation can be summarized in the following:

  • Cost: Automation can be expensive to implement.
  • Skills: Automation can require specialist skills, which may be in short supply.
  • Integration: Automation can be difficult to integrate into existing systems.
  • Safety: Automation can pose a risk to safety if not implemented correctly.


In a prior article on our future agenda, we had discussed the progression of the industrial revolution using the following diagram:


Therein, we had discussed our progression staring in the 1800s from an agrarian society to migrating to the urban metropolises of today. We started by harnessing energy using steam and electricity to build products, used it to for mass manufacturing via assembly lines, just-in-time Kanban processes to optimize production to where we are current with digitized automation. The next iteration will consist of mass customization and more coordinated human-machine interactions that will be the basis for Industry and Services 5.0.

At every stage of our industrialization, we added more and different processes to automate the production process thus increasing its complexity so that today we need specialized machines and high-end computers to manage the systems. The plethora of data generated by these processes now needs to be managed so that it can be used to improve the performance of the systems. This is where data science comes in. Data science is the process of managing and extracting value from data. Data science is the application of mathematics, statistics, and computer science to data so that insights can be gleaned, and decisions made.


Since buying equipment to do manufacturing is capital intensive, several strategies have evolved to account for managing automation.

  1. In what are called brown-field initiatives, existing equipment is retrofitted with smart sensors coupled with data gathering devices to enable protocols like predictive maintenance and quality control measures to reduce time and optimize operations.
  2. A green-field approach is sometime desirable by starting afresh by building a platform from ground up. An example of this green-field development in the recent past is how Tesla Motors, a new entrant in the auto industry who after only 10+ years in the domain has become a trillion-dollar company, wealthier than all of the other automotive companies combined. Albeit not perfect – they still have a lot of issues when it comes to service and support – they initially created a master plan and have followed it for two versions to stick to their purpose of building compelling next products following a theme of “the machine to build the machine” as how Elon Musk “eloquently” puts it.
  3. For our forays into Industry and Services 5.0, we will follow what we call the blue-sky approach wherein there will be a tight coupling between the process of creating and the products and services that issue. As such, all of our solutions will conform to the same DNA so that all aspects of make, manage, move, market. and maintain are coordinated.

Future-proofing designs for all of the above endeavors entails creating blueprints that move beyond current 2D and simplistic 3D animations, to constructs that morph – dynamically change, evolve, and articulate with the conditions and the environment, and are embedded with intelligence that will learn, consider, and evolve.

The diagram below shows the top Industrial IoT based use cases that would be the basis for themed automation in production environments.



𝐃𝐢𝐠𝐢𝐭𝐢𝐳𝐞/𝐄𝐱𝐭𝐫𝐚𝐜𝐭/𝐓𝐫𝐚𝐧𝐬𝐟𝐨𝐫𝐦 – Digitize, extract, transform raw data for analysis, sharing, archiving, or distribution

  • 𝐀𝐧𝐚𝐥𝐲𝐳𝐞/𝐃𝐞𝐭𝐞𝐜𝐭/𝐃𝐢𝐚𝐠𝐧𝐨𝐬𝐞 – Analyze the data – typically to identify anomalies
  • 𝐎𝐩𝐭𝐢𝐦𝐢𝐳𝐞 – Tune a set of parameters to find an optimum point/output (often done through reinforcement learning)
  • 𝐆𝐞𝐧𝐞𝐫𝐚𝐭𝐞 – Given a set of parameters, generate acceptable solutions (often done through the use of generative adversarial network (GAN))
  • 𝐏𝐫𝐞𝐬𝐜𝐫𝐢𝐛𝐞- Given a set of events/data points, prescribe the next course of action
  • 𝐏𝐫𝐞𝐝𝐢𝐜𝐭 – Use historical data to predict the likelihood of future events
  • 𝐃𝐨/𝐀𝐜𝐭Identify an appropriate action given a condition

As products become more connected, we need to mature as platforms to offer new services to customers and partner relationships. We need to be able to manage massive amounts of data from various sources, components, and devices seamlessly through the digital thread to realize the connected value stream through the product and service continuum. Architecting today’s Enterprise involves integrating core functionalities between the customer and the business using cloud provisioning, intelligent information and connected services so that activities and processes can be optimized and harmonized.

The plethora of Customer Data Platforms (CDP) tools available today and MACH principles – Microservices, API-first, Cloud-native, and Headless, have evolved considerably to make the IT/OT divide less daunting but there is need to improve further.

Our Perspective

We believe that the following three key areas will be important in the future as we continue to strive to achieve the ultimate goal of a fully connected product and service continuum:

  1. Platform Integration and Management: We need to be able to manage massive amounts of data from various sources, components, and devices seamlessly through the digital thread to realize the connected value stream through the product and service continuum.
  2. Intelligent Data Analytics: We need to be able to make sense of all the data that is being generated in order to make better decisions about what needs to be done with the product or service.
  3. Connected Services and Applications: We need to be able to provide new services and applications to our customers in order to improve their experience and get more value out of our platform.

