Enterprise Design Engineering
Science beats intuition and experience and Engineering beats craftsmanship
In a business we have three groups of professionals: engineers, MBA-educated, and Computer Science-educated. The only group that uses science in their projects are engineers. Both IT enterprise software professionals and business management use intuition and experience to run projects. Their projects are information-centric. And there is no true complete and comprehensive information and system theory to help them.
The main goal for the enterprise design engineering is to change that. It introduces a new information and system theory that is the subject of a new field called Physics of Information. To help with the application of the enterprise design in practice, its knowledge domain introduces three frameworks and a methodology. They target the business processes, the digitization of the business, and the talent development together with career planning. The methodology is the traditional waterfall extended to projects that are information-centic.
Modern Control System Theory, Lyapunov Stability Equations, and Model-Based System Engineering Applied to Business Management
The trillion dollars management consulting industry and current business management approach are facing obsolescence and require fundamental transformation. Imagine that you are one of the 30 passengers on a plane like Boeing 307 from 1940s. Pilot skills were essential to get you from point A to point B. Would you use the same approach, relying mainly on pilot skills, to carry 200 to 300 passengers on today’s planes? Definitely not. The difference in safety and comfort of today’s air travel would not be possible without the advances in modern control system theory. Then why we should let today’s executives and management consultants run companies, many of them with tens of thousands of employees, using only their skills and experience? The constant rollercoaster of hiring and layoffs we find in many large corporations is a sign that we need to find a better way to do this job, and as in the past, the modern control system theory should be able to help. Equating a business with a complex dynamic system and being able to apply the 100-year-old very successful control system theory to manage it, should have an impact order of magnitude higher on the overall economy than any AI-based technology, regardless of how advanced that may be.
Jay Forrester, a former MIT professor, founded the field of system dynamics during 1950s. Originally this field was developed to help corporate managers to improve their understanding of business processes. The main idea was to extend the lessons learned from the system engineering field to business. System engineering has been very successful in designing, integrating, and managing complex systems in practice. This discipline has been crucial in sectors such as aerospace, defense, telecommunications, and other fields of industries.
Nevertheless, the field of system dynamics failed to play any significant role in the business. The hope that this will change in the future has been expressed by Jay Forrester himself in his 1998 paper called “Designing the Future.” He wrote that “Several decades of progress in system dynamics point to a new kind of management education. Such a future education will train a new kind of manager for the future. I anticipate future management schools devoted to enterprise design. Such business schools would train enterprise designers.” He continued by writing that “Education, in present management schools, trains operators of corporations. There is almost no attention to designing corporations.”
Why despite all these decades of progress, the field of system dynamics is no closer to the main goal of “designing the enterprise” than during 1950s? The answer is simple. When a complex system such as a plane is designed using system engineering, there is always an abstract model that is capable of describing the target as “a set of interconnected elements that interact according to defined rules or relationships, forming a whole whose behavior emerges from those interactions.” If we go one step further, in the control system theory, the target is described as something whose state evolves over time. None of the existing models used by the system dynamics qualify for this description.
On top of this, all models used by system engineering are rooted in scientific theories. There is no such scientific theory that is used in describing the dynamic of a business. Currently, the only approach used by executive teams to decide on changes is the case method promoted by Harvard School of Business since 1921.
Introducing a new information theory
Before we look into defining an abstract model for an enterprise, we need to identify first the main driver for its dynamics. In our approach we consider that the driver is information processing. This is because the entire cycle that leads to operational changes is driven by executive decisions, which are the result of processing various pieces of information.
As a consequence, to create an abstract model of an enterprise that qualifies as a system, you need a complete theory of information first. The first true scientific theory of information has been introduced by Claude Shannon in his 1948 paper called “A Mathematical Theory of Communication.”
Though the theory has been very successful in creating the digital world we are surrounded by, he acknowledged that it covers only one attribute out of two. He mentioned that while messages have also “meaning”, or semantic content, they are not covered by his theory. And processing these semantic aspects of communication are at the core of decision-making cycle.
The next step in completing the abstract model of an enterprise is to extend Shannon’s theory to cover the semantic attribute. The paper called “Extending Shannon’s Theory of Information to Viable Dynamic Systems, and the Emergence of a New Physics of Information” was proposed at the MIT System Dynamics Conference from August 2025 with the stated goal to fill this gap. An extended version of this paper can be downloaded from here.
Systems Engineering and Business Management
The use of system engineering in designing complex systems has two advantages. One of them is helping to understand the set of interconnected elements and the rules driving their relationship. The second advantage is far more important, which is the ability to apply the theory of system control to shape the behavior of a dynamic system. At its heart, control theory provides the mathematical and conceptual tools to make a system behave the way you want, despite uncertainty, disturbances, or internal complexity. And this is the main reason why the most important part of the enterprise design engineering field is the introducing of a way to manage and control the enterprise change cycles using the same equations and methods aircraft engineers use to design the airplane control platforms.
