OR/MS Today, June 2011
Linear Programming Survey
Table 13
|
Product |
New Features |
Other Techniques |
Comments |
| AIMMS | Robust Optimization, Constraint Programming, Remote Optimization platform, Open Office, Dynamic model editing, Etc. | Advanced IDE, Interactive GUI builder, most MP types, Advanced integration including web services, Custom algorithms, Outer Approximation, Multi-start, Etc. | AIMMS is being developed and designed to succeed in its mission: bring the benefits of OR to society and help you apply OR successfully. |
| AMPL | Updated solver support | General nonlinear solver support including 2nd derivatives; support for detailed solver-specific directives and results; user-defined functions; MATLAB interfaces | Flexible handling of sets and indexing for handling complex models naturally and large models efficiently. Includes a scripting language for iterative optimization schemes. Free experimentation available through the NEOS Server. |
| C-WHIZ | Matrix supersparsity is uniquely exploited to minimize memory consumption; this lets you solve a larger problem in a | ||
| CBC | |||
| CLP | |||
| CoinMP Open-Source Solver | New release offers library interfaces for VB, C#, and Python. Linux/UNIX versions with automake/configure support. | The source is precompiled, to generate CoinMP.dll/CoinMP.so libraries that can easily be embedded into application projects. | CoinMP is a free open-source C-API interface library that includes the CLP (Coin LP), CBC (Branch and Cut), and CGL (Cut Generation) projects. |
| Coopr | Coopr is collection of software packages that support optimization modeling and analysis in Python. | Key drivers for Coopr development are Pyomo, a Pythonic modeling tool, and PySP, Python stochastic programming solvers. | Coopr is an open source project. The Coopr software is freely available, and it includes packages created by government and academic developers. |
| DATAFORM | The DATAFORM hierarchical, virtual database holds all of the model data: data tables, matrices, and solutions. | DATAFORM is a model database manager and a data manipulation language. It is rich enough to do most data processing as well as the model generation, solution control, and report writing. | |
| FICO Xpress Optimization Suite | Auto multi-threading, Auto multi-threading, Ultra-large scale true 64bit support, Distributed solving, V Large performance increase, Much more.. | Constraint programming, User heuristics, UI builder | FICO offers the leading modeling tools, fastest solvers according to recent benchmarks and the best services and support for your projects. |
| Frontier Analyst 4 | Comparisons allow side-by-side review of different analysis options. | Implements all common Data Envelopment Analysis (DEA) models. | Frontier Analyst is a DEA tool to help managers learn about the efficiency in their organisation. Full graphics as well as in depth tables. |
| GAMS | http://www.gams.com/docs/release/release.htm | stochastic optimization (scenario reduction, importance sampling, Benders decomposition), scenario solver | links to Gnuplot and matlab |
| GENO | Interval Calculus | Interval Calculus | GENO solves single/multi-agent, static/dynamic optimization problems, with or without equation constraints for continuous or disrete solutions |
| GIPALS - Linear Programming Environment | SSE2 and Multi-Core CPU support | Reads and writes MPS files. | A free 30-day trial version of GIPALS is available without registration at Optimalon's website. |
| Gipals32 - Linear Programming Library | SSE2 and Multi-core CPU support. | Read and write MPS files. | A free 30-day trial version of Gipals32 is available without registration at Optimalon's website. |
| GLPK (GNU Linear Programming Kit) | |||
| Gurobi Optimizer 4.5 | Interior point for LP and QP. Convex QP and MIQP support. .NET interface. Free academic licenses, Cloud and Pay -By-The-Day licenses. | Call backs available. Python interactive shell with support for all model and algorithm features. | |
| IBM ILOG CPLEX Optimization Studio | Keyless licensing; nonconvex QP; SPSS Modeler connector; parallel for LP and for MIP root node; MIPKappa; CP multi-objective | Constraint programming; modeling language (OPL); Integrated Development Environment; APIs for C++, Java, .NET, Python, and C | IBM ILOG CPLEX Optimization Studio covers the entire development process from editing/testing/tuning/profiling/deploying optimization models. |
| KNITRO | |||
| LINDO API | Stochastic programming, K-Best solutions, Infeasibility analysis (LP, IP, QP, & NLP), Chance Constrained Programming. | Automatic detection of problem type. | Suite of fast callable solvers for creating customized linear, integer, nonlinear, quadratic, stochastic and global optimization applications. |
| LINGO | Stochastic programming, K-Best solutions, Infeasibility analysis (LP, IP, QP, & NLP), Chance Constrained Programming. | Automatic detection of problem type. | Fast linear, integer, nonlinear, quadratic, stochastic and global solvers and a comprehensive modeling language with convenient data options. |
| LOQO | |||
| Mathematical Modeling System | Professional Report Generator, Row generation techniques, IIS-infeasibility search | Integrated drawing library, Web service version, Automatic Modeling (Latex) documentation | Complete modeling system, free basic version |
| Microsoft Solver Foundation | Microsoft Solver Foundation is a .Net platform for mathematical optimization and modeling. Solver Foundation makes it easier to build and solve real-world optimization models by providing a .Net API and runtime for model creat | ||
| MOSEK | A Python interface has been added. | MOSEK ships with interfaces to Java, .NET and Python. | |
