| 摘要 |
<p>The invention provides a method for optimizing a Business Process Management (BPM) model which couples mixed integer programming and discrete event simulation. The BPM is associated with a BPM model and a plurality of key performance indicators (KPI) depending on a control parameter vector comprising a plurality of control parameters. The method comprises separating the control parameters into a linear control parameter vector (X) and a non-linear control parameter vector (X'), iteratively calculating a set of candidate values for the control parameter vector, each candidate value being determined for a preselected value of the non-linear control parameters (X') from the execution of a mixed integer program (MIP) model associated with the business BPM model, the plurality of key performance indicators (KPI) and the preselected value of the non-linear control parameters (X'), and adjusting the BPM model based on at least one of said candidate values for the control parameter vector.</p> |
