Input data in a convenient format for optimization Apps

[tim-shea]

User story

As a product manager, I want to be able to input data in a format that is convenient, without having to write any complicated data transformation logic to make my data compatible with the Lava-Optimization Apps. For example, if I am using a scheduler app, I should be able to easily input tasks, resources (employees, vehicles, physical systems, etc.), and capabilities (restrictions/enablements on which resources can achieve which tasks) from a standard format such as Google ORTools.

Providing input data that describes an instance of my problem should be the default behavior, as opposed to the Optimization App generating a synthetic instance of the class of problem. When provided in the standard format, problem setup/conversion to a suitable Q matrix or solver configuration should be as fast as possible.

Conditions of satisfaction

• Lava-Optimization Apps accept a standard, easy-to-use, format for problem data.
• Apps expect the user to provide problem data as the default mode of operation. Generating synthetic data is a secondary mode.
• Conversion of standard format input data to internal/native problem structure is as fast as possible and ideally can be done in much less time than it takes the solver to solve the problem.