Covering shadows with a smaller volume

For n?2 a construction is given for convex bodies K and L in Rn such that the orthogonal projection Lu onto the subspace u^⊥ contains a translate of Ku for every direction u, while the volumes of K and L satisfy Vn(K)>Vn(L).A more general construction is then given for n-dimensional convex bodies K and L such that each orthogonal projection Lξ onto a k-dimensional subspace ξ contains a translate of Kξ, while the mth intrinsic volumes of K and L satisfy Vm(K)>Vm(L) for all m>k.For each k=1,…,n, we then define the collection Cn,k to be the closure (under the Hausdorff topology) of all Blaschke combinations of suitably defined cylinder sets (prisms).It is subsequently shown that, if L∈Cn,k, and if the orthogonal projection Lξ contains a translate of Kξ for every k-dimensional subspace ξ of Rn, then Vn(K)?Vn(L).The families Cn,k, called k-cylinder bodies of Rn, form a strictly increasing chain

Cn,1⊂Cn,2⊂?⊂Cn,n−1⊂Cn,n,     

The stochastic location model with risk pooling

In this paper, we present a stochastic version of the location model with risk pooling (LMRP) that optimizes location, inventory, and allocation decisions under random parameters described by discrete scenarios. The goal of our model (called the stochastic LMRP, or SLMRP) is to find solutions that minimize the expected total cost (including location, transportation, and inventory costs) of the system across all scenarios. The location model explicitly handles the economies of scale and risk-pooling effects that result from consolidating inventory sites. The SLMRP framework can also be used to solve multi-commodity and multi-period problems.     

Retrial queue with discipline of adaptive permanent pooling

A novel customer service discipline for a single-server retrial queue is proposed and analysed. Arriving customers are accumulated in a pool of finite capacity. Customers arriving when the pool is full go into orbit and attempt to access the service later. It is assumed that customers access the service as a group. The size of the group is defined by the number of customers in the pool at the instant the service commences. All customers within a group finish receiving the service simultaneously. If the pool is full at the point the service finishes, a new service begins immediately and all customers from the pool begin to be served. Otherwise, the customer admission period starts. The duration of this period is random and depends on the number of customers in the pool when the admission period begins. However, if the pool becomes full before the admission period expires, this period is terminated and a new service begins. The system behaviour is described by a multi-dimensional Markov chain. The generator and the condition of ergodicity of this Markov chain are derived, and an algorithm for computing the stationary probability distribution of the states of the Markov chain is given. Formulas for computing various performance measures of the system are presented, and the results of numerical experiments show that these measures essentially depend on the capacity of the pool and the distribution of the duration of the admission period. The advantages of the proposed customer service discipline over the classical discipline and the discipline in which customers cannot enter the pool during the service period are illustrated numerically.     

Error-correcting pooling designs associated with some distance-regular graphs

Motivated by the pooling designs over the incidence matrices of matchings with various sizes of the complete graph K_2n considered by Ngo and Du [Ngo and Du, Discrete Math. 243 (2003) 167–170], two families of pooling designs over the incidence matrices oft-cliques (resp. strongly t-cliques) with various sizes of the Johnson graph J(n,t) (resp. the Grassmann graph J_q(n,t)) are considered. Their performances as pooling designs are better than those given by Ngo and Du. Moreover, pooling designs associated with other special distance-regular graphs are also considered.     

Optimal design of efficient acoustic antenna arrays

Minimax optimal design of sonar transducer arrays can be formulated as a nonlinear program with many convex quadratic constraints and a nonconvex quadratic efficiency constraint. The variables of this problem are a scaling and phase shift applied to the output of each sensor. This problem is solved by applying Lagrangian relaxation to the convex quadratic constraints. Extensive computational experience shows that this approach can efficiently find near-optimal solutions of problems with up to 391 variables and 579 constraints. This work was supported by ONR Contracts N00014-83-C-0437 and N00014-82-C-824.     

Economical experiments: Bayesian efficient experimental design

We propose and implement a Bayesian optimal design procedure. Our procedure takes as its primitives a class of parametric models of strategic behavior, a class of games (experimental designs), and priors on the behavioral parameters. We select the experimental design that maximizes the information from the experiment. We sequentially sample with the given design and models until only one of the models has viable posterior odds. A model which has low posterior odds in a small collection of models will have an even lower posterior odds when compared to a larger class, and hence we can dismiss it. The procedure can be used sequentially by introducing new models and comparing them to the models that survived earlier rounds of experiments. The emphasis is not on running as many experiments as possible, but rather on choosing experimental designs to distinguish between models in the shortest possible time period. We illustrate this procedure with a simple experimental game with one-sided incomplete information.We acknowledge the financial support from NSF grant #SES-9223701 to the California Institute of Technology. We also acknowledge the research assistance of Eugene Grayver who wrote the software for the experiments.     

