DOE - Design Of Experiment

Broad ranges of engineering activities are widely applied in both pre-development and product development, and in construction of many industrial branches, in power engineering, ecology, medicine and sports.

DOE allows the solution of complicated and complex task, when the final result is influenced by numerous factors. It is therefore necessary to carry out a comprehensive analysis utilizing a tool that is able to incorporate all the factors. The aim is to identify critical factors, their optimization, minimizing variability and in the end create a robust model in which the characteristics are resistant to external natural influences. This doesn't concern only purely statistical methods, but linking of analytical and statistical techniques.

In terms of pre-development stages, using DOE techniques can be utilized for research of construction and manufacturing technical solution alternatives with regard to their interaction and influence. The main reason is the prediction of system behavior in the future and show possibilities of improvement. DOE applications at this stage can prevent quality problems in final production and thus again contribute to cost reduction. The methodology can be certainly used in every planning stage, during product development process and during serial production.

DOE methodology is one of the tools that require absolutely precise preparation prior to experiments. Firstly it is necessary to analyze the task, determine a precise specification of it and prepare basic inputs for the actual experiments planning. Time spent on such preparation activities is easily gained back, because thanks to the preparation we execute only necessary experiments.

CFD - Computational Fluid Dynamic

  • Computational - employing mathematics with computation technique
  • Fluid Dynamics - dynamics of moving particles

CFD is a computational technology that allows us to see the fluid dynamics. By using CFD it is possible to create virtual computational models, which represent general systems or devices and which enable us to predict their relations, such as vortexes, limiting layers etc. CFD programs provide a powerful tool for simulating fluid and gas flow, temperature or mass transfer, interactions between the solid and air component of the task and mechanical waves. Using CFD analysis, we can create a virtual prototype of a system or a process, which can be viewed in chronological sequence and then analyze the behavior response to different stimuli. CFD program offers the user unlimited number of outputs in the form of data or graphic representation, from which the behavior of modeled system can be easily determined.

FEM - Finite Element Method

The finite element method is a numerical method used to simulate the behaviour of stress, strain, frequencies, heat flow, electromagnetism and so on by creating a physical model. Its principle lies in the discretization of continuum to certain (final) number of elements. The surveyed parameters are being determined at certain nodal point. FEM is used primarily for the control of already designed equipment or to determine critical (most loaded) point in a construction. Although the principle of this method has been known for a long time, its mass utilization started only due to the development of modern computer technology.

General process of CFD/FEM simulations

The whole modeling process is always divided into at least three phases. The first phase involves Pre-processing of the task, i.e. preparation of the model (preparation of geometry and computational mesh, definition of physical and material model, determination of limiting conditions etc.), the second consists of the calculation itself (Solver) and the final involves post-processing, i.e. visualization, evaluation and interpretation of results.


This is the first step in creating a new computational model. Involves the creation or modification of geometric models using CAD (Computer-Aided Design) programs, the creation and control of surface and volume meshes and export to computing software.

Most used programs can import a previously created CAD geometry. Subsequently we utilize series of tools that simplify the insignificant details of the model, which unnecessarily increase the demand on computer technology and are unimportant for the analysis. On the geometry of the finished model is then applied the surface and volume mesh, which respects the specific requirements of numerical simulations.

Computation (Solver)

CFD program carries out the required computation on the basis of mesh and given boundary conditions and stores the results in files that are used for further processing.

Most of today used solvers offer many possible settings :
  • multi-phase flow
  • material response
  • shift and deformities of objects, induced by pressure load
  • turbulent models
  • radiation, convection, conduction
  • acoustic, mechanical wave propagation


This is the final step of the analysis. Based on calculated values, images or animations we carry out their interpretation. Correct interpretation of results is very important and requires an experienced professional, who is able to execute a correct analysis based on his experiences and expertise.

For post-processing it is usually possible to use tools of the computational program, as well as specialized software, which usually offers much wider possibilities.


Optimization is a process of finding the best possible solution based on suitable purpose functions in an area determined by set of limitations.

The basis of any optimization is the parametric model, which allows to modify the model of solved component by a rapid change of parameters. At first it is necessary to define the target function and set restrictive conditions. For automatic control of the optimization process there are today many commercial systems available. These programs usually already include a large number of optimization algorithms, i.e. the classical gradient, genetic methods, DOE, their combination, approximation function and many other tools. In general, the optimization programs try to determine the extremes (minimum / maximum) of defined target function by changing given parameters under the same restrictive conditions. In case of 3D shape optimization it is appropriate to use methods that require the lowest possible number of computing cycles.

Optimization scheme
Methods of general optimization :
  • RSM/DOE (Response Surface Method)
  • Gradient methods
  • Evolution methods
  • NN (Neural Network)
Methods of demanding computational optimization :
  • MACK (Multivariate Adaptive Crossvalidating Kriging)
  • FBM (Falling Ball Method)

Today, a different approach is more and more used, which allows maintaining the original mesh and the shape change is made directly by shifting the nodes of individual elements, the so-called mesh morphing. SCULPTOR and ANSA products are examples of such technology.

CAE - Computer Aided Engineering

CAE is a computer analysis of components, mechanisms and machines. It includes simulation, verification and optimization of products and their production. CAE tools are nowadays a relatively common part of development and design process and are becoming an important source of information needed for qualified decision making during development of new products and production planning.

Areas, covered by CAE tools, include :
  • Computation of components and sets stress by using FEA (Finite Element Analysis)
  • Analysis of heat flow and propagation by using CFD (Computational Fluid Dynamics)
  • Kinematics
  • MES, Mechanical Systems
  • BMX, Behavioral Modeling Extension
  • Analytical tools for simulating production processes such as casting, forming and moulding.

Project Management

Project Management is a way of planning and implementation of complex, usually one-time operations, which need to be executed in due time and with planned budget in order to achieve set targets. Project Management can be in brief described as an effort to maximize effective and efficient project execution.

To increase the probability of project success, Project Management uses many methods representing proven and described procedures, addressing different phases of the project, such as project initiation, risk analysis and change management.

FMEA - Failure Mode and Effect Analysis

FMEA is a method that can be used to prevent or lower the risks that occur during development and manufacture of a product. Its purpose is to identify potential risks with respect to the importance of detection probability.

FMEA is systematic sequence of activities aimed to :
  • Detection and evaluation of potential defects in a product or a process and their effect
  • Identification of actions preventing or limiting the conditions of possible defects occurrence during the documenting process