Modeling by object-driven linear elemental relations: A user's guide for MODLER. Disk included (Q687360)

The MODLER is a software system for representing linear programming models and generating instances of models. The basic input to MODLER is a model file, and its basic output is a matrix file for optimizers and for other software systems, like ANALYZE, to provide computer-assisted analysis. One of the philosophies underlying the design of MODLER is the separation of models and instances. An instance can be obtained by reading data and by randomization, which enables instances to be generated without actual data. The aim of doing so is either for rapid prototyping, to see model properties for instances, or for experimentation with algorithms. MODLER is a language for representing linear programming models. The model is defined by objects and relations among its objects. These subdivide according to their properties and their roles in the model specification. The model file structure reflects this design philosophy. Data objects are constants, parameters, tables and sets. The parameter and the table can be specified with range list, which is used for data checking and serves the purpose of giving a numerical range for randomization. Decision objects are either binary variables, or continuous-valued variables. A decision object can have a domain restricted by relations between data objects. Relations in a model are of two types: conditions and constraints. Conditions include domain restrictions and admissible values. Arithmetic constraints are equations, bounds and limits. Logical constraints are represented by \(\text{IF}\dots\text{THEN}\) statements, which make them easier to read. The MODLER can be considered as a perspective field of research related to the work of \textit{A. Charnes} and \textit{W. W. Cooper} [`Management models and industrial applications of linear programming', Vols. I and II (New York 1961)] and \textit{F. Glover}, \textit{D. Klingman} and \textit{N. V. Phillips} [`Network models in optimization and their applications in practice' (New York 1992)].