Page 13

Linear Programming Survey
Table 13

Product New Features (since June 2015): Other Techniques (available as part of the package): Comments / Description:
AIMMS Forecasting Library, R integration, Collaborative Data Management, Improved Outer Approximation and Multi-Start, and more AIMMS Cloud Platform, Tableau Integration, Extended charting and workflow in WebUI, and more AIMMS is used by AirLiquide, GE, Heineken etc. to deploy optimization Apps across their organizations for better decision making by end users
AMPL New QuanDec interface builder creates a web-based, multi-user app for sharing, analyzing, and deploying your AMPL models. Integrated scripting language; nonlinear function library with automatic 1st & 2nd derivatives; user-defined functions AMPL is designed for quick development & reliable deployment. It combines a powerful modeling language with access to the most advanced solvers.
AMPL API Released a corrective version of the API with many bug fixes, these has been tested against a large number of examples. Each API has been implemented following the idioms of the target languages. This approach makes the tool developer friendly. A set of APIs to interact with AMPL: Java, MATLAB, C++, C#, Python, R.  they enable execution control, model instantiation and interrogation.
AMPLDev Generic and dynamic data instantiation using SQL and MDDB layer; followed by data visualisation and display of model outputs. Stochastic extensions of AMPL are included also retaining features of AMPL. It is bundled with  SP and SIP solver FortSP. Contains IDE, Data connectivity, modeling support for an established AML. Includes advanced features: workspace management and model exploration.
Analytic Solver Five major releases: Apache Spark, Power BI, Tableau, RASON support; simulation, data mining enhancements; cloud integration. Comprehensive analytics: Conventional and stochastic optimization, simulation and risk analysis, data mining, text mining. Best-selling Excel Solver upgrade now offers comprehensive analytics, integrates with Big Data, cloud versions, SDK, RASON modeling and REST API.
Analytica Optimizer   Visual hierarchical influence diagrams; Intelligent multi-D array abstraction; Uncertainty & sensitivity analysis. Add-on solver engines also available: Frontline engines (LSLP, LSGRG, LSSQP) Gurobi, Knitro, XPress, Mosek, OptQuest. New as of September 2016, enhanced each calendar quarter thereafter. Comprehensive analytics: Conventional and stochastic optimization, simulation and risk analysis, data mining, text mining. Leading cloud platform for predictive and prescriptive analytics (50,000 users Q1 2017) inter-operates with desktop Excel, Solver SDK and RASON.
Artelys Knitro MISQP algorithm, R interface, Specialized Least Square API, Hessian derivative-checker Derivative-free optimization, black box optimization Professional support provided by Artelys experts
Cbc (COIN-OR Branch and Cut)      
Clp (COIN-OR LP Solver)      
CMPL (<Coliop|Coin> Mathematical Programming Language) Plug-in for SolverStudio, Cmpl is bundled with the SolverStudio package, see: The CMPL distribution contains Coliop (IDE for Cmpl), pyCmpl and jCmpl (APIs for Python and Java). The package contains also CMPLServer, which is an XML-RPC-based web service for distributed and grid optimisation.
CoinMP Open-Source Solver CoinMP release 1.8 offers new library interfaces and algorithmic updates. Linux/UNIX versions with automake/configure support CoinMP is available as precompiled CoinMP.dll and (Linux/Unix) libraries that can be easily embedded into 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.
Dip (Decomposition in Inteeger Programming)   A decomposition-based solver and modeling language supporting column generation and advanced decomposition-based algorithms.  
FICO® Xpress Optimization Suite Parallel innovations: task based mip search, parallel crossover, parallel black-box and derivative evaluations. Rapid development of web-based optimization apps, no web programming needed. Integrates with Tableau® for visualization. Xpress is a comprehensive set of fast solvers with modeling and optimization solution building, deployment and management - on cloud and on prem.
FortSP Branch and cut addition to SIP processing capability. The enhanced solver has been tested on SSD long/short portfolio models. Benders method refined by using level decomposition and other regularization approaches. Models are set up via SMPS standard. FortSP uses L-shaped method to process single stage and two-stage stochastic programming models. The embedded solvers are: CPLEX, Gurobi, FortMP.