The main topic for this article is the introducing of modern control system theory to business management. The main goal for this introduction is to go beyond building an abstract model of the enterprise and to look into how we are representing the separation between the controlled and the controller mechanisms found in a business, and how they relate. If we are able to use the science of control system theory to drive changes in a business, this will completely reshape the $1.0 trillion a year management consulting industry.
Introducing Modern System Control theory and Lyapunov method for business management
The history of control system theory can be divided into two phases, the before and the after late 1950s. The age of modern control system theory was born with the launch of the first sputnik satellite in 1957. This advancement was acknowledged by all the researchers in the field, and by worldwide consensus the first conference of the International Federation of Automatic Control was scheduled in 1960 to take place in Moscow.
The new approach to control system theory was different. Instead of frequency-domain method of Bode and Nyquist, stability was approached via the Lyapunov theory. Differential equations replaced transfer functions for describing the dynamics of a system. Calculating optimal solutions was further performed using calculus of variation developed by Pontryagin instead of Wiener Hops methods.
Before we are looking into how the control theory applies to business, we need to identify first the system we need to control and the controlling mechanism. To have a better understanding of the way control works in a dynamic system, we are using the analogy with the behavior of a classical pendulum. How you split these two fundamental parts is key to the correct application of Lyapunov stability equation and Pontryagin method.
The most important task confronting the control system analyst is developing a mathematical model of the process of interest, in this case the target is the enterprise main operations.
This mathematical model is used to predict how the various devices will behave in response to various inputs. There are two modeling and analysis approaches in customary use for linear systems: the transfer-function or frequency-domain approach, and the state-space approach. The modern control theory uses the state-space approach. The notion of the state of a dynamic system is a fundamental notion in physics. The state of a dynamic system is a set of physical quantities, the specification of which completely determines the evolution of the system. The state-space is the mode of representation of a dynamic system that would be most natural to the mathematician or the physicist.
Mathematical model of enterprise operations and LTI System
In the above diagram we are looking at a few models used to represent the enterprise as a dynamic system under control. The controlled system is represented by the value creation cycle, or the main supply chain workflow. The controller mechanism follows the steps of strategic plan implementation.
State-space representation from control theory has simpler mathematical equations if the system is linear and time-invariant [LTI].
-
Linearity means that the relationship between the input x(t) and the output y(t), both being regarded as functions, is a linear mapping. If α is a constant, then the system output to input αx(t) is αy(t)
-
Time invariance means that whether we apply an input to the system now or T seconds from now, the output will be identical except for a time delay of T seconds. That is, if the output due to input x(t) is y(t), then the output due to input x(t-T) is y(t-T). Hence, the system is time invariant because the output does not depend on the particular time the input is applied.
In practice linearity of enterprise operations is relatively easy to demonstrate. The main controlling parameter is the amount of financial resources. If we increase the budget associated with the development of a new product, the outcome is likely to be a product of higher value for the business.
The time invariance is also relatively easy to demonstrate. By delaying the start of a new product development with a reasonably short period, then the outcome will not be going to change.
To apply control theory in practice, an LTI dynamic system has to be also observable and controllable. Controllability and observability are dual concepts. A system (A,B) is controllable if and only if (AT,BT) , the transposed system is observable.
-
Controllable: A system is controllable if there always exists a control input, u(t), that transfers any state of the system to any other state in finite time. Controllability asks: can input drive the internal state where we want?
-
Observable: A system is observable if we can calculate its state from the output. If we can reconstruct the entire internal state, then the system is considered fully observable. Observability asks: can output reveal the internal state uniquely?
By going back to enterprise operations, if we are able to reconstruct the entire path of the enterprise change cycle from its output, then we are calling its operations observable. This requires that executive management should have a good understanding of the path, the amount of resources and time required to get a product or a service on the market.
If we add to this the ability of the executive management to adjust the allocated resources with the main goal to reach a new target, then we are also calling the enterprise operations controllable.
To conclude, the modern control theory can be applied to enterprise operations. Enterprise operations can be modeled using the Linear Time-Independent equations, and internal mechanism has to be controllable and observable.
The only difference from basic theory is that in practice we are going to use a time-discrete system equations to describe the dynamic nature of an enterprise change management cycle. In this case, system evolution is defined at distinct, separate points in time, rather than continuously. The controller acting in a discrete-time system is like a machine that wakes up every T interval, reads its inputs, acts on existing state, then it goes back to sleep until the next tick. The entire control feedback works on sampling data.
Enterprise Dynamic System Model – Controller and Controlled
The first representation of an enterprise as a dynamic system was introduced by Stafford Beer in 1972, and it is called the Viable System Model. In this model it is easy to identify the controlled system, and the controller mechanism. The controlled system is called “Process” in the diagram. This component is where the entire value creation cycle lives. This is also where we find the supply chain cycle. Changes to its operations are driven by a two-stage controller. The Adjuster/Organizer analyzes the outside input, which could be changes to the market, while the Controller is directly responsible for implementing changes to the value creation cycle.
The simplest model is the one in which an enterprise is represented by two information flows. One of the flows carries the enterprise change cycle, and the other one carries the value creation cycle.
The last model takes the information flows to the next level. “Process” component becomes a three-steps workflow, while the operational change cycle becomes a six-steps implementation plan. This model, called the “Action-Verb,” reflects the way an enterprise is divided into physical departments. The first step called “KNOW” is completed by the marketing team. They explore the market and the customer needs for possible sources of new products and services. Once these sources have been aggregated into proposed business models, they are presented to the executive team. They then decide which ones can be used as a target to “AIM” the future business operations. Once that decision is made, the next step is to “EVALUATE” if the business has the portfolio of resources required to reach the target. If it passes this phase, the engineering team gets to work on “DEVELOPING” the product or the service. Once completed the engineering work, the production team designs the engineering changes required for “IMPLEMENTING” the operational changes. The last step before introducing the new product or service in production is to schedule it. The “PLANNING” step is where we find the Master Scheduler used to control the entire supply chain cycle.
Enterprise Model with Steady Operations
The vast majority of businesses are very cautious when it comes to disrupting their operations, especially if they are profitable. The equilibrium is usually achieved by a steady flow of profits from current supply chain cycle.
To compare this with traditional controlled system we use the pendulum analogy. If there is no perturbation, then the gravitational force will maintain the pendulum system in a steady state. There are no changes to its state-space representation.
There are many businesses that operate with this model. The executive team mainly works to maintain the flow of products or services.
Enterprise Controller and Its Main Cycle
In order for the executive management to use control theory as the main tool to drive change in an enterprise, the entire workflow path has to be observable and controllable.
If we use the pendulum analogy, the gravitation force is acting as a main factor that brings back the system to a steady state. In the enterprise, the factor that brings the business back to a steady state after responding to an external perturbation is the generation of profits.
It is important to note that the two cycles, enterprise change and operations, are running two separate information flows.
Sate-Space Theory, Lyapunov method, PID controller, and Region of Attraction
Being able to read on every step during the change cycle how far the internal state is from targeted one, and what would be the amount of resources required to bring it back on track are only two of the benefits you get from applying control theory. Modern control theory can reach beyond the direct control of the change cycle.
As we mentioned previously, in control engineering and system engineering, a state-space representation is a mathematical model of a physical system that uses state variables to track how input shapes system behavior over time through equations. These state-variables change based on their current values and input, while output depends on the current and previous states, together with current input. The state-space is also called time-domain approach and is equivalent to phase-space in some dynamical phase.
When it comes to enterprise operations, the physical system is the process associated with the development of a new product or service, from initial idea to the end of its life. The state-space representation uses financial resources as the state variable, and time as a way to capture various phases associated with the entire process.
For the mathematical model and control theory to apply, we mentioned before that the system behavior has to be linear, time-independent, observable, and controllable. As a consequence, in order for the control system theory and Lyapunov stability equations to apply, the enterprise operations have to be organized first to be linear time-independent, observable, and controllable.
Next is a proposed state-space representation of a new product or service lifecycle.
The ultimate goal for a change in operations is to reach a certain target for profitability. This takes place only after a certain number of products or services are delivered to customers. The initial batch is always used to recover the initial investment from the development phase.
The entire lifecycle of an operational change has two phases. The first one is the development phase. Over this phase we have total control because the process is both observable and controllable. During this time, we can use the PID [Proportional, Integral, and Derivative] Controller as the feedback mechanism. This is the most widely used feedback-control mechanism in engineering.
The second phase is more complicated as the control is shared with the customer base. The main goal for the executive management is to establish the ideal price, one that is liked by both the business and by consumer. To calculate this price, we can use the Lyapunov equations to determine the region of attraction for the product or service price.
As we can see from the diagram, there are two curves. One of them is the product or service lifecycle forecast. These values were finalized during the evaluation phase. The second curve is the actual values. The difference between them is called delta. The main goal for entire control mechanism is to minimize the delta values during the entire product or service lifecycle.
Model-Based System Engineering
Management consulting is way overdue for a major refresh. This field has over a trillion dollars in revenue every year, but there are signs that they reach their limits in understanding and solving the current problems businesses have. The case method is over 100 years old, and as Jay Forrester said, the current MBA programs, all 13,000 of them, are targeting only business operations, instead of enterprise design.
The only way forward is to piggyback on the success the field of system engineering has. The next diagram shows a workflow engineers are following when they are targeting the design of a car. This method is called Model-Based System Engineering, and it has already been used with great success in designing many complex industrial products.
It is clear that the use of MBSE in designing a car cannot be easily translated to the design of an enterprise. In the next diagram there is a proposed workflow for enterprise designers that combines the modern control system theory with Model Based System Engineering.
There are a lot of steps that require a lot of details. Some of these concepts are covered by an Udemy course called “Enterprise Design Engineering - Fundamentals - A New Era for Business Management - The Roadmap from Business Operators to Enterprise Designers” The complete enterprise design map goes beyond the product or service lifecycle. It includes concepts like information stores, fly-by-wire platform for technology, and visual skill map for managing employee competencies.
More about the MBSE and the way control theory applies will be detailed in a future course.
New Product Development and Calculated Risk
When an enterprise change is initially planned, the executive team creates a cost vs. profit forecast that covers the entire product or service lifecycle. When the teams are starting to implement various steps, the executive team collects the actual values and compares them with those forecasted.
Many times, various factors, internal or external, will have an impact on actual costs or prices. In the next diagram we represent three scenarios. The first one is the forecast, and the other two are versions of actual paths.
The first one is the forecast. Lyapunov equations are used in this case to calculate the region of attraction for targeted prices.
The second scenario is the desired one. The collection of actual values indicates a process that can be brought under control. Using a PID mechanism and Lyapunov equations, the executive team can continuously adjust the actual values to match the forecast.
The third scenario is the one that completely misses the forecast. The collection of actual values indicates that the new product or service is unlikely to reach profitability. In this case Lyapunov equations can be used to predict the new region of attraction for prices that can be used to calculate the losses.
Based on these three scenarios we are marking in the next diagram two regions for risk. If we find that actual values fall in one of these two regions then there is a risk that profitability for the new product or service is out of reach. If the higher cost is reached during development [region I], then the risk is considered medium. But if the higher cost than the one that was forecasted is reached during production and selling [region II], then the risk of reaching profitability is high.
In both scenarios, control system theory together with the PID controller mechanism can be used to minimize the losses, and even bring the new product or service to become profitable.
Functional Envelope Map and Calculated Risk
Managing the overall cost and timeline using control system theory and PID controller mechanism are not the only tools to manage risk.
In the development of a new product or service there is a fundamental principle related to functionality. It is very likely that certain functions of great value to consumer costs a lot more, and it takes longer time to implement because of their complexity. For instance, a smartphone touch screen display is far more complex than the previous generation which used buttons for input. As a consequence, the higher the value for consumer, the higher the risk that the cost in resources to develop such feature will not be recovered.
This relationship between benefits/complexity and risk is represented by a Functional Envelope Map.
When the executive team does the forecast for a new product or service, such envelope map can be used together with the PID controller to lower the risk during the development phase.
Conclusion and Future Plans
A new course about the MBSE and control theory applied to enterprise design engineering will be followed by a software application that will help with implementation [patent pending]. Such an application can be used by a management consultant to standardize all its services.
Business Genome Map
You always wanted to have a clear map of a business with all its processes and how they interact, regardless of the size, or industry. Business Genome Map is a complete and comprehensive record of all business processes and their interaction. Other than a map of all processes, the Business Genome Map has two elements associated. One of them is the network of communication channels, also called the enterprise information highways. The second element is a set of components called information stores. They manage the lifecycle of main business entities that exist in an enterprise.
Dynamically Stable Enterprise
Digitization of the enterprise is on every business strategy agenda. It is also one of the most complex and costly task. Its main goal is to acquire, design, and implement a complete and comprehensive set of enterprise applications that will support business operations. The Dynamically Stable Enterprise provides both a concept and a reference architecture to help complete this task.
Visual Skill Map
Despite all the progress in technology, employees are still in the driving seat when comes to the value creation. Within this context, there is one element that drives a lot of frustration in a business, both on the employee and on the management side. This happens when there is a mismatch of skills between an employee and the tasks associated with his role. The enterprise design introduces a science-driven tool called the Visual Skill Map that will help understanding where this mismatch originates and how to fix it. It gives the management a tool to evaluate the skills required for a role, and it gives the same tool to the employee to help bridge the gap.
Viable Enterprise Operating Platform
The enterprise design engineering field brings together science, tools, and a method to help run all information-centric projects in a business. The overall method is called the Viable Enterprise Operating Platform and it includes Physics of Information, frameworks like Business Genome Map, Dynamically Stable Enterprise, and Visual Skill Map. This method is an extension of the traditional waterfall engineering method to information-centric projects.