DSpace Angular :: Browsing CESE - Other Publications by Issue Date

Item

An Advanced Agent-Based Order Planning System for Dynamic Networked Enterprises

( 2004) César Toscano ; Jorge Pinho de Sousa ; Américo Azevedo ; António Lucas Soares

Item

Cooperative planning in dynamic supply chains

( 2005) César Toscano ; Jorge Pinho de Sousa ; Américo Azevedo

Item

Solving Irregular Strip Packing Problems by Hybridising Simulated Annealing and Linear Programming

( 2006) José Fernando Oliveira ; António Miguel Gomes

Item

Polyhedral study of simple plant location problem with order

( 2013) Vasilyev,I ; Xenia Klimentova ; Boccia,M

This paper is addressed to the generalization of simple plant location problem where customer's preferences are taken into account. Some basic polyhedral studies and a new family of facet-defining inequalities are given. The effectiveness of the proposed approach is illustrated by the computational experience.

Item

Dimensions to consider in the development of Hospital Centres performance evaluation framework

( 2013) Goncalves,S ; Ana Cristina Simões ; Américo Azevedo ; 2160 ; 5521

Item

New General Discrete-Time Scheduling Model for Multipurpose Batch Plants

( 2013) Samuel Moniz ; Barbosa Povoa,AP ; Jorge Pinho de Sousa

This work deals with the optimal short-term scheduling of general multipurpose batch plants, considering multiple operational characteristics such as sequence-dependent changeovers, temporary storage in the processing units, lots blending, and material flows traceability. A novel Mixed Integer Linear Programming (MILP) discrete-time formulation based on the State-Task Network (STN) is proposed, with new types of constraints for modeling changeovers and storage. We also propose some model extensions for addressing changeovers start; nonpreemptive lots; lots start and sizes; alternative task-unit and task-unit-layout assignments. Computational tests have shown that the proposed model is more effective than a similar model based on the Resource-Task Network (RTN).

Item

Special issue on "Cutting and Packing"

( 2013) António Miguel Gomes ; José Fernando Gonçalves ; Alvarez Valdes,R ; de Carvalho,V

Item

The Dotted-Board Model: A new MIP model for nesting irregular shapes

( 2013) Toledo,FMB ; Maria Antónia Carravilla ; Cristina Ribeiro ; José Fernando Oliveira ; António Miguel Gomes

The nesting problem, also known as irregular packing problem, belongs to the generic class of cutting and packing (C&P) problems. It differs from other 2-D C&P problems in the irregular shape of the pieces. This paper proposes a new mixed-integer model in which binary decision variables are associated with each discrete point of the board (a dot) and with each piece type. It is much more flexible than previously proposed formulations and solves to optimality larger instances of the nesting problem, at the cost of having its precision dependent on board discretization. To date no results have been published concerning optimal solutions for nesting problems with more than 7 pieces. We ran computational experiments on 45 problem instances with the new model, solving to optimality 34 instances with a total number of pieces ranging from 16 to 56, depending on the number of piece types, grid resolution and the size of the board. A strong advantage of the model is its insensitivity to piece and board geometry, making it easy to extend to more complex problems such as non-convex boards, possibly with defects. Additionally, the number of binary variables does not depend on the total number of pieces but on the number of piece types, making the model particularly suitable for problems with few piece types. The discrete nature of the model requires a trade-off between grid resolution and problem size, as the number of binary variables grows with the square of the selected grid resolution and with board size.

Item

Dimensions as Virtual Items: Improving the predictive ability of top-N recommender systems

( 2013) Domingues,MA ; Alípio Jorge ; Carlos Manuel Soares

Traditionally, recommender systems for the web deal with applications that have two dimensions, users and items. Based on access data that relate these dimensions, a recommendation model can be built and used to identify a set of N items that will be of interest to a certain user. In this paper we propose a multidimensional approach, called DaVI (Dimensions as Virtual Items), that consists in inserting contextual and background information as new user-item pairs. The main advantage of this approach is that it can be applied in combination with several existing two-dimensional recommendation algorithms. To evaluate its effectiveness, we used the DaVI approach with two different top-N recommender algorithms, Item-based Collaborative Filtering and Association Rules based, and ran an extensive set of experiments in three different real world data sets. In addition, we have also compared our approach to the previously introduced combined reduction and weight post-filtering approaches. The empirical results strongly indicate that our approach enables the application of existing two-dimensional recommendation algorithms in multidimensional data, exploiting the useful information of these data to improve the predictive ability of top-N recommender systems.

Item

Proceedings of the International Workshop on Meta-learning and Algorithm Selection co-located with 21st European Conference on Artificial Intelligence, MetaSel@ECAI 2014, Prague, Czech Republic, August 19, 2014