1. Introduction

Optimization technology is a technique used to find the best solution among many possibilities. It is one of the core tools in modern industrial and commercial decision-making and is widely used in industries such as power energy, industrial manufacturing, transportation logistics, retail, finance, cloud computing, and scientific research. Examples include designing the shortest project schedule, the lowest total cost power generation plan, the lowest risk level of capital allocation, the least harmful radiotherapy plan for healthy organizations, the most fuel-saving rocket launch plan, the shortest transportation route, etc. All these problems seeking the “optimal” solution are application scenarios of optimization technology.

In practice, people need to analyze the system structure, logical relationship, and desired goals of actual problems. Then, using functions, equations, inequalities, simulation programs, and other tools to formally describe them, this process is commonly referred to as Mathematical Modeling. The more complex the problem, the larger the optimization model obtained from mathematical modeling. Therefore, the complexity of many actual application optimization problems has far exceeded the range that general tools can handle. At this time, an Optimization Solver is needed to solve these problems and find the optimal solution.

MindOpt is an optimization solver suite. Centered around the MindOpt optimization solver, we have developed several software and platform services to meet various needs of users in intelligent decision-making scenarios and help enterprises reduce costs and increase efficiency. Our current products include the MindOpt optimization solver, MindOpt APL modeling language, MindOpt Tuner parameter tuner, and MindOpt optimization modeling platform.

  • MindOpt Optimization Solver is an efficient optimization solving software that currently supports solving linear programming, mixed-integer linear programming, convex quadratic programming, convex quadratically constrained programming, mixed-integer quadratic programming, mixed-integer quadratically constrained programming and semidefinite programming problems. For details on its usage and instructions, please refer to Overview.

  • MindOpt APL (MindOpt Algebraic Programming Language, MAPL) is a general optimization modeling language that establishes optimization models in a form closer to algebraic language. It supports the modeling of general linear, non-linear, and mixed-integer problems, and enables the utilization of over 20 different optimization solvers for problem-solving, thereby reducing the barrier to entry for modeling and offering users more choices. For its instructions, please refer to MAPL.

  • MindOpt Tuner is a tool for hyperparameter optimization of users’ optimization problems to achieve the best performance of the solver. For more introduction, please Contact Us.

  • MindOpt Studio is a cloud platform. We provide users with a cloud modeling optimization platform through cloud services, integrating a series of software and services including optimization solvers, optimization modeling languages, and parameter tuners, and using cloud elastic resources to eliminate the need for users to install and configure the environment. Additionally, users can learn modeling methods for different application problems through numerous examples provided within the optimization platform. For its usage, please Contact Us.

In addition, more products and capabilities are under development. We welcome everyone to continue following our progress.