| MPL Modeling System | Increased speed and scalability; New solver versions: CPLEX 12.3, Gurobi 4.5, CoinMP 1.5, Lindo 7, Knitro 7; Stochastic Prog; | With the OptiMax Library, MPL models can be embedded into end-user applications using VB, VBA, C#, C/C++, Java, Python, PHP. | Maximal is now giving away for FREE development copies of MPL. We also offer free MPL for Academics and with solver purchase. |
| OML (Optimization and Modeling Library) | OML is a complete tool kit for linear and mixed-integer optimization as well as model management in a callable library. | ||
| OMP Plus | Strongly improved Parallel Branch & Cut. | Leading supply chain solution including planning, scheduling, promising, forecasting, cutting stock solving. | |
| OptiMax Component Library | New releases for Python, VB.Net, and CSharp with multiple new objects, methods and properties added for application dev. | OptiMax is an object-oriented Component Library, designed to help embed optimization models into end-user applications. | Maximal is now giving away for FREE development copies of MPL/OptiMax. We also offer free MPL for Academics and with solver purchase. |
| OptimJ | OptimJ GUI, a compiler-generated graphical user interface for rapid prototyping, easily extensible for stunning graphics. | Object-oriented modeling (models are expressed directly over the existing OO data model). Bulk data preprocessing primitives. | A powerful algebraic modeling language extending Java, giving direct access to all Java libraries and state-of-the-art development tools. |
| Oracle Crystal Ball Suite | Improvement in the optimization algorithm. | ||
| PICO | PICO is a parallel MIP solver that has scalable parallelism suitable for large-scale parallel computers. | ||
| Premium Solver Platform | Solver and method improvements for 2X to 10X+ faster solutions; 64-bit, multi-core processor, and cluster computing support | Convex and non-convex smooth nonlinear optimization, non-smooth optimization, global optimization, if/min/max linearization | Quickly solve every type and size conventional optimization problem in Excel. Expand w/plug-in engines and upgradeable to Risk Solver Platform. |
| Premium Solver Pro | Solver and method improvements for 2X to 10X+ faster solutions; 64-bit, multi-core processor, and cluster computing support | Convex and non-convex smooth nonlinear optimization, non-smooth optimization, and global optimization | A great first upgrade to the Basic Excel Solver. Capable of solving larger problems in less time with a flexible UI and in-product assistance |
| QMS | QMS is designed as a pedagogical tool that allows the student to focus on formulating Decision Science optimization problems and interpreting the results. | ||
| Risk Solver Platform | Solver and method improvements for 2X to 10X+ faster solutions; 64-bit, multi-core processor, and cluster computing support | Nonlinear/non-smooth optimization, global optimization, simulation optimization, stochastic programming, robust optimization | Most powerful, fastest, and easy to use Excel Solver upgrade for the best insight & decisions on the largest & most complex real-world problems |
| SAS | Use of named models, network simplex solver, crossover for interior point solver. | OPTMODEL language supports LP, MILP, QP, and NLP with unified syntax; custom solution algorithms are supported. Nonlinear optimization includes large-scale solvers and multi-start capability. | Fully integrated with other SAS analytics--statistics, data mining, forecasting, etc.--and with SAS strengths in data integration and business intelligence. |
| SCIP | framework, customizable via callbacks; reads more than 10 different input formats; HowTos, FAQ, mailing list, examples | SCIP is a standalone MIP solver and easily customizable branch-cut-and-price framework oriented towards the needs of OR/MS. | |
| Solver SDK Platform | Solver and method improvements for 2X to 10X+ faster solutions; 64-bit, multi-core processor, and cluster computing support | Smooth and non-smooth optimization, global optimization, monte carlo simulation, simulation optimization, load/solve .xlsx | Create models in your choice of platforms, deploy on desktops, servers, or in the cloud, and solve using the fastest and best solvers available |
| Solver SDK Pro | Introduced in 2011 | Smooth and non-smooth optimization, global optimization, monte carlo simulation, and simulation optimization | Create models in your choice of platforms using an object-oriented API and advanced solvers, and deploy on desktops, servers, or in the cloud |
| SoPlex | SoPlex is a C++ implementation of the revised simplex algorithm for Linear Programming. | ||
| SOPT (Smart Optimizer) 4.2 | Making an optimization progress file available for web-based applications. | Extended rounding and search capabilities are included to find feasible solutions to large-scale integer programs with automatically generated cuts. Search depth and times can be controled by user parameters. | Heuristic search algorithms are implemented to find integer feasible solutions quickly. Quadratic or nonlinear problems are solved fast by an interior-point algorithm. Integer variables can be handled with nonlinear constraints. |
| Vanguard Global Optimizer | No restrictions on linearity, discontinuity, stochasticity, and model characteristics. Designed for Grid Computing. | ||
| What'sBest! | Stochastic programming, K-Best solutions, Infeasibility analysis (LP, IP, QP, & NLP). | Automatic detection of problem type. | WhatˆásBest is a large scale optimization add-in for Excel; powerful enough for real world models and ideal for building models for clients. |
| XA | Multi-core support | Binary/Integer probing | Siginficant speed improvement |
| YALMIP | Bilevel programming | Geometric programming, Convex programming, Robust optimization, Multiparametric optimization, Global optimization, Sum-of-squares |
Linear Programming Software Survey Pages:
Introduction | Page 1 | Page 2 | Page 3 | Page 4 | Page 5 | Page 6| Page 7 | Page 8 | Page 9 | Page 10 | Page 11 | Page 12 | Page 13 | Vendor List | Accompanying Article