We consider the problem of designing delivery routes for vehicles where the vendor has the choice of how much of the demand from a customer to fulfill. The customer demand is known a priori only as a probability distr...
详细信息
We consider the problem of designing delivery routes for vehicles where the vendor has the choice of how much of the demand from a customer to fulfill. The customer demand is known a priori only as a probability distribution. Exact customer demand is known only after visiting the customer. Different customers are able to negotiate different prices for each unit of product with the vendor. Given a route, the objective is to decide at each customer location, how much demand to satisfy so as to maximize expected profit taking into account a linear penalty cost for unfulfilled demand and the vehicle routing costs. In this article, we develop several new structural results for this problem. We illustrate how these structural results can be embedded in different heuristic frameworks commonly used for deterministic vehicle routing problems. This helps develop efficient routes for a single vehicle as well as a multiple vehicle scenario for this stochastic variant. For small-sized problems that allow for exhaustive enumeration, we demonstrate the effectiveness of the illustrated heuristic. For larger problem instances, based on structural results, we develop methods that allow the heuristic to run more efficiently than otherwise. Results are reported on instances based on benchmark instances drawn from literature for upward of 100 customers and vehicle capacity up to 600 units. Computational times needed to heuristically solve such problems are within 1 100 s.
We consider the problem of minimizing the number of bends in a planar orthogonal graph drawing. While the problem can be solved via network flow for a given planar embedding of a graph, it is NP-hard if we consider al...
详细信息
We consider the problem of minimizing the number of bends in a planar orthogonal graph drawing. While the problem can be solved via network flow for a given planar embedding of a graph, it is NP-hard if we consider all planar embeddings. Our approach for biconnected graphs combines a new integer linear programming (ILP) formulation for the set of all embeddings of a planar graph with the network flow formulation of the bend minimization problem fixed embeddings. We report on extensive computational experiments with two benchmark sets containing a total of more than 12,000 graphs where we compared the performance of our ILP-based algorithm with a heuristic and a previously published branch & bound algorithm for solving the same problem. Our new algorithm is significantly faster than the previously published approach for the larger graphs of the benchmark graphs derived from industrial applications and almost twice as fast for the benchmark graphs from the artificially generated set of hard instances.
In a recent paper, the authors introduced a method to estimate optical parameters of thin films using transmission data. The associated model assumes that the film is deposited on a completely transparent substrate. I...
详细信息
In a recent paper, the authors introduced a method to estimate optical parameters of thin films using transmission data. The associated model assumes that the film is deposited on a completely transparent substrate. It has been observed, however, that small absorption of substrates affect in a nonnegligible way the transmitted energy. The question arises of the reliability of the estimation method to retrieve optical parameters in the presence of substrates of different thicknesses and absorption degrees. In this paper, transmission spectra of thin films deposited on non-transparent substrates are generated and, as a first approximation, the method based on transparent substrates is used to estimate the optical parameters. As expected, the method is good when the absorption of the substrate is very small, but fails when one deals with less transparent substrates. To overcome this drawback, an iterative procedure is introduced, that allows one to approximate the transmittance with transparent substrate, given the transmittance with absorbent substrate. The updated method turns out to be almost as efficient in the case of absorbent substrates as it was in the case of transparent ones. (C) 2002 Elsevier Science B.V. All rights reserved.
In this paper, the problem of restoration of cloud contaminated optical images is studied in the case when we have no information about brightness of such images in the damage region. We propose a new variational appr...
详细信息
In this paper, the problem of restoration of cloud contaminated optical images is studied in the case when we have no information about brightness of such images in the damage region. We propose a new variational approach for exact restoration of optical multi-band images utilising Synthetic Aperture Radar (EOS - Spatial Data Analytics, GIS Software, Satellite Imagery - is a cloud-based platform to derive remote sensing data and analyse satellite imagery for business and science purposes) images of the same regions. We prove existence of solutions, propose an alternating minimisation method for computing them, prove convergence of this method to weak solutions of the original problem and derive optimality conditions.
This paper addresses a problem that is typical of multi-period capacity expansion equilibrium models: plants or sectors have different risk exposures that may warrant different costs of capital. The paper examines mod...
详细信息
This paper addresses a problem that is typical of multi-period capacity expansion equilibrium models: plants or sectors have different risk exposures that may warrant different costs of capital. The paper examines modifications of a capacity expansion model interpreted in equilibrium terms to account for asset-specific costs of capital.
This is a summary of the author's PhD thesis, supervised by Prof. Domenico Conforti and defended on 26-02-2010 at the Universita della Calabria, Cosenza. The thesis is written in Italian and a copy is available fr...
详细信息
This is a summary of the author's PhD thesis, supervised by Prof. Domenico Conforti and defended on 26-02-2010 at the Universita della Calabria, Cosenza. The thesis is written in Italian and a copy is available from the author upon request. This work deals with the development of a high-level classification framework which combines parameters optimization of a single classifier with classifiers ensemble optimization, through meta-heuristics. Support Vector Machines (SVM) is used for learning while the meta-heuristics adopted and compared are Genetic-Algorithms (GA), Tabu-Search (TS) and Ant Colony Optimization (ACO). Single SVM optimization usually concerns two approaches: searching for optimal set up of a SVM with fixed kernel (Model Selection) or with linear combination of basic kernels (Multiple Kernel Learning), both issues were considered. Meta-heuristics were used in order to avoid time consuming grid-approach for testing several classifiers configurations and some ad-hoc variations to GA were proposed. Finally, different frameworks were developed and then tested on 8 datasets providing reliable solutions.
The nonlinear problem of beef production is modeled by Signomial Geometric programming. The technical data have been taken from the article written by Epplin and Heady (1984). Inverse transformations and methods of co...
详细信息
In this paper, we deal with the just -in -time job shop scheduling problem with sequence -dependent setup times and release dates. Given a set of jobs characterized by release and due dates, the goal is to execute the...
详细信息
In this paper, we deal with the just -in -time job shop scheduling problem with sequence -dependent setup times and release dates. Given a set of jobs characterized by release and due dates, the goal is to execute them by minimizing a weighted sum of their earliness, tardiness, and flow time (i.e., the difference between completion and start time of each job). We develop new destroy and repair operators by exploiting the structure of the problem, and we use them within a reduced variable neighborhood search matheuristic. Computational experiments carried out on several sets of instances show that the proposed algorithm outperforms existing solution methods.
暂无评论