Announcing LINGO 21, Algebraic Modeling Language and Solvers

 

LINGO 21.0 includes a wide range of performance enhancements and new features:

 

• Improved reproducibility when solving a problem repeatedly, particularly with the concurrent option choosing the fastest of Barrier, Primal simplex, and dual simplex.

• Global solver offers improved handling of IF statements and polynomials.

• New symmetry detection capabilities have been added to the integer (MIP) solver. This can dramatically reduce the time needed to prove optimality on some models with integer variables.

• Improved Multi-start for quickly finding good solutions to difficult nonconvex problems.

• External Solvers Improvements. Increased support for the external solvers, including linear solvers such as COPT, CLP and HIGHS. Improved support for not only CONOPT, but also external Nonlinear solvers such as IPOpt

• Enhancements to the Windows version: 

     o Improved model editor, with the ability to split the edit window into multiple panes.

     o New auto-completion of common keywords when typing.

     o Larger Undo Redo buffer with support for more operations.

     o Faster find and replace method.

     o Larger toolbar buttons.

     o Improved Unicode handling.

 

Announcing LINGO 20 : New Features 

 

 

 

      + Improved API  interface. It is now much easier to incorporate your LINGO

 

         model(s) into your own custom designed system. 

 

         The standard LINGO distribution has examples to illustrate how to do this.

 

 

 

      + LINGO in Excel.  It is now easy to integrate a LINGO model into an Excel 

 

           spreadsheet.  The end user sees it as just a smart Excel spreadsheet that can solve say, 

 

           a cutting stock problem for metal fabrication, or vehicle routing and delivery problems, or 

 

           a supply chain shipping assignment problem, and more.

 

 

 

      + New REST API  distributed computing interface. If you want to have an

 

         optimization based application accessible on the web from smart phones and

 

         other devices, this is now easier to do.

 

 

 

      + Improved capability to create a  Docker Image.  If you like to be able to move an 

 

         application from one server to another with minimum hassle, Docker has proven to

 

         be a popular way of achieving such portability. 

 

 

 

      + Improved interface to R. The R statistical package is a popular way of doing 

 

         statistical analysis.  It is now much easier to establish an interface between your 

 

         data in R and your optimization model in LINGO.

 

 

 

      + Improved interface to Python.  Python is one of the most popular languages for  

 

         doing general computations. It is now much easier to establish an interface

 

         between your Python application and your optimization model in LINGO.

 

 

 

      + Improved support for implied set names such as J01..J99,  Jan..Oct, etc.

 

 

 

+ Improved ODBC connection to databases allowing import of sets (in addition to 

 

   attributes) within Calc sections.

 

 

 

+ Ability to generate alternate optimal solutions to linear programming models.

 

   This may be done interactively, or programmatically within Calc sections using the 

 

   @NEXTALTOPT() function.

 

 

 

Performance Improvements 

 

+  Linear and Integer Solver Improvements.

 

Improved heuristics for general integer programs.

 

Average performance improvement of 2-3% on our standard test set.

 

 

 

+  Nonlinear and Global Solver Improvements

 

Faster (order of magnitude) solution of linear fractional programs (ratio 

 

     objectives)

 

 

 

+  Support for additional useful but “problematic” functions: 

 

    Power utility function (x^g-1)/g  and

 

    exponential ratio function (exp(g) – 1)/g, are important in some 

 

    situations modeling consumer behavior. The solver can now robustly avoid the 

 

    numerical problems that would otherwise occur when g approaches 0.

 

 

 

 + Linearization Improvements

 

More expressions can be automatically linearized, so you can now use a fast

 

     linear solver where otherwise a much slower (30x?) nonlinear solver might be

 

     required. Improved linearization of certain IF expressions. 

Announcing What’sBest! 19, the Optimization Add-in for Excel

 

Release 19 includes a wide range of performance enhancements and new features:

• Advancements in the Nonlinear and Global Solvers including improved handling of IF statements and polynomials.

• Improved Multi-start Solver for quickly finding good solutions to difficult non-convex problems.

• Improved reproducibility when solving a problem repeatedly, particularly with the concurrent option when solving with fastest of Barrier, Primal simplex, and Dual simplex solver.

• Improved handling of Blackbox Excel models, that have either

              o All variables binary,

              o All functions in the optimization model are linear.

In these situations, all available Excel functions can be used.

 

New What’sBest! functions 

• WBPOLYNOMIAL to efficiently handle univariate polynomials, and 

• WBCENSOR to efficiently handle piecewise linear S shaped functions as in neural nets.

 

• Improved warning messages for improperly formulated models.

• External Solvers Improvements. Increased support for the external solvers, including linear solvers such as COPT, CLP and HIGHS. Improved support for not only CONOPT, but also external Nonlinear solvers such as IPOpt

 

 

Linear-Integer solver:

   + Improved heuristics for general integer programs.

    + Average performance improvement of 2-3% on our standard test set.

   + Improved method for generating all alternative optima to a linear program. 

      The first call to LSgetNextBestSol() creates a prespecified number of corner points  

       via pivoting on the optimal solution set. 

       Subsequent calls will nonredundantly return successive corner points. 

       Standard solution query methods can be used to access primal-dual vectors 

       following each call to LSgetNextBestSol. 

 

Linearization

     + Support for Indicator constraints, e.g., z = 0 implies x + y <= 0;

     + more expressions can be automatically linearized, so you can now use a fast linear 

        solver where otherwise a much slower (30x?) nonlinear solver.might be required.

     + advanced linearization of QP and Conic models

     + improved linearization of certain IF expressions. 

Nonlinear and Global solver:

      + faster (order of magnitude) solution of linear fractional programs (ratio objectives).

      + improved bound tightening process in preprocessing of nonlinear models.

      + auxiliary variables generated automatically to improve performance with

          complicated expressions. 

      + Support for additional useful but “problematic” functions:

         Power utility function  (x^g-1)/g  and the exponential ratio function (exp(g) – 1)/g,

         are important in some situations modeling consumer behavior.

         LINDO API can now robustly avoid the numerical problems that would otherwise 

         occur when g approaches 0.

 

Interfaces:

       + Julia/JuMP is now supported officially.

       + Python interface installations now easier via pip (pypi.org)

       + Matlab interface now has two alternative methods for linear and integer

          optimization, LSlinprog and LSintprog. Argument lists follow their counterparts, 

           linprog and intprog, in Matlab’s optimization toolbox.