CADLM accelerates product design and development via real-time parametric simulations 

A Technology company

We develop innovative computational and mathematical approaches for solving complex engineering, scientific, industrial and data problems. This is a multi-displinary approach based on a wide variety of fields including statistics and data science, machine and deep learning, control, optimization, numerical analysis, applied mathematics, physics, biomechanics and material sciences.


What we do is unique, revolutionary even. We basically design algorithms based on fusion of intelligence, capability and creativity to create innovative solutions to your problems. Most of our solutions are customized, however we have suite of tools that enables you as a beginner or advanced user to create your own unique solutions as well.

Stock Exchange

The environmental cost of computing worldwide is reaching that of transport. Increasing number of solvers (often decoupled in terms of the underlying physics), size of models and their CPU cost is hindering innovations, forcing engineers to start from zero at every design cycle thus not exploiting past experience. Numerous design iterations are creating computing bottlenecks in spite of huge advances in hardware and storage solutions.

ODYSSEE, allows to remedy this problem by reducing the effective number of CPU intensive computations, replacing them by real-time equivalents capable of being run on small size laptops.

Computer Processor


Reducing the cost of production and design requires optimization of designs and processes leading to efficient manufacturing. While a great majority of engineers are aware of the advantages of optimization techniques, few of them actually have the resources and the opportunity to employ them. Numerous iterations are necessary either over the solver solutions or their approximate surrogates. The existing solutions often ignore the fact that many runs may be avoided using “learning” instead of calculating every single possible combination of design parameters.

ODYSSEE, allows to remedy this drawback by combining optimization and learning in order to provide very precise surrogate models which may be employed at very low computing cost. The gain in total computing effort may be in the order of 1 to 1000 or even more. Robust optimization can in particular benefit from this and allow for running thousands of runs in seconds or minutes.


Rethinking Design COST

Computing, Optimization, Simulation, Time

A design space is identified by variables or parameters. The associated results are simulations outputs, experimental measurements or both. The primary purpose of simulations is to replace real-life experiments by lower cost and easier to control counterparts. This however, has it limits due to often too complex physics or lack of reliable models. We simply don’t have an intermediate solution allowing to learn from experiments or simulations and establish a link between the two, allowing for a “digital twin” or simply a “coupled simulation-experimental” model. Additionally, modeling is often confronted with issues related to licensing, multi-scale considerations and confidentiality, especially in an industrial environment.

ODYSSEE, establishes the missing links and allows for the above enhancements to be made in order to improve the performance, the precision and the feasibility of simulation based engineering.

Workers at Their Computers

Even in the case where all the above barriers are overcome, the speed with which new solutions can be obtained and explored remains a major obstacle. Engineering is about “what-if” interrogations and incremental learning. Many decisions do not necessary require a detailed FE model nor an elaborated experiment. This is especially the case in early stages of the design in “V”. We can simply learn from our past experience and data and provide real-time answers.

ODYSSEE technology is based on real-time computing; therefore you need little-computing effort for parametric studies and optimization, exploring new horizons in data science in order to capture the most important part of the solution, leaving the details to more elaborate and CPU intensive computing, optimization and simulation technology.

Vintage Wood Clocks