The diagram below shows the different system streams that are the key value drivers in a new enterprise to attain maturity to manage their processes. In a typical implementation, we approach maturity by looking at 3 horizons for growth based on our strategy – Groundwork, Battlefield and Breakthrough.


Our reference architecture, the Digital TwineTM enables the composition of these new architectures that are needed for more real-time, secure, and frictionless interactions by coordinating all the streams – upstream, midstream, and downstream, in a combination of Digital Threads and Digital Twins that account for all integrations between the different processes in an enterprise. The Diagram below depicts the interaction model for the Digital Twine that harmonizes processes between up-stream, mid-stream, and down-stream activities in a typical production setup.

This will be further enabled using our other three tenets: The ManthanTM Design Philosophy for Innovation, Tendril ConnectorTM for interactions between systems, and the TAU Codex OrchestratorTM for multi-modal actionable intelligence.

The diagram below shows the composition of the Digital Twine reference architecture that corresponds with the different enterprise systems to create a holistic view of all processes and their interactions. It is based on the four themes: Design, Build, Evolve and Harmonize to collaborate between congruent streams of enterprise activity to enable and integrate consistency to optimize processes in an enterprise both within its boundaries as well as when connecting with partners, customers, and third-party services.

The above composition perfectly represents three phases of industrial automation in that as we move from left to right:

  1. We ideate and iterate on the possibilities in the product development phase up-stream
  2. We linearize on getting the product built in the mid-stream phase of manufacturing
  3. We try and test out markets and gather customer feedback to market and sell down-stream

We have utilized the above reference architecture to blueprint more than 15 use cases to define brown-field, green-field and blue-sky initiatives as shown in the diagram below. These use-cases are derived from more than 50+ projects that we have done for mostly Fortune 500 clients in the past 20+ years, and our intention as Numorpho Cybernetic Systems (NUMO) to help and implement new brown, green and blue initiatives.

Industrial Use Cases

The Premise: Connected digital manufacturing enterprises are collecting more and more data related to stages of the product life cycle downstream from product design. Linking these data back to choices made during the design process is a large opportunity for improvement as 70-80% of costs are determined there. The Industry is in-need of artificial intelligence-powered tools that can ingest historical data and correlate it to new or revised designs in order to socialize expertise and improve life cycle outcomes.


As seen above most of the use cases fir into the top two quadrants of being Strategically important and providing a quick payoff for investments. In this section, we discuss three pertinent use case blueprints for automation in the new.


Use Case 4 is our quintessential use case for automating actionable intelligence in a factory floor. Our basis for this is in three parts – Gear Up, Ascension and Genesis that will provide the impetus for other modalities to interface, integrate and interact with to provide a strong basis for Robotic Process Automation (RPAs) in the factory floor and a coordinated ecosystem for our smart and connected products and services. The diagram below shows the blueprint for implementing it.

More details on this particular use case will be provided in a separate whitepaper.


Innovating new products requires a holistic approach to design that includes feedback from mid-stream processes (production) and down-stream use (customer expectations). Since 70-80% of the life-cycle costs are determined by decisions made in the design phase, it is imperative to utilize knowledge and best-practices to influence the upstream activities of design and product development.

In this use case (Use Case 8), we present an approach to use a guidance system created using trained AI and statistical methods to better inform designers on mid-stream and down-stream implications. This is based on historical data used in conjunction with simulation techniques to provide the basis for next generation products and services. Previously, this was achieved by doing Parametric modeling based on Statistical and Regression analysis using benchmark, test, and field data to create new variations of products prior to prototyping. These were termed Expert Systems and were the bane of the existence of Knowledge Management frameworks in engineering centric activities. In this iteration a more robust approach using sensor data from edge servers from the manufacturing floor based on the utilization of IoT.

In the diagram above the two tracks for Adaptive Engineering and Data Management will be used to set the functionalities needed for a design advisor that will utilize cloud provisioning and API components), dashboards to do intelligent analysis and a cognitive toolkit to model ML training. In addition, we will use the Engineering software stack from different Engineering and Manufacturing system providers to create a robust modular platform for all engineering-based data.

The basis behind the Design Advisor would be to utilize:

  • Cold Data – Simulation results from CAE analysis
  • Hot Data – streaming data from the manufacturing systems that have been aggregated and processed – statistically or to train neural networks

to have an alternative to real experts to provide meaningful insights to the Design process to reduce churn and rework and coordinate upstream activities with midstream and downstream processes. Taking the analogy of real notebooks, the goal of the project from NUMOs side is to provide explicit “chapters” with sections on how to do definite design tasks that use data gathered from manufacturing processes (IoT/sensor – what we call hot data) and combine them with cold/black box simulation runs to better guide the product development intent.

