Jij Inc. (Headquarters: Bunkyo-ku, Tokyo; CEO: Yu Yamashiro) has released a new Open Source Software (OSS) called "OMMX" (Open Mathematical prograMming eXchange, pronounced "omyx") on July 17, 2024. This tool is designed as an open ecosystem to support developers and researchers in mathematical programming and optimization. OMMX aims to simplify complex optimization workflows and facilitate data exchange between different tools, thereby improving productivity for all players in the field of mathematical optimization.Go to the repository: https://github.com/Jij-Inc/ommxCurrent State of Mathematical Optimization Experiments and Challenges OMMX SolvesBackground and ChallengesThe field of mathematical optimization is facing increasingly complex real-world problems. Many practical optimization problems involve a mix of discrete and continuous variables with a vast number of variables, making it difficult to solve effectively with a single solver. While advanced technologies such as quantum and Ising optimization computations hold great potential, their current application range is limited, and there are still challenges in full-scale implementation in real-world scenarios. Furthermore, the field of mathematical optimization has diverse tools and platforms, each adopting its own data format, making data exchange and integration between different tools difficult. This lack of interoperability significantly reduces the efficiency of research and development. Additionally, it's challenging to maintain the relationship between optimization models and their solutions after storing them in files or databases, and there's a lack of mechanisms to efficiently manage large amounts of data and reference them appropriately when needed.Problems OMMX SolvesOMMX provides a comprehensive solution to these challenges:Standard Data Exchange Format: Establishes a language and framework-independent standard data exchange format, enabling seamless data exchange between different tools and platforms.Metadata and Reference System: Introduces metadata and a reference system, allowing efficient management while maintaining relationships between data. This promotes reproducible experiments and effective collaboration between different teams.Support for Hybrid Algorithms: Accelerates the development and implementation of quantum-classical hybrid algorithms, supporting the creation of innovative solutions for complex optimization problems.Streamlined Development Process: Realizes seamless integration between different tools, significantly improving the efficiency of the entire development process.Scalability and Flexibility: Provides scalability and flexibility that allow user-defined annotations and customization, addressing diverse needs.Key Features of OMMXOMMX elevates mathematical optimization workflows to a new level by defining two types of data:OMMX Message: Enables storage of optimization models and solutions independent of language or framework, based on Protocol Buffers data schema.OMMX Artifact: Realizes storage of data with metadata and exchange as container images, based on OCI (Open Container Initiative) artifact-based packaging and distribution specifications.Use CasesOMMX plays a crucial role in the mathematical optimization research process, enhancing productivity in modeling, data management, solver integration, and interoperability.OptOps Cycle consists of seven key stages in the optimization process, which OMMX supports throughout:Design: Create hybrid algorithms capable of addressing diverse optimization problems.Development: Provide an efficient environment for implementing designed algorithms.Testing: Benchmark and evaluate the performance of developed algorithms against various optimization problems.Deployment: Apply verified algorithms to real-world problems and conduct proof-of-concept experiments.Operations: Operate algorithms in actual industrial settings and monitor their performance.Optimization: Continuously improve algorithm performance based on operational data.How to UseTo run notebooks locally, you need to install the required packages:# Optional: create a virtual environmentpython -m venv .venv && source .venv/bin/activate# Install required packages (including Jupyter)pip install -r requirements.txt# Start Jupyterjupyter labTutorialsDetailed tutorials are provided to facilitate the introduction and utilization of OMMX. Tutorials for OMMX Message, OMMX Artifact, and Cookbook are available in Jupyter Notebook format, accessible via Binder and Google Colab.Community and SupportActive community support is essential for the development and dissemination of OMMX. Discussions and contributions on the GitHub repository are welcome.Developer Comment"We are pleased to release OMMX today. We hope this tool will be widely adopted as a standard data exchange format in the field of mathematical optimization, significantly improving the efficiency of researchers and developers' work. OMMX is provided under a dual license of Apache License 2.0 and MIT License, allowing for various uses including commercial applications. As an open-source project, anyone can participate in improving the source code and adding features. We plan to continue improving and expanding functionality, so we look forward to your feedback and contributions.Stay tuned for the future development of OMMX!"Go to the repository: https://github.com/Jij-Inc/ommxThis project was supported by the Cabinet Office's Strategic Innovation Promotion Program (SIP) "Promoting Social Implementation of Advanced Quantum Technology".Development of Quantum Hybrid Optimization Algorithm Foundation: https://www.qst.go.jp/site/bridge/r5-04-bridge-r6.htmlAbout BRIDGE ProjectBRIDGE (Beyond-5G Realization and Demonstration of Innovative Gen-beyond technologies for high-performance En-to-end systems) is a significant initiative aimed at advancing quantum technology implementation in society. Here are the key aspects of the BRIDGE project:Objective: The primary goal is to promote the social implementation of quantum technologies, with a particular focus on realizing practical applications of quantum and Ising optimization computation technologies.Support: It is implemented as part of the Cabinet Office's Strategic Innovation Promotion Program (SIP) "Promoting Social Implementation of Advanced Quantum Technology".Relation to OMMX: OMMX was developed as part of the BRIDGE project, serving as a schema for mathematical models. It aims to establish a unified standard for mathematical models in optimization.Key Initiatives: Development of a quantum hybrid optimization algorithm foundation Application of quantum and Ising optimization technologies to real-world problems Promotion of market creation and startup businesses in quantum technologyDistinctive Feature: The project aims to build algorithm development capabilities that enable easy implementation of tight coupling between conventional computers and quantum/Ising optimization platforms.The BRIDGE project represents a crucial effort to accelerate the practical application and social implementation of quantum technologies. OMMX stands as one of the concrete outcomes of this initiative, contributing to the broader goals of advancing quantum computing and optimization in real-world scenarios.