Classical facility location models like the P-median problem (PMP) and uncapacitated fixed-charge (UFLP) implicitly assume that, once constructed, facilities chosen will always operate as planned. In reality, however, “fail” from time to due poor weather, labor actions, changes of ownership, or other factors. Such failures may lead excessive transportation costs customers must be served much farther than their regularly assigned facilities. this paper, we present for choosing locations...

We present an end-to-end framework for solving the Vehicle Routing Problem (VRP) using reinforcement learning. In this approach, we train a single model that finds near-optimal solutions problem instances sampled from given distribution, only by observing reward signals and following feasibility rules. Our represents parameterized stochastic policy, applying policy gradient algorithm to optimize its parameters, trained produces solution as sequence of consecutive actions in real time,...

Problem definition: The beer game is widely used in supply chain management classes to demonstrate the bullwhip effect and importance of coordination. a decentralized, multiagent, cooperative problem that can be modeled as serial network which agents choose order quantities while cooperatively attempting minimize network’s total cost, although each agent only observes local information. Academic/practical relevance: Under some conditions, base-stock replenishment policy optimal. However,...

The two most widely considered measures for optimization under uncertainty are minimizing expected cost and worst-case or regret. In this paper, we present a novel robustness measure that combines the objectives by while bounding relative regret in each scenario. particular, models seek minimum-expected-cost solution is p-robust; i.e., whose no more than 100p% We p-robust based on classical facility location problems. solve both problems using variable splitting, with Lagrangian subproblem...

article Free AccessA Linear Time Algorithm for Deciding Subject Security Authors: R. J. Lipton Department of Computer Science, Yale University, 10 Hillhouse Avenue, New Haven, CT CTView Profile , L. Snyder Authors Info & Claims Journal the ACMVolume 24Issue 3July 1977 pp 455–464https://doi.org/10.1145/322017.322025Published:01 July 1977Publication History 131citation1,015DownloadsMetricsTotal Citations131Total Downloads1,015Last 12 Months63Last 6 weeks7 Get Citation AlertsNew Alert...

This paper proposes a Real-Time Pricing (RTP)-based power scheduling scheme as demand response for residential usage. In this scheme, the Energy Management Controller (EMC) in each home and service provider form Stackelberg game, which EMC who schedules appliances' operation plays follower level sets real-time prices according to current usage profile leader game. The sequential equilibrium is obtained through information exchange between them. Simulation results indicate that our can not...

We study an integrated supply chain design problem that determines the locations of retailers and assignments customers to minimize expected costs location, transportation, inventory. The system is subject random disruptions may occur at either supplier or retailers. Analytical numerical studies reveal effects these on retailer customer allocations. In addition, we demonstrate numerically cost savings from considering phase (rather than tactical operational phase) are usually significant.

We consider mechanisms to optimize electricity consumption both within a home and across multiple homes in neighborhood. The are assumed use energy management controllers (EMCs) control the operation of some their appliances. EMCs, which feature emerging SmartGrid, prices user preferences power usage home. first present simple optimization model for determining timing appliance take advantage lower rates during off-peak periods. then demonstrate, using simulation, that resulting solution may...

The newsvendor problem is one of the most basic and widely applied inventory models. If probability distribution demand known, can be solved analytically. However, approximating not easy prone to error; therefore, resulting solution may optimal. To address this issue, we propose an algorithm based on deep learning that optimizes order quantities for all products features data. Our integrates forecasting inventory-optimization steps, rather than solving them separately, as typically done,...

The Folklore is replete with stories of "secure" protection systems being compromised in a matter hours. This quite astounding since one not likely to claim that system secure without some sort proof support the claim. In practice, provided and reason for this clear: although primitives are apparently simple, they may potentially interact extremely complex ways. Vague informal arguments, therefore, often overlook subtleties an adversary can exploit. Precision merely desirable systems, it mandatory.