Figure 4

Flowchart representing the algorithm for viable volume estimation, and the acquisition of a set of viable parameter points. A set of viable parameter points found by OEAMC and MEBS (uppermost set of blue points in the figure) which are nonuniformly distributed over the whole viable space (area covered by the red curve in the figure) seeds the algorithm. Then, the method groups these points into k clusters (k = 3 in the hypothetical example shown), and calculates the ellipsoids with minimum volume that enclose the points in each cluster (11). After that, the algorithm performs a Monte Carlo integration of every ellipsoid (the intersections between ellipsoids are sampled only once) (12, 13). The result of the algorithm is a set of uniformly distributed viable parameter points (bottom set of blue points in the figure), from which the viable volume can be estimated.