top of page


Process informatics business

アンカー テスト

What is process informatics

This is a method of efficiently searching for desired conditions in the shortest time possible by utilizing artificial intelligence technology for process design.
With iCrystal's unique approach, even at manufacturing sites where it is difficult to accumulate big data, analysis and optimization can be achieved in a short period of time using limited past data.

Provided by iCrystal
Characteristics of process informatics

Along with limited past experimental data, learning the intentions of experimental conditions determined by skilled engineers dramatically improves AI learning efficiency. It is also possible to optimize manufacturing process conditions using AI in computer simulation data.
By repeating the game of catch in collaboration between engineers and AI,
Improve the accuracy of the digital twin and propose optimal conditions.


What is a digital twin

A digital twin is a technology that collects information in real (physical) space using IoT, etc., and reproduces the real space in cyber (virtual) space based on the transmitted data. It is a mirror-like image that copies the real world environment into the virtual space, and is called a digital twin, meaning "digital twin".
Unlike conventional virtual spaces, digital twins are characterized by the ability to reproduce a more realistic space in real time. Behind this is the evolution of IoT and AI. By sending various data acquired by IoT to a server on the cloud in real time and analyzing and processing by AI, digital twins can reproduce physical space in real time.
Digital twins make it possible to simulate future changes in physical space in virtual space, and prepare for changes in physical space that will actually occur in the future.

process informaticsWhat you can do with


Optimizing existing manufacturing equipment and facilities to their performance limits

It is possible to search for optimal setting conditions within the constraints of existing manufacturing equipment without investing in new equipment.

  • Improve quality and accuracy

  • Increased speed

  • Decrease defect rate (improve yield)

If a process that used to take 10 minutes can be reduced to 5 minutes, it will be possible to process the same number of products even if the equipment is halved.
Improved yields reduce material costs and time, and improve profitability.



Improvement and new development of manufacturing equipment

In addition to finding the optimal solution within the constraints of existing manufacturing equipment, it is also possible to propose changes to manufacturing equipment specifications.

  • Since the specifications of digital twins in cyberspace can be freely changed, AI can propose conditions that cannot be changed due to the constraints of existing manufacturing equipment.

  • If the accuracy of the digital twin is high, it will be possible to calculate the amount of new investment associated with the improvement and new development of manufacturing equipment and the benefits thereof with a certain degree of accuracy, making it possible to make bold capital investments.

In addition to changing the manufacturing process, it is now possible to consider a series of optimizations in a short period of time, including equipment improvements, material changes, design changes, and changes in pre- and post-processes. accuracy is improved.


Ultra-fast computer simulation

​AI models replace high-performance simulator calculations, making it possible to produce results instantly.

  • ​When performing simulations with various combinations of conditions using a high-performance simulator, calculations can take dozens of hours or more even with an expensive computer. On the other hand, using an AI model makes it possible to produce results instantly, so it is possible to test various conditions in a short time.



Promoting digitization of expert technical know-how

We formulate tacit knowledge and experiences that are difficult to verbalize.

  • Expert judgment formulation

  • Analysis of expert decision-making grounds

We formulate the tacit knowledge and experience of experts, and contribute to the reduction of the burden on skilled engineers and the succession of technology.


Case Study

Process informatics can be applied to all kinds of manufacturing.
iCrystal has a track record in various fields such as semiconductors, metal processing, glass, and agriculture.

metal processing

Find conditions that allow quick machining while prolonging tool life.


Molding process for resin used in car interiors

Find the resin impregnation conditions that do not warp, distort, or bubble.

glass lens

Find the pressing conditions that produce uniform, distortion-free lenses.


Flow of consultation


Overview of optimization using process informatics


Consideration of the target equipment and target process by the customer and pre-contract consultation with our company


NDA (non-disclosure agreement) concluded


Sharing issues and discussing parameters to be optimized and optimization methods(and on-site visit if necessary)


Contract conclusion (issues to be solved, implementation details, target values, period, costs, etc.)


​Our analysis of past experimental data and simulation data provided by customers


Investigation of various processes and search for next experimental conditions


​Optimization experiments by customers and feedback to our company


​Hold technical meetings as appropriate and repeat steps ⑦ and ⑧ to narrow down optimal conditions and reach target values.



Please feel free to contact us.

bottom of page