Invited Session Fri.1.MA 144

Friday, 10:30 - 12:00 h, Room: MA 144

Cluster 22: Stochastic optimization [...]

Stochastic network design and reliability


Chair: Nedialko Dimitrov



Friday, 10:30 - 10:55 h, Room: MA 144, Talk 1

Paul Kantor
Layered screening at public events: Models and challenges

Coauthors: Endre Boros, Fred Roberts, Brian Thompson


Public events must screen against many threats. Stakeholders’ interests diverge - owners seek revenue, patron experience, and safety against minor disturbances, and major attacks. Patrons seek to enjoy the game, to drink, etc. Events are protected by layers of agents (county police, city police, stadium workers etc.) under distinct authorities, and each with its own screens. For each "violation'' there are several terminal remedies. A guest with a knife, or beer may be asked to surrender the object, or to go home. A cost matrix C(a,s;J) depends on the state of the patron (s=harmless; … etc.) and the terminal action imposed by the authorities (a=confiscate but admit; send home; arrest; etc.), as seen by stakeholder class J.
Given imperfect screens (metal detector; wanding; dogs; pat-down; etc.) we seek optimal routing rules for arriving patrons. Assignment to further screens, and choice of actions, depends on results of earlier screens. One screened for metal may be labeled (admit; wand; pat-down; etc.). We need throughput for all patron classes, with acceptable costs for all. We give a polytope approach, a dynamic
programming approach, rigorous results, and simulations.



Friday, 11:00 - 11:25 h, Room: MA 144, Talk 2

Christian Klaus
Increasing network reliability by introducing warehouses

Coauthor: Nedialko Dimitrov


Humanitarian assistance cargo is shipped by filling available space on regularly scheduled transportation routes. Regularly scheduled transportation can be modeled as a network consisting of source, destination, and transshipment nodes, where the edges represent the scheduled transportation routes. The edge capacities are discrete random variables, the space available on a particular mission. An empirical distribution of the capacities can be obtained from historical data.
Because of the stochastic nature of the network, an analysis of the network's reliability in shipping cargo is done by sampling. The analysis shows insufficient ability to deliver the cargo.
In order to increase the reliability of the network to ship cargo, we consider creating warehouses on selected transshipment nodes, where goods can be stored until space for further shipment is available. The introduction of the warehouses can be modeled using a network with multiple time layers.
We address the question of how to select the best warehouse. The problem can be modeled as a two stage stochastic program, where the first stage decision variables select warehouse locations, including storage capacity.



Friday, 11:30 - 11:55 h, Room: MA 144, Talk 3

Melih Çelik
The post-disaster debris clearance problem with uncertain debris amounts

Coauthors: Ozlem Ergun, Pinar Keskinocak


In this study, we focus on the clearance stage of post-disaster debris management process, which spans the first few days following the disaster, when clearance resources are extremely limited. Given that a set of roads in the network are blocked, the objective is to determine the road clearance sequence in each period so that the total expected penalty due to unsatisfied relief commodity demand over all periods is minimized. We assume that the amount of debris to be cleared is known for only a certain set of blocked roads, where as for the remaining roads, initial beliefs exist and are updated as clearance proceeds. For this problem, we formulate a Partially Observable Markov Decision Process (POMDP) model to find the optimal solutions, and explore performance bounds compared to the case of solving a deterministic model with expected debris amounts. Due to the high computational burden of applying the POMDP model, we propose a heuristic procedure based on sampling of alternative actions and possible observations of actual debris amounts. We test the performance of this procedure on randomly and structurally generated instances on certain types of graphs.


  Payday Loans In Texas. On the global pharmaceutical market this medicine was issued in 2003 by two companies - Eli Lilly and ICOS. Initially, permission to sell Cialis was obtained in Europe, Australia, New Zealand.