Forum & resources

Access to forum and resources is limited to registered members.
Please log on or apply for membership

Share this page

Share this page

Linked in ViadeoTwitter Facebook
Google+

Join us, Get more

Join the Artelys community and get a full access to :

• Tips and tricks
• Upcoming developments
• Ressources
• Client case
• Experience sharing

Register

KNITRO

 

Robust and powerful solver for non-linear optimization and large complex problem solving

Zinea Knitro

Non-linear optimization problems arise in numerous business and industry applications, such as optimizing a portfolio in the financial industry or getting the optimal pricing in the energy industry. Customers in hundreds of sites worldwide rely on KNITRO to deliver high performing and powerful optimization solutions for the most challenging applications in all industries: financial, energy, communications, science and engineering.

KNITRO is developed by Ziena Optimization. Artelys distributes and provides support worldwide. Ziena was founded in 2001 by world-renowned scientists from Northwestern University to provide superior software tools and responsive professional technical support for difficult non-linear optimization problems.

KNITRO is available as a standalone component and with the AMPL modeling environment. Artelys provides additional modeling and integration services.

 

The most advanced optimization techniques

KNITRO is an ideal tool for solving problems, such as pricing or revenue management, parameter identification in financial and industrial models or design and operation of electricity and gas transportation networks. It can efficiently handle non-convex non-linear models with thousands of variables. KNITRO is the first non-linear optimization tool that offers two options for inequality constrained problems: "interior points" and "active set." These options reinforce the efficiency for a wide range of models with multiple structures.

Artelys and Ziena Optimization offer partnership programs with universities, public training and research centers to promote teaching and research in optimization. Academic partners get KNITRO and AMPL licenses at discounted prices.

For further information on academic partnership, please contact us.

Key benefits & Features

Key benefits

• Solve complex non-linear problems: handle large-scale, complex problems with millions of variables and constraints 
• Offers the leading combination of computational efficiency and robustness 
• Includes high accuracy solutions via the Active Set algorithm 
• Offers the ability to choose the best algorithm among three options
• Flexibility of use

Key features

• Unique solver able to solve a wide range of problems
• Three different algorithmsBoth interior-point (barrier) and active-set solvers
• Derivative-free, 1st-derivative and 2nd-derivative options
• Multi-start feature for seeking the global solution
• Options to remain feasible throughout the optimization
• Both iterative and direct approaches for computing Newton-like steps

Problems classes solved by KNITRO

• General Nonlinear Constrained Problems (NLP), including Non-convex
• Linear Problems (LP)
• Quadratic Programming Problems (QP), Both Convex and Non-convex
• Least Squares Problems/Regression, Both Linear and Nonlinear
• Systems of Nonlinear Equations 
• Mathematical Programs with Complementarity Constraints (MPCCs/MPECs)
• Mixed Integer Nonlinear Problems (MIP/MINLP)

Programming interfaces

 

Modelling language interfaces

 

Operating Systems

 

Business applications

KNITRO is used in all type of industries such as:

Financial industries:

• Portfolio optimization
• Optimal pricing and risk management
• Credit risk
• Strategic bidding and auctions (Nash equilibrium)
• Demand optimization
• Nonlinear least squares (data fitting)

The Energy sector:

KNITRO is used in various power transmission problems such as:

• Nonlinear OPF (optimal power flow) problem
• Optimize generator costs and line losses 
• Nonlinear AC power flow equations
• Transmission and other technical constraints

 

Please feel free to contact us to receive more information regarding our business applications.

 



User manuals

KNITRO version 8.0 is described in the user manual: KNITRO 8.0 User's Manual, December 2011.

Older releases:
KNITRO 7.0
KNITRO 6.0
KNITRO 5.2
KNITRO 5.1
KNITRO 5.0

 

Ziena License Manager User

Ziena License Manager User 7.0

Download a trial version

step 1Select version

  • This version is for students who want to use KNITRO for educational purposes.

  • This version is for evaluating KNITRO in the business world. Cannot be used for gain nor profit.

  • This version is for degree awarding institutions that want to evaluate KNITRO for research or educational purposes.

DO NOT REMOVE

Frequently asked questions

This section has a selection of general issues encountered during the installation and the use of KNITRO and our answers to help you. For specific help, please contact us.

Q : What should I do to purchase KNITRO?
A : Please contact KNITRO sales team.

Q: What should I do when I have difficulties installing KNITRO?
A: Read section 1.3 and 1.4 (Installation and Troubleshooting) of the user manual. If you cannot resolve your problem, please contact KNITRO support team (for users under maintenance only).

Q: What should I do to make KNITRO solves my problem faster?
A: Read section 2.15 (Tips and Tricks) of the user manual. If you cannot resolve your problem, please contact KNITRO support team (for users under maintenance only).

Q : How do I get KNITRO 8.0 to behave similarly to KNITRO 7.0?
A : KNITRO 8.0 can behave more like KNITRO 7.0 by setting the following non-default options in KNITRO 8.0: presolve=0, bar_switchrule=0.

Q : How do I get KNITRO 7.0 to behave similarly to KNITRO 6.0?
A : KNITRO 7.0 can behave more like KNITRO 6.0 by setting the following non-default options in KNITRO 7.0: blasoption=0, linsolver=4, bar_directinterval=some very large number (e.g. 10000).

KNITRO support

Artelys provides technical assistance worlwide for KNITRO, either as a standalone component or with AMPL modelling environment.

More information