GAMS Mixed complementarity problems (MCP), Extended Mathematical Programs (EMP)  
GCG new and improved primal heuristics, cutting planes, and column management   GCG is a generic Branch-Cut-and-Price solver for mixed integer programs. It is based on SCIP and part of the SCIP Optimization Suite.
GENO 2.0 Now solves systems of linear/nonlinear equations; it also accommodates real discrete variables and specific search points.   GENO solves uni-/multi-agent, static/dynamic optimization problems, with or without equation constraints, for continuous or discrete solutions.
GLPK (GNU Linear Programming Kit)      
hsol New product Presolve, crash and advanced basis start Terms of open-source availability yet to be decided. Source is available on application
IBM Decision Optimization on Cloud IBM Decision Optimization on Cloud is a solve service on IBM HW with different VM sizes/access    
IBM ILOG CPLEX Optimization Studio model assistance, variability tester, automated benders decomposition, cloud solve service, object-oriented Python API modeling/solving for scheduling CPLEX Optimization Studio features a Development Env. for OPL and 2 engines, CPLEX (math progr.), CPOptimizer (constraint progr., scheduling)
IBM Decision Optimization Center IBM Decision Optimization Center contains IBM ILOG CPLEX Optimization Studio(Eclipse plugin),server component, enduser GUI    
JuMP Syntax improvements, user-defined nonlinear functions, semidefinite programming.   Fast and extensible open-source algebraic modeling language.
LINDO API AllDifferent constraint support. Improved performance on fixed charge/min batch size/cardinality constraints. Stochastic & Chance Constraints, Guaranteed Global, Nonlinear Least Squares Suite of fast callable solvers for creating customized linear, integer, nonlinear, quadratic, stochastic and global optimization applications.
LINGO AllDifferent constraint support. Improved performance on fixed charge/min batch size/cardinality constraints. Stochastic & Chance Constraints, Guaranteed Global, Nonlinear Least Squares Fast linear, integer, nonlinear, quadratic, stochastic and global solvers and a comprehensive modeling language with convenient data options.
LocalSolver Dynamic roadmap to deliver new features and performance improvements: two versions per year released since 2012. LocalSolver integrates many different optimization techniques automatically combined through learning during the search. LocalSolver is an all-terrain, all-in-one mathematical solver, for large-scale, real-life combinatorial and numerical optimization.
LPL Modeling System      
MATLAB and Optimization Toolbox Improved speed and robustness for mixed-integer, linear, quadratic and nonlinear solvers. Linear and nonlinear least squares and nonlinear equation solving. Combine Optimization Toolbox with MATLAB data preprocessing, predictive modeling, and graphics to build and deploy optimization applications.
MibS (Mixed Integer Bilevel Solver)   Supports solution of general mixed integer bilevel linear programs  
MOSEK Optimization Suite Version 8 was released in 2016.    
MPL Modeling System New MPL release 6.0, offers a large MPL C-API Callable Library with over 500 functions, and full support for Multi-threading New MPL REST Server allows MPL to be called from online web, mobile and tablet applications with support for JSON/XML data. Maximal offers full-size development copies of MPL for FREE, through our MPL Free Development and MPL Academic Programs.
OMP Plus   Leading integrated, interactive and intelligent Supply Chain Planning solution from strategic level to operational level.  
OpenSolver Improved non-linear support. Available for Macintosh as well as Windows. Better solver integration. Usability enhancements.   Open source Excel plug-in for building & solving optimization models in spreadsheets. Comes with open source solvers. Supports Gurobi & NEOS.
OpenSolver for Google Sheets Newly released product.   A plug-in for solving linear and integer programmes using Google Sheets. Supports multiple solvers.
OptiMax Component Library New release MPL Optimax with extensive new objects, full multi-threaded support, and ability to import JSON/XML data formats. With the new MPL REST Server, OptiMax  models can now easily be deployed on Web sites and Mobile/Tablet application platforms MPL OptiMax is an Object-Oriented Component Library, designed for embedding optimization models into end-user applications and online services.
