Research Article
A Two-Stage Method of Dimensioning and Scheduling Service Providers under Patient Demand Uncertainty
(1) | X : set of data points | (2) | k: number of clusters | (3) | Randomly initialize 1 cluster centers(centroids) from X | (4) | Select k-1 cluster centers(centroids) systematically with a probability that is proportional to their contribution to the overall error from X | (5) | Repeat | (6) | Expectation: assign each point to its closest centroid. | (7) | Maximization: the mean of all points belonging to each cluster specifies the new centroid | | Until the centroid positions converge to a constant value |
|