• Solver Systems
• FortMP
• CPLEX
• FortSP
• FortMP-MEX
• MOPS
• GUROBI
• MOSEK

 

Solver Systems - An Overview

Solvers are software to solve a wide variety of optimisation problems. Typically, such problems could vary from solving fuel blending combinations, portfolio optimisation problems, production planning, Asset and Liability management, risk management to logistics problems. The algorithms used in solvers are distinguished by their ability to solve large-scale linear, variable separable, mixed integer, quadratic and quadratic mixed integer problems. Solvers may be available independently or may be bundled as packages incorporating a number of optimization algorithms. Solvers are provided with preprocessor routines and facility for scaling and model reduction techniques in order to enhance the speed of operation. Solvers need to communicate with the various AMPL packages which will enable the transfer of the data seamlessly and efficiently. Typical solvers OptiRisk offers are FortMP, CPLEX, FortSP, FortMP- MEX, MOPS.

OptiRisk solvers are robust yet affordable. In addition, OptiRisk optimization experts would work with you to develop ‘bespoke’ decision support system to meet your enterprise decision optimization needs, if you choose as to be the service provider / partner. To increase the ROI (return on Investment) on your purchases, we offer training programs as well on (1) how to use our modelling tools and solver systems independently and (2) how to integrate our modelling tools and solver systems with your existing enterprise decision support system (DSS) or to develop a new one from grounds-up. 

 

FortMP: FortMP is a state of the art optimisation system designed to solve various mathematical programming modules which include large-scale linear, variable separable, mixed integer, quadratic and quadratic mixed integer problems. Primal and dual simplex algorithms with sparse matrices are the main algorithms used by FortMP. This optimisation system is available in diverse formats and works seamlessly with other modelling systems like AMPL, AMPL Studio and AMPL COM. FortMP is available either as a stand-alone program or as a static library or as a DLL enabling it to be embedded in other software environments. It is suitable for modularisation, finely tuned for serial or parallel platforms. Input data files may be in standard MPS (extended) layout or in a free-form equivalent.

… more

CPLEX: AMPL CPLEX solver interface supports linear, nonlinear, and mixed integer models with no built-in size limitations. This interface supports many of the features used by advanced solvers to improve performance and solution accuracy, such as piecewise-linear constructs, representation of network problems, and automatic differentiation of nonlinear functions.

… more

FortSP: FortSP is a solver for stochastic programming (SP) problems – Optimization problems under uncertainties. It is used as the solver for the SPInE modelling system. FortSP can also be used on its own to solve problems presented in SMPS format. It is available either as a standalone program or as a library with interface in the C programming language. It has a powerful plug-in system for connecting external solvers through the COIN-OR Open Solver Interface (OSI). FortSP is designed to solve both two-stage and multi-stage recourse problems and solving the class of problems commonly referred to as “Here and Now” problems such as Bender’s decomposition, variant of level decomposition and nested Bender’s decomposition. These algorithms make it possible to solve problems with large number of scenarios. FortSP can also formulate deterministic equivalents of two-stage problems with chance constraints and integrated chance constraints besides supports evaluating WS and EV models.

… more

FortMP-MEX :FortMP-MEX is an add-on to Matlab, which is an optimisation solver system capable of solving large-scale industrial applications ranging from supply chain to portfolio planning and asset liability management in finance. It provides a versatile way of solving problems by offering various alternatives of specifying a problem. FortMP provides an industrial strength optimiser. It provides a very transparent means for Matlab users to interact with an industrial strength solver. It solves large scale Linear Programming (LP) Problems using SSX and Barrier / IPM, mixed scale integer Programming (MIP) problems, Quadratic problems (QP), Quadratic Mixed Integer Programming problems (QMIP) . FortMP-MEX finds application in Portfolio Optimisation and Asset Liability Management.

… more

MOPS :MOPS is a high performance LP and MILP solver system based on state-of-the-art algorithms and their efficient implementation. It is a proven system used in large companies since 1990. It is equipped with advanced LP-preprocessing algorithms to remove redundancies in LP/IP-models. It has Primal and dual simplex algorithms with efficient LU-technology. It has Fast Interior point (barrier) algorithm with optimal basis identification and has sophisticated IP-Preprocessing with most known mathematical techniques to strengthen the LP-relaxation of the IP-model. Initial Integer solutions are possible by using various heuristic algorithms.

… more

GUROBI :The Gurobi Optimization Software is a unique complex mathematical solver system which can be used effectively to solve Linear as well as Mixed Integer Problems. It has the unique feature to exploit modern multi-core processors. Parallelism is included with every Gurobi license. The Gurobi Parallel MIP Optimizer is deterministic: two separate runs on the same model will produce identical solution paths.

… more

MOSEK :MOSEK Optimization Software is designed to solve large-scale mathematical optimization problems.

… more