ParaSCIP, FiberSCIP     Implemented by using Ubiquity Generator Framework(UG).(
Pulp     supported by and part of
Pyomo Releases 4.1, 4.2 , 4.3, 4.4, 5.0 and 5.1.  Added support for Python 3.6, support for new solvers and conda installation.   Pyomo is an open-source software package that supports a diverse set of capabilities for formulating and analyzing optimization problems.
RASON: Restful Analytic Solver Object Notation Major releases in March 2017, mid 2016, early 2017. Modeling language enhanced, higher performance in every Solver Engine. Monte Carlo simulation, simulation optimization, robust optimization, stochastic optimization, rich data mining, text mining. RASON models, embedded in JSON and solved in the cloud via a REST API, radically simplify the creation of optimization-based web and mobile apps.
SAS Optimization 8.1 New offering. MILP: distributed concurrent mode. Network optimization: 2 distrib. algorithms, distrib. BY-group processing.   Part of SAS Viya platform. Integrated with broad suite of products for descriptive and predictive analytics and data exploration/visualization.
SAS/OR 14.2 Solver algorithms performance improvements. Added features for network algorithms, CLP algorithm, and local search algorithm. Discrete-event simulation, critical path and resource-constrained scheduling, bill of material analysis. Integrated with full spectrum of SAS analytic products and solutions. Note: additional license required for distributed features.
SCIP,   available as package together with LP solver SoPlex, modeling language ZIMPL, generic branch-and-price solver GCG, massive parallel framework UG
Solver for Excel Online and Google Sheets New user interface tuned for tablets and browsers; automatic algebraic diagnosis of optimization models; new reports. Primarily for cloud use, but also works in latest desktop Excel versions. First cloud spreadsheet Solver (mid 2013),  the leading optimization tool for Excel Online and Google Sheets, with over 125,000 users in Q1 2017.
Solver SDK Major releases in 2016 and 2017; new support for R, Python and RASON; higher performance in every Solver Engine. Monte Carlo simulation, simulation optimization, robust optimization, stochastic linear programming. The only SDK/callable library that can open, modify and solve Excel and RASON optimization models, as well as models in 8 programming languages.
SolverStudio Now also supports Julia/JUMP   Use modelling languages such as AMPL, Julia/JUMP, PuLP, Pyomo & GAMS to build & solve optimization models within Excel.
SoPlex LP solution polishing, decomposition based dual simplex, persistent scaling    
The Gurobi Optimizer Support for multiple objectives, MIP solution pools and new general constraints. Significant python modeling enhancements. Compute Server w/client-server, queuing & fail-over capabilities. Instant Cloud w/easy instance launch & manage capabilities. Superior optimization algorithms, flexible interfaces, technical support from optimization experts, transparent pricing and flexible licensing.
UG: FiberSCIP and ParaSCIP     UG is a generic framework to parallelize SCIP in a distributed (ParaSCIP) or shared memory (FiberSCIP) computing environment.
Vanguard System Vanguard's optimzation capaiblity is integrated in our flagship planning and forecasting application, Vanguard IBP.   Vanguard is used by over 3,400 customers in 68 countries and for supply chain, finance, sales, human capital, IT, and advanced analytics.
What'sBest! AllDifferent constraint support. Improved performance on fixed charge/min batch size/cardinality constraints. Stochastic & Chance Constraints, Guaranteed Global, Nonlinear Least Squares What'sBest! is a large scale optimization add-in for Excel; powerful enough for real world models and ideal for building models for clients.
XA Professional Linear Programming System Raspberry Pi Edition in support of Internet Of Things (IoT)   Supports 32/64bit IoT micro-controllers
ZIMPL     GNU LGPL license

Linear Programming Software Survey Pages:
Introduction | Page 1 | Page 2 | Page 3 | Page 4 | Page 5 | Page 6Page 7 | Page 8Page 9 | Page 10 | Page 11 | Page 12 | Page 13 | Vendor List | Accompanying Article