State Task Network for the scheduling of multiproduct batch plants.

STN is a scheduling model that can be applied to a large number of batch processes that are specified by recipes, it is concerned with short-term scheduling in which demands of products are specified at various points in time in the form of deadlines.

The discrete-time MILP model that is used was developed by Kondili et al. (1993), and has the following three major capabilities:

  1. Assignments of equipment to processing tasks need not to be fixed.
  2. Variable size batches can be handled with the possibility of mixing and splitting.
  3. Different intermediate storage and transfer policies can be accommodated as well as limitations of resources.

The major assumption in this model is that the time domain can be discretized in intervals of equal size. This network has two types of nodes: (a) state nodes that correspond to feeds, intermediates, and final products; and (b) task nodes that represent processing steps.

The continuous-time MILP that is used was developed by Maravelias and Grossmann (2003), and also uses the State-Task Network as the underlying representation. However, the time domain is represented by variable time points that eliminates the need of fixed time intervals. The specification of the number of time points, however, is non-trivial.

STN was developed by Gabriela Garcia-Ayala under the supervision of Ignacio E.Grossmann.