The design of an efficient load balancing algorithm employing block design

In order to maintain load balancing in a distributed system, we should obtain workload information from all the nodes on network. This processing requiresO(v ²) communication overhead, wherev is the number of nodes. In this paper, we present a new synchronous dynamic distributed load balancing algorithm on a (v, k+1, 1)-configured network applying a symmetric balanced incomplete block design, wherev=k ²+k+1. Our algorithm needs only $O(v\sqrt v )$ communication overhead and each node receives workload information from all the nodes without redundancy. Therefore, load balancing is maintained since every link has the same amount of traffic for transferring workload information.     

The design of an efficient elevator operating system

A method is developed for finding an efficient operating policy for an office building elevator system. The method was applied to a particular eleven story building in which there were four elevator shafts. A queuing model was formulated in which the characteristics of passenger arrivals and destinations were time variable. The objective involved the minimization of the weighted sum of ratios of actual to minimum possible travel time between all pairs of floors. Simulation was used to analyze several logical routing policies for each of two methods in which demand information was used to alter the elevator route. The best policy was found to be almost twice as efficient as most of the other policies which were studied and over 25 times more efficient than another seemingly logical operating policy.     

An efficient algorithm for optimal design of area traffic control with network flows

An equilibrium network design (EQND) is a problem of finding the optimal design parameters while taking into account the route choice of users. This problem can be formulated as an optimization by taking the user equilibrium traffic assignment as a constraint. In this paper, the methods solving the EQND problem with signal settings are investigated via numerical calculations on two example road networks. An efficient algorithm is proposed in which improvement on a locally optimal search by combining the technique of parallel tangents with the gradient projection method is presented. As it shows, the method combines the locally optimal search and globally search heuristic achieved substantially better performance than did those other approaches.     

Allocating the gains from resource pooling with the Shapley Value

To make their cost structure more efficient, firms often pool their critical resources: small divisions of a large firm may negotiate a joint contract to benefit from volume discounts; or firms may outsource their call centres to an independent provider who is able to increase utilization by reducing variability since demand is now pooled. Since pooling demand reduces total joint costs, an immediate question is how the realized savings should be shared. We model the problem as a cooperative game and use the resulting allocation schemes to distribute the savings. One popular scheme is the Shapley Value, which always exists and, we show, represents each player's incremental value to the pool. When the pooled savings depend on the sum of each player's demand, we label the game coalition symmetric and propose, for those games, an algorithm that makes pseudo-polynomial the computation of the Shapley Value.     

Allocation of jobs and identical resources with two pooling centers

We examine the resource allocation problem of partitioning identical servers into two parallel pooling centers, and simultaneously assigning job types to pooling centers. Each job type has a distinct Poisson arrival rate and a distinct holding cost per unit time. Each pooling center becomes a queueing system with an exponential service time distribution. The goal is to minimize the total holding cost. The problem is shown to be polynomial if a job type can be divided between the pooling centers, and NP-hard if dividing job types is not possible. When there are two servers and jobs cannot be divided, we demonstrate that the two pooling center configuration is rarely optimal. A heuristic which checks the single pooling center has an upper bound on the relative error of 4/3. The heuristic is extended for the multiple server problem, where relative error is bounded above by the number of servers.      

Large-scale standard pooling problems with constrained pools and fixed demands

We present a new variant of the standard pooling problem in which demands are fixed and there are specific constraints on the intermediate pool. We propose a new formulation composed of proportion-flow variables, and we design an exact branch and bound algorithm by combining existing algorithms. Difficult instances have been generated to demonstrate the efficiency of our method, and our results are compared with those of Couenne, a generic MINLP solver.     

Value of capacity pooling in supply chains with heterogeneous customers

We investigate the value of pooling capacity in supply chains that serve product demands of different variabilities. We build and analyze models that integrate production queuing models with base stock inventory systems serving demands with different inter-arrival time distributions. The first model combines hyperexponential and exponential demand inter-arrival time distributions. Exact analysis of the model allows us to develop insights into the impact of the difference in demand variabilities on the value of pooling capacity. Simulation experiments allow us to validate these insights for more general settings. We then find one special case that combines exponential and deterministic demand arrivals with deterministic service, where pooling capacity results in increasing the total cost.     

An efficient strategy for reliability-based multidisciplinary design optimization of twin-web disk with non-probabilistic model

The twin-web disk holds big promise for increasing efficiency of the aircraft engine. Its reliability-based multidisciplinary design optimization involves several disciplines including fluid mechanics, heat transfer, structural strength, and vibration. The solution to this optimization problem requires three-loop calculations including loops for optimization, reliability, and interdisciplinary consistence often making its computational cost unacceptably high. The lack of sufficient amount of probabilistic data, especially for this brand-new turbine disk, makes matters worse. In this paper, the non-probabilistic uncertain variables are described by an evidence theory-based fuzzy set method, which we extend to general structure of uncertain data. We also propose two modifications of the active learning kriging model: one of them for the purpose of optimization with respect to the distance from the optimum point and another one for the purpose of assessing reliability by introducing the importance concept. Applications of these two modifications are demonstrated in this paper. Finally, a multi-adaptive learning kriging strategy for non-probabilistic reliability-based multidisciplinary design optimization of twin-web disk is proposed to improve its power efficiency and reliability in a computationally effective way.     

