Capacity-Insensitive Algorithms for Online Facility Assignment Problems on a Line

In the online facility assignment problem [Formula: see text], there exist text] servers on a metric space where each has an integer capacity and request arrives one-by-one. The task of algorithm is to irrevocably match current with one vacancies before next arrives. As special cases for we consider line, which denoted by latter case equidistant servers. this paper, perform competitive analysis above problems. natural generalization greedy grdy, introduce class algorithms called MPFS (Most Preferred Free Servers) show that any capacity-insensitive property, i.e., alg if text]-competitive when then general text]. By applying property derive matching upper lower bounds ratio grdy To investigate capability algorithms, at least Then, propose new idas (Interior Division Adjacent most best possible in all algorithms. We also give numerical experiments performance performs better than distribution sequences locality.