The goal of this use case is to ascertain several key benefits from a Design perspective:

  • Guidance for better designs utilizing a data driven intelligent approach
  • Utilize data created by IoT in smart manufacturing processes to better inform design
  • Use standards (STEP, XML, etc.) transfer data between systems so that it is interoperable and cohesive
  • Extensive use of Metadata to dimension the data based on type and category of information
  • Use of AR/VR technology to create dynamic blueprints of process flow that would superimpose 3D rendering of Digital Twins to simulate conditions during manufacturing
  • Investigate blockchain technologies to maintain the immutability of records so that data is not duplicated and susceptible to change

Our First PaaS

Merging Additive Manufacturing, Ecommerce and understanding the implications of Supply Chain, we are in the process of building our first Platform-as-a-Service (PaaS) offering – an Omni-channel Connected and Adaptive Marketplace (OCCAM).


This platform will appropriately connect innovators and customers with providers having industrial design expertise, 3D printer OEMS and contract manufacturing facilities to enable production in small or large scale. It will have built in intelligence to initiate the appropriate handshake and coordination of quoting, contract negotiations and material procurement and will provide the basis for the next generation flexible ERP, CRM and SCM systems merged into one. Use cases for Contract Manufacturing, B2C2B Commerce and Fast Fashion have already been defined and we are in the process of constructing the platform using Microsoft toolsets – Azure, Dynamics, Power, Cognitive Toolkit and utilizing OpenAI for the intelligence. More details on this and how we are utilizing the services of one of our partners in the Commerce provisioning industry is available in a separate whitepaper.

The Digital Twine blueprint for this ecosystem is represented below:


Following with the metaphor “of building the machine that builds the machine”, here is our concept for an end product that fits with our methodology of evolving a solution that has the same DNA as our ecosystem.

Premise: The global e-bike market size is projected to grow to USD 70.0 billion by 2027 from USD 41.1 billion in 2020, at a CAGR of 7.9%. Government support and initiatives to increase the sale of e-Bike would drive the global e-Bike industry.

Our first product in the micro-mobility space (Personalized Active Transportation) – a smart e-bike, a pedelec for urbanists. The diagram below showcases the manifesto of the bike – the “should” conditions that match up with its feature set that will be developed in an MVP progression.

In this next diagram, we have animated the transitional, collapsible, and foldable aspects of the bike to showcase its uniqueness and adaptability to various modes that the bike would be subject to during its use.

A separate document will detail its design, engineering, and manufacturing considerations. Adjacent products that will be developed in conjunction – foldable helmets, articulating handlebar, width-changing seats and telescopic fender will also be detailed in this document.

Subsequent articles will elaborate on our perspective on Automation by evolving adjacencies that need to be fully explored:

  • In Engaging Disengagement, we will review what should happen in conjunction with automation within our development process and for our connected products, thus re-affirming our Outside-Inside theme. This will talk about rules and regulations and what we plan to do to take control of processes when needed to intervene so that things do not get out of hand.
  • Balanced Supply Chain will coordinate our basis for Move in our Make-Manage-Move-Market-Maintain philosophy. Here we will merge Procurement, Just-in-Time, Zero Based, Last Mile and other logistics considerations to arrive at a holistic perspective for our implementation.
  • Insightful Analytics will delve into our basis for Actionable Intelligence wherein we utilize our forays in AI/ML and Cybernetics to define Existential Intelligence. Cybernetics is a wide-ranging field concerned with regulatory and purposive systems. This will prescribe, predict, and portray outcome based on multi-modal compositions based on our Tau Codex Orchestrator to enable appropriate and contextual feedback.


We stand on the brink of a technological revolution that will fundamentally alter the way we live, work, and relate to one another. In its scale, scope, and complexity, the transformation will be unlike anything humankind has experienced before. We do not yet know just how it will unfold, but one thing is clear: the response to it must be integrated and comprehensive, involving all stakeholders of the global polity, from the public and private sectors to academia and civil society. This Fourth Industrial Revolution and its successor, the Fifth are being driven by a staggering range of new technologies that are blurring the boundaries between people, the internet, and the physical world. It’s a convergence of the digital, physical, and biological spheres. These spheres are coming together in ways that are fundamentally altering the way we live, learn, work, and socialize.

The Fifth Industrial Revolution has the potential to raise human living standards and address many of the world’s pressing challenges, including poverty, inequality, climate change, and pandemics. It also poses significant risks, including job losses, cyber attacks, and the weaponization of artificial intelligence. We need to develop a comprehensive understanding of these risks and take steps to mitigate them. We also need to ensure that the opportunities created by the Fifth Industrial Revolution are equitably distributed and that the benefits are shared by all.

To realize the full potential of these revolutions, we need to act now. Numorpho Cybernetic Systems, with our tenets for themed automation plans to be in the forefront of this change to elevate our status as a civilization and define our progression as humanity.

NI+IN UCHIL Founder, CEO & Technical Evangelist

, ,

3 responses to “The Hard Problem of Automation”

Leave a Reply

%d bloggers like this: