Process control plays a dominant role in driving your company’s bottom line. Every time an operator makes a set point change, it can affect plant profitability

For over 40 years industry has invested significant resources in process optimization, from early PID systems to modern APC and AI analytics. But the truth is that industrial complexity has been growing faster than technological solutions can scale — until now.

New technological breakthroughs within AI have enabled intelligent process optimization solutions to tackle modern industrial challenges.

Key pain points of industry
Key pain points of

Quality issues

Yield targets

High cost of scrap

High cost of energy and environmental impact

Industrial complexity has been growing faster than technological solutions can scale.

Process Control Optimization reality
Process Control Optimization

Improving the productivity of a plant by using advanced technologies to increase production rates while keeping costs and environmental footprint down is a continuous goal of a plant manager. Traditionally, a common approach has been to invest a lot of resources in implementing modern process control systems such as MPC, MLC, expert-based systems, fuzzy logic, genetic algorithms, and so forth. Such systems are usually referred to as APC (Advanced Process Control).

To improve productivity, a common approach has been to apply Advanced Process Control (APC) solutions. 

Traditional APC solutions do not scale to the challenges of industry, suffer from performance degradation, and have a high cost of ownership.

APC solutions offer considerable benefits right after installation, but suffer from a slow degradation of performance over time. The reason for this is that APC solutions have always been challenged by the very nature of industrial processes, which are dynamic and high-dimensional. What’s more, the realities of industry—degradation of equipment, replacement of instruments, and frequent reconfiguration of production and recipes—make the challenge even greater. APC software makers struggle to close the gap, but outcomes are often far from optimal.

Typical performance degradation of installed APC solution over time

Key difficulties with traditional process control systems


Costly and time-consuming to configure, customize (manually build mathematical models) and maintain.


Neither fully tuned nor adaptive to a dynamic process environment. Requires costly and time consuming manual intervention.


Operator decision-making has become increasingly complex, even for senior experts, due to the many conflicting factors involved.


System performance affected by differing competence and experience levels of operators.


Traditional systems operate with a set of deterministic instructions derived for predictable environments, yet process dynamics continuously change.


Systems oscillate at their set-point (steady state error) and are slow to adapt to changing set-points. They also struggle with trade-offs between speed and quality.

Industry is continuously changing, while traditional industrial solutions are static and require high investment and expertise to operate and to adapt to new dynamics, resulting in a high cost of ownership. Simply, traditional process control systems do not scale to the challenge.

Unlock the Potentintial
Unlock the

SentianController is based on new generation of Artificial Intelligence (AI) with a focus on goals to secure the long-term success and sustainability of your industrial process.

SentianController learns the details of your process on its own, freeing you to set the high-level objectives. It’s the outcome that matters.

SentianController unlocks the potential of achieving high benefit


Automatic Dynamics Model

Automatic inference of dynamics model from operational data, eliminating the need for costly manual modeling and tuning.

Continuous Learning & Adaptation

True AI learning from live data. It continuously adapts with drifts and changes in process dynamics.

Continuous Exploration

Discovery of more optimal set-points through continuous exploration of never before explored state space.

Data Efficient

Works even in the cases of limited data availability.

Scenarios Analysis

Analyzing hundreds of possible future scenarios/trajectories for finding the single most optimal action.

Set-points Regularization

Regularization model keeps actions within stable operational space.


Maximize yield

By dynamically adapting to changes in the processes and environment, while continuously optimizing towards the goals to get the best yield and throughput. 

Reduce cost of operations

By including key resources (such as additives, chemicals, water, heating and cooling) in the optimization process, thereby ensuring that resource consumption is optimal given your objectives.

Use expertise and human resources optimally

By capturing and using your expert operators’ operational knowledge and experience in generating process control optimization results, and thereby helping more junior operators achieve higher performance, while at the same time reducing risks caused by retirement.

Increase autonomy

Many industrial processes are not well understood by process experts and need constant monitoring as they are too dynamic and complex. Many can now be semi-automated or even made fully autonomous.

Together we can improve your process 


*According to McKinsey, the US department of Energy, and our experience

SentianController addresses competitive opportunities and threats, efficiency, and pain points — all with a compelling, measurable return on your investment.