New Version

ILOG OPL Development Studio 5.5

• Support for new ILOG CPLEX features:

• Generated solutions are collected and displayed in the problem browser, where values of both decision variables and objective functions can be compared.
• The performance tuning tool can be launched from the ILOG OPL independent development environment (IDE), and progress can be monitored. The recommended parameter settings are available as a standard ILOG OPL setting file. ILOG OPL Interfaces and Script are available through OPLRun.
• Parallel solving on multi-core and multi-CPU machines is possible. All ILOG CPLEX options are available, including the choice of new deterministic or traditional opportunistic parallel mode.

• Sorted and ordered sets—Sets can be kept in constructed-, sorted-, or reverse-sorted order.

• Improved integration of modeling language and scripting—Modeling functions can be used from scripting.

• Launch of OPLRun directly from the IDE—A run configuration can be launched as a separate process, IDE overhead is eliminated, and 64-bit mode execution is permitted.

• Labeled asserts—Debugging information has been improved.

• Improved organization of ILOG OPL examples—Variations on a model are now located in the same ILOG OPL project for easier comparison.

Previous Updates

ILOG OPL Development Studio 5.2

Support for ILOG CP Optimizer:

• ILOG CP Optimizer is a Constraint Programming (CP)-based optimization engine for sequencing, resource allocation and timetabling problems. ILOG CP Optimizer-based models can now be developed, debugged and tuned in the ILOD OPL integrated development environment (IDE). This allows you to model and solve problems using the latest in constraint programming techniques, or to decompose a problem and use ILOG CP Optimizer- and ILOG CPLEX-based models in sequence to solve them.
• The ILOG OPL modeling language has been extended to support the definition of ILOG CP Optimizer-based models, including arithmetic linear and non-linear constraints, logical constraints, and specialized constraints that define relationships between large sets of variables.
• The ILOG OPL time and memory profiler, the console, log and solve progress displays all support ILOG CP Optimizer

Ports:

• The Windows 32 bit port now supports Windows Vista
• Red Hat Enterprise Linux 5
• SUSE Enterprise Linux Server 10 (SLES 10)

ILOG OPL Development Studio 5.1

• Performance improvements
  Model generation in ILOG OPL is now on average 10 percent faster and requires 10percent less memory than ILOG OPL 5.0.
• External function calls
  It is now possible to make calls to external functions written in Java from scripting. This allows you to combine ILOG CPLEX with external algorithms and is useful for decomposition schemes and iterative solving procedures.
• Table loading
  Database tables can now directly be loaded into ILOG OPL arrays through the DBread statement
• Arrays of decision expressions
  Decision expressions (dexpr) was introduced in ILOG OPL 5.0 and ILOG OPL 5.1 now allows arrays of such expressions to be defined – resulting in compact, readable models and improved performance.
• New ports
  Windows 32 bit port now supports both Microsoft Visual Studio 2003 and 2005
  Windows 64 bit now available as a deployment port (no 64 bit IDE)

ILOG OPL Development Studio 5.0

• Improved model debugging
  Conflicts and possible relaxations are displayed for infeasible models, allowing quick identification and resolution of model inconsistencies. A mouse-click takes users to the point of conflict, and a display of relaxed constraints required for feasibility helps users identify and make corrections in the model.
• Logical constraints
  Logical constraints in the modeling language support more compact modeling and create easier model maintenance. As an alternative to manual linearization, which often requires using big-M constraints, constraints including “AND,” “OR,” “NOT,” “MIN,” “MAX” and “ABS” are now modeled and solved natively in ILOG OPL and ILOG CPLEX.
• Advanced flow control
  Scripting now supports warm start and direct control of the ILOG CPLEX matrix. This allows for iterative algorithms for better performance.
• Data integration and manipulation
  New tuple keys and tuple references provide automatic data consistency checking, better performance of model generation, and a more natural modeling approach.