Abstract

The generation of models that represent real production systems has difficulties that vary from case to case. In general, the construction of these models is highly difficult and requires relevant specialized resources. Therefore, a general framework is presented that seeks to propose a series of steps to select models that represent the reality of a system using adequate resources for its construction and analysis. This proposal focuses on systems in which there is uncertainty in the key variables, with complex interaction between subsystems. There are also important differences between the perspectives of the system's behavior of the areas that manage it and the system's inputs have characteristics that are difficult to represent, among others. As a result of this proposal, it is expected to identify different model alternatives, specifying for each of them, the expected level of precision in their results and the resources necessary for their generation. In this way, it will be possible to select a type of model that meets the objective of the study with the most efficient use of resources.