Zhouxing Su

(21% acceptance rate) Solutions for the optimal camera placement (OCP) and unicost set covering problem (USCP), which guarantee that every sampling point in the target area is covered and minimizes the number of cameras.

(12.6% acceptance rate) Solutions for the single-vehicle, multi-period, single-depot, deterministic-consumption classical inventory routing problem, which guarantee that no customer will run out of stock and minimize the sum of transportation cost and inventory holding cost.

(12.6% acceptance rate) Solutions for the classical p-center problem, which determine the location of a given number of centers to serve multiple clients, and minimize the serving radius while keeping each client under the covered area of any centers. This problem can be applied to the location of base transceiver stations in cellular network, datacenters, or ballot collection sites.

Solutions for the inventory routing problem in the final phase of ROADEF2016. It is able to generate an accurate-to-minute schedule which involves over 10 drivers, 10 trailers, 300 customers and lasts over 30 days, which guarantees that no customer will run out of stock and no drivers will work overtime in the planning horizon and minimizes the logistic ratio.

Solutions for the nurse rostering problem in INRC2. Unlike classic problems in the literature, it comes with immutable histories and short-term predictions instead of overall information, which is much more adaptive to real world applications.