Multi Disciplinary Optimization (MDO) is a particular case of black-box optimization with multiple intrinsically interconnected computational blocks widely used in complex engineering scenarios. We present a process implementing the Individual Discipline Feasible approach for handling an MDO problem with two disciplines.
This workflow models (a simplified version of) a process generating a timetable of lectures at a university.
This example has been generated with the help of the Department Electrical Mechanical and Management Engineering (DIEGM) of Udine University (Italy). Professor A. Schaerf is an expert in this field both for the “human” part of the process (contacting lecturers, checking room availabilities, interacting with the secretary office, etc) and for the computational part (scheduling algorithms, databases, etc).
This is the C.1.1 test from MIWG suite, drawn for the 2015 Execution Demo in Berlin. In the original version of the model, there are a lot of extension elements needed to actually automate the process on some execution engines. Here we post only the pure BPMN model, as produced for the “export” test”. The model has three User Tasks that interact with a human, and one Service Task that interacts with a machine, usually a web service.
This A.3.0 MIWG test introduces two interesting elements: the Sub-Process and Boundaries.
The Sub-Process can be collapsed into a simple task-like structure in order to hide its possibly complex process, as seen in the preview image.
Since this test is not an executable BPMN model, the Sub-Process is actually empty. Two Boundary elements are placed at its perimeter : an Intermediate Boundary Non-Interrupting Message Event and an Intermediate Boundary Interrupting Escalation. Please refer to BPMN specifications for a detailed description of these elements.