Browse Prior Art Database

Method and System for Building Mathematical Optimization Model using Interactive Graphical Pipeline

IP.com Disclosure Number: IPCOM000237716D
Publication Date: 2014-Jul-04

Publishing Venue

The IP.com Prior Art Database

Abstract

A method is provided to model mathematical optimization models using interactive graphical pipeline. Modular graphical optimization components are designed to accelerate the optimization model development process. Optimization model developers can interactively drag and drop the modular components into a graphical canvas UI, and assemble them into an optimization pipeline, so as to build the full optimization model. Constraint templates harvested from previous modeling best practice are provided to lower the technical barrier and ensure the modeling quality. The optimization graphical pipeline is stored in a XML model. An automatic code generator is used to transform the XML model into the executable optimization programming codes. Model developer can further customize the model in programming mode. The customized code can be stored in private repository or public repository for future reuse. The disclosure also provides a mechanism for business users to define an optimization model quickly with minimal optimization knowledge.

This text was extracted from a PDF file.
This is the abbreviated version, containing approximately 28% of the total text.

Page 01 of 11

Method and System for Building Mathematical Optimization Model using Interactive Graphical Pipeline

Mathematical optimization is the selection of a best element (with regard to some criteria) from some set of available alternatives. In the simplest case, an optimization problem consists of maximizing or minimizing a real function by systematically choosing input values from within an allowed set and computing the value of the function. The generalization of optimization theory and techniques to other formulations comprise a large area of applied mathematics. More generally, optimization includes finding "best available" values of some objective function given a defined domain, including a variety of different types of objective functions and different types of domains.

To solve the real world problems using optimization techniques, optimization models need to be built based on the business requirements. Optimization tools, e.g. ILOG OPL Studio, LINGO, AIMMS, etc., are usually used to develop and maintain optimization models. It bridges the gap between mathematical model formulation and solving algorithms. The popular optimization modeling methods include algebraic modeling languages (e.g. MPL, OPL, GAMS, etc.) and general programming languages (e.g. java, c++, etc.). Generally, to develop an optimization using current available methods and tools, it requires highly skillful researcher (e.g. PhD in Operations Research), which results in the following drawbacks: 1. Niche users - The skillful operations researchers (usually PhD) are rare resource in the industry. This greatly limits the adoption of optimization techniques in the industry.

2. Steep learning curve - It usually takes long time for developers to master the current optimization tools to build proper optimization models.

3. Long development lead time - Development of an optimization model from scratch usually need long time to complete using the current optimization modeling approaches/tools.

4. Hard to control the quality of the optimization model - Optimization modeling is an art. In an optimization model development, usually there are different coding representations for the same business requirement. Very often, different coding representation will affect the model solving performance greatly.

Some modern optimization modeling tools, e.g. ILOG OPL Studio, LINGO, Frontline, AIMMS, etc., provide simple UIs in the integrated development environment (IDE) to help on the model construction. However, these features are not easy to use by the developers, especially junior developers, which results in low productivity and poor modeling quality.

1. ILOG OPL studio and LINGO provide IDEs for developers to program the optimization model using algebraic modeling languages. To construct a model in these IDEs, developers need to be very skillful in algebraic modeling languages. There is very few graphical UI features in OPL studio to help developers on the modeling process.

2. AIM...