Risk pooling strategy in a multi-echelon supply chain with price-sensitive demand

This paper studies a nonstationary inventory and pricing problem. We consider a two-echelon supply chain with one supplier and two retailers, in which the supplier carries all inventory to supply the retailers. Both the reserved and pooled inventory systems are analyzed. Results with normally distributed demands are compared. Assuming the random demand at each retailer is price-sensitive, we further consider the cases when the retailers have and do not have service level requirements. We start with analyzing inventory and pricing strategies for the supplier in a one-period scenario. Then we extend our analysis to both the backlogging and lost-sale scenarios in an infinite planning horizon. The first author’s research is sponsored by Grant No. 70502009 and No. 70432001 of the Chinese National Natural Science Foundation and the second author’s research is sponsored by Grant #W911NF-04-D-0003 of the US Army Research Office and Grant #DMI-0553310 of the US National Science Foundation.     

An efficient approximation algorithm for the survivable network design problem

The survivable network design problem (SNDP) is to construct a minimum-cost subgraph satisfying certain given edge-connectivity requirements. The first polynomial-time approximation algorithm was given by Williamson et al. (Combinatorica 15 (1995) 435–454). This paper gives an improved version that is more efficient. Consider a graph ofn vertices and connectivity requirements that are at mostk. Both algorithms find a solution that is within a factor 2k – 1 of optimal fork 2 and a factor 2 of optimal fork = 1. Our algorithm improves the time from O(k ³n⁴) to O ). Our algorithm shares features with those of Williamson et al. (Combinatorica 15 (1995) 435–454) but also differs from it at a high level, necessitating a different analysis of correctness and accuracy; our analysis is based on a combinatorial characterization of the redundant edges. Several other ideas are introduced to gain efficiency. These include a generalization of Padberg and Rao's characterization of minimum odd cuts, use of a representation of all minimum (s, t) cuts in a network, and a new priority queue system. The latter also improves the efficiency of the approximation algorithm of Goemans and Williamson (SIAM Journal on Computing 24 (1995) 296–317) for constrained forest problems such as minimum-weight matching, generalized Steiner trees and others. © 1998 The Mathematical Programming Society, Inc. Published by Elsevier Science B.V.A preliminary version of this paper has appeared in the Proceedings of the Third Mathematical Programming Society Conference on Integer Programming and Combinatorial Optimization, 1993, pp. 57–74.Research supported in part by NSF Grant No. CCR-9215199 and AT & T Bell Laboratories.Research supported in part by Air Force contracts AFOSR-89-0271 and F49620-92-J-0125 and DARPA contracts N00014-89-J-1988 and N00014-92-1799.This research was performed while the author was a graduate student at MIT. Research supported by an NSF Graduate Fellowship, Air Force contract F49620-92-J-0125, DARPA contracts N00014-89-J-1988 and N00014-92-J-1799, and AT & T Bell Laboratories.     

A construction of pooling designs with surprisingly high degree of error correction

It is well known that many famous pooling designs are constructed from mathematical structures by the “containment matrix” method. In this paper, we propose another method and obtain a family of pooling designs with surprisingly high degree of error correction based on a finite set. Given the numbers of items and pools, the error-tolerant property of our designs is much better than that of Macula?s designs when the size of the set is large enough.     

Capacitated location model with online demand pooling in a multi-channel supply chain

This paper presents a location model that assigns online demands to the capacitated regional warehouses currently serving in-store demands in a multi-channel supply chain. The model explicitly considers the trade-off between the risk pooling effect and the transportation cost in a two-echelon inventory/logistics system. Keeping the delivery network of the in-store demands unchanged, the model aims to minimize the transportation cost, inventory cost, and fixed handling cost in the system when assigning the online demands. We formulate the assignment problem as a non-linear integer programming model. Lagrangian relaxation based procedures are proposed to solve the model, both the general case and an important special case. Numerical experiments show the efficiency of our algorithms. Furthermore, we find that because of the pooling effect the variance of in-store demands currently served by a warehouse is an important parameter of the warehouse when it is considered as a candidate for supplying online demands. Highly uncertain in-store demands, as well as low transportation cost per unit, can make a warehouse appealing. We illustrate with numerical examples the trade-off between the pooling effect and the transportation cost in the assignment problem. We also evaluate the cost savings between the policy derived from the model, which integrates the transportation cost with the pooling effect, and the commonly used policy, which is based only on the transportation cost. Results show that the derived policy can reduce 1.5–7.5% cost in average and in many instances the percentage of cost savings is more than 10%.