What is GENO?
GENO is an acronym that stands for General Evolutionary Numerical Optimizer. GENO is a real-coded genetic algorithm for solving single or multi-objective optimization problems that may be static or dynamic in character; unconstrained or constrained by functional equalities or inequalities, as well as by upper and lower bounds on the variables; the choice variables themselves may assume real or discrete values in any combination. In short, the algorithm does not require the problem to have any special structure.
GENO has proven its worth in the market place: the client base includes individual researchers, various university departments and research institutes, major central banks, an oil company, an insurance company and a major car manufacturer; GENO computes solutions that are regarded as benchmarks for other algorithm designs to emulate.
New GENO 2.0 Features
Internal genetic operators have been re-designed resulting in a vast improvement in performance
Provision for solving nonlinear systems of equations, as well as goal programming models have been added
Program is much easier to use because some data and parameter requirements have been automated and internalized
User may now specify that a particular point in the search space should be included in the solution process
Scope of Application
GENO may be specialized in situ to solve various classes of problems by mere choice of a few parameters. It applies to:
systems of linear or nonlinear equations
static or dynamic optimization problems, with or without functional and/or set-constraints
single, as well as multi-objective problems
It may be set to generate real or integer-valued solutions, or a mixture of the two as required.
The scope of its application includes:
Mixed Variable Optimization
Nonlinear Equation Systems
Documention: GENO is well documented and easy to use; the product includes a large number of examples programs to help kick-start user experience.
Testing: The algorithm has been tested on a wide range of real-life and artificial problems from well-known test suites; it has also been tested against well known algorithms that are embedded in popular computational systems including Mathematica, Global Optimization and MathOptimizer; it consistently out-performs many evolutionary algorithms, and the quality of its final solution is at least as good as several specialist deterministic algorithms in many cases.
Practical Example Problems: A partial list included in the product is as follows:
The Economic Dispatch Problem
The Alkylation Process
Decentralized Economic Planning
Heat Exchanger Optimization
Asset Portfolio Optimization
Job Shop Scheduling
Market Equilibrium Problem
Chemical Process Synthesis
General Resource Allocation
Operating Systems: Windows, Linux, Mac OSX
Requires: GAUSS 10 or later