Optimization Direct at INFORMS, Las Vegas 2-4 April, 2017
Technology Workshop: CPLEX Optimization Studio Modeling, Theory, Best Practices and Case Studies
Sunday April 2nd
Recent advancements in Linear and Mixed Programing give us the capability to solve larger Optimization Problems. CPLEX Optimization Studio solves large-scale optimization problems and enables better business decisions and resulting financial benefits in areas such as supply chain management, operations, healthcare, retail, transportation, logistics and asset management. In this workshop using CPLEX Optimization Studio we will discuss modeling practices, case studies and demonstrate good practices for solving Hard Optimization Problems. We will also discuss recent CPLEX performance improvements and recently added features.
/ Alkis Vazacopoulos: Introduction & Case Studies, Combining Predictive and Prescriptive Analytics
/ Vincent Beraudier: Develop a Data Science project for a Marketing campaign planning with Python and CPLEX.
/ Robert Ashford: Recent computational experience with ODH-CPLEX
/ Yiannis Gamvros: Industrial Maintenance Scheduling: Challenges, Solution, Benefits
Technology Tutorial: An Introduction to ODH-CPLEX and Recent Computational Results
Monday, April 3, 9:10-10am
Location: Octavius 21
ODHeuristics is a general purpose program built on CPLEX for obtaining good feasible solutions to MIPs. It is intended for use on large scale MIP models, many of which are so computationally onerous that it is not possible to raise the best bound at all beyond the root solve. ODHeuristics is a general purpose program built on CPLEX for obtaining good feasible solutions to such MIPs.
Whilst these good solutions are useful they do not provide the optimality guarantee that many users require. ODH-CPLEX is the CPLEX optimizer in which ODHeuristics is embedded using the standard CPLEX API. On computers with many cores, it delivers the benefits of ODHeuristics whilst using CPLEX to provide optimality measures (the 'gap'). Providing good solutions early can accelerate the CPLEX solve.
On small scale test sets such as MIPLIB2010, ODH-CPLEX performs on average as well as CPLEX alone, i.e. the benefits of ODHeuristics compensate for the resources spent on it.
On large scale MIPs it provides good solutions and optimality measures that are often beyond the reach of traditional optimization methods.
It is designed for scheduling problems but works for any MIP which has a reasonable number of integer feasible solutions. It has been deployed effectively on packing problems, supply chain and telecoms as well as scheduling applications.
This talk reviews ODH-CPLEX performance on standard test sets and on large scale user MIPs. Features of models which suggest that ODH-CPLEX might work well are identified and the benefits of parallelism explored.