The Application is King

Regardless of "Fancy-Doo-Dad" words for cryptic mathematical algorithms, you must know how to use them properly. It is the application of any technology to solve a need that is valuable.

Mesh Modeling

Neural networks have turned out to be of limited utility in most industrial applications. We have instead invented our own modeling algorithms after decades of experience. We call these algorithms "Meshes". There are no academic papers on these algorithms. They are our invention, they are secret, and they work very well. Don't ask. We won't tell.

Genetic Algorithms
We have been the leaders in Genetic Algorithm technology for 20 years, having created and sold the first such library for Windows back in the 1980s. Over the years we have learned and refined our optimization and search technologies to arrive at GA-CPG algorithms.

Swarm Technology
Our most recent addition is our Swarm Technology found in BioComp Dakota. This techology adaptively walks-forward collections of systems, adapting their performance determined over a look-back period. This proves to be effective in financial trading applications.

On-Going Innovation
We are constantly thinking, trying, prototyping, evaluating and developing new innovative technologies and strive to prove their worth in commercial applications.

  • Platform Technologies
    Microsoft .NET 2.0
  • Microsoft Windows
  • Intel / AMD Processors
  • Single, Multi-CPU and
  • Distributed Multi-Processing
  • Language Technologies
    Visual Studio 2008
  • Visual Studio 6
  • C, C++, C# and Visual Basic
Core Technology Statement

Please note that many of our core technologies are vital to our business and are protected by trade secret. Accordingly, we do not divulge their inner workings so that we do not lose our competitive advantage. Instead, we provide you the ability to use them easily and effectively.

 

For over 20 years, the key BioComp technology staff have researched, applied, and learned the utility of different algorithms. Every algorithm has its strengths and weaknesses. We look for and invent "robust" algorithms, ones that are flexible and powerful enough to be applied in many ways with little or no modification. Through this process we have modified select algorithms to improve their performance and results.

Model
Modeling is the act of capturing relationships within data to perform a useful purpose, such as understanding those relationships, making predictions or "inverting" the model for process, product and performance control. We have created "Mesh" modeling algorithms for this purpose. These models are automatically created through a search process that seeks combinations of input variables and models that maximize the performance of the models. This includes a 3 to 4 fold model validation. Our models will out perform any other vendors across a spectrum of useful ("real world") applications.

 

Applications

  • Virtual Sensors
  • Customer Purchase Behavior Modeling
  • Product Performance Estimators
  • Maintenance Expectations and Failure Probabilities

(endless others)

Understand
We provide various understanding technologies including data visualization (e.g., 2D and 3D scatter plots, histograms, temporal correlation plots, etc.), highly capable model sensitivity analysis and a technology we refer to as Causal Analytics that fuses together time-based correlation, sequential event ordering and model response to arrive at prioritized rankings of causal variables over time against a target variable. While this gives an indication, causality can only be proved through disturbing a process.

 

Applications

  • Why Customers Buy Products or Services
  • What Drives Customer Satisfaction
  • Why Does Product Performance Fail
  • Why Machines Fail
  • What Drives the Financial Markets at this Time

(endless others)

Optimize
We provide an array of optimization technologies, including our GA-CPG Genetic Algorithm (it is so good we use it in nearly every product), model inversion for optimal control, and "Change-Sense-Change" technologies (self-sensing search controllers).

 

Applications

  • Increase Customer Satisfaction
  • Increase Sales
  • Improve Product Performance
  • Avoid Machine Failures
  • Determine the Best Strategies

(endless others)

Adapt
Our most recent addition is our Swarm Technology found in BioComp Dakota. This techology adaptively walks-forward collections of systems, adapting their performance, determined over a look-back period. This proves to be effective in financial trading applications.

 

Applications

  • Track Shifts in Buying Behavior
  • Adapt to Changing Financial Markets Behavior

(endless others)