Papers by Carlos Cardonha
Cornell University - arXiv, Oct 21, 2018
Bookmarks Related papers MentionsView impact
Proceedings of the International Conference on Automated Planning and Scheduling
Even in today’s world of increasingly faster storage technologies, magnetic tapes continue to pla... more Even in today’s world of increasingly faster storage technologies, magnetic tapes continue to play an essential role in the market. Yet, they are often overlooked in the literature, despite the many changes made to the underlying tape architecture since they were conceived. In this article, we introduce the LINEAR TAPE SCHEDULING PROBLEM (LTSP), which aims to identify scheduling strategies for read and write operations in single-tracked magnetic tapes that minimize the overall response times for read requests. Structurally, LTSP has many similarities with versions of the Travelling Repairman Problem and of the Dial-a-Ride Problem restricted to the real line. We investigate several properties of LTSP and show how they can be explored in the design of algorithms for the online version of the problem. Computational experiments show that the resulting strategies deliver very satisfactory scheduling plans, which in most cases are clearly superior (potentially differing by one order of ma...
Bookmarks Related papers MentionsView impact
Cornell University - arXiv, Dec 13, 2021
Bookmarks Related papers MentionsView impact
Operations Research, Mar 21, 2019
Bookmarks Related papers MentionsView impact
ArXiv, 2021
We study optimization problems where the objective function is modeled through feedforward neural... more We study optimization problems where the objective function is modeled through feedforward neural networks with rectified linear unit (ReLU) activation. Recent literature has explored the use of a single neural network to model either uncertain or complex elements within an objective function. However, it is well known that ensembles of neural networks produce more stable predictions and have better generalizability than models with single neural networks, which suggests the application of ensembles of neural networks in a decision-making pipeline. We study how to incorporate a neural network ensemble as the objective function of an optimization model and explore computational approaches for the ensuing problem. We present a mixed-integer linear program based on existing popular big-M formulations for optimizing over a single neural network. We develop two acceleration techniques for our model, the first one is a preprocessing procedure to tighten bounds for critical neurons in the ...
Bookmarks Related papers MentionsView impact
INFORMS Journal on Computing, 2021
This paper provides a novel framework for solving multiobjective discrete optimization problems w... more This paper provides a novel framework for solving multiobjective discrete optimization problems with an arbitrary number of objectives. Our framework represents these problems as network models, in that enumerating the Pareto frontier amounts to solving a multicriteria shortest-path problem in an auxiliary network. We design techniques for exploiting network models in order to accelerate the identification of the Pareto frontier, most notably a number of operations to simplify the network by removing nodes and arcs while preserving the set of nondominated solutions. We show that the proposed framework yields orders-of-magnitude performance improvements over existing state-of-the-art algorithms on five problem classes containing both linear and nonlinear objective functions. Summary of Contribution: Multiobjective optimization has a long history of research with applications in several domains. Our paper provides an alternative modeling and solution approach for multiobjective discre...
Bookmarks Related papers MentionsView impact
Integration of Constraint Programming, Artificial Intelligence, and Operations Research, 2021
Bookmarks Related papers MentionsView impact
ArXiv, 2019
Bin Packing with Minimum Color Fragmentation (BPMCF) is an extension of the Bin Packing Problem i... more Bin Packing with Minimum Color Fragmentation (BPMCF) is an extension of the Bin Packing Problem in which each item has a size and a color and the goal is to minimize the sum of the number of bins containing items of each color. In this work, we introduce the BPMCF and present a decomposition strategy to solve the problem, where the assignment of items to bins is formulated as a binary decision diagram and an optimal integrated solutions is identified through a mixed-integer linear programming model. Our computational experiments show that the proposed approach greatly outperforms a direct formulation of BPMCF and that its performance is suitable for large instances of the problem.
Bookmarks Related papers MentionsView impact
The Vehicle Positioning Problem (VPP) is a classical combinatorial optimization problem that has ... more The Vehicle Positioning Problem (VPP) is a classical combinatorial optimization problem that has a natural formulation as a Mixed Integer Quadratically Constrained Program. This MIQCP is closely related to the Quadratic Assignment Problem and, as far as we know, has not received any attention yet. We show in this article that such a formulation has interesting theoretical properties. Its QP relaxation produces, in particular, the first known nontrivial lower bound on the number of shuntings. In our experiments, it also outperformed alternative integer linear models computationally. The strengthening technique that raises the lower bound might also be useful for other combinatorial optimization problems.
Bookmarks Related papers MentionsView impact
ArXiv, 2019
A graph is chordal if every cycle of length at least four contains a chord—that is, an edge conne... more A graph is chordal if every cycle of length at least four contains a chord—that is, an edge connecting two nonconsecutive vertices of the cycle. Several classical applications in sparse linear syst...
Bookmarks Related papers MentionsView impact
This dissertation is dedicated to the Vehicle Positioning Problem (VPP), a classical combinatoria... more This dissertation is dedicated to the Vehicle Positioning Problem (VPP), a classical combinatorial optimization problem in public transport in which vehicles should be assigned to parking positions in a depot in such a way that shunting moves are minimized. We investigate several models and solution methods to solve the VPP and the VPPp, a multi-periodic extension of the problem which was not previously studied. In the first part of the thesis, the basic version of the problem is introduced and several formulations, theoretical properties, and concepts are investigated. In particular, we propose a mixed integer quadratic constrained formulation of the VPP whose QP relaxation produces the first known nontrivial lower bound on the number of shunting moves. The second part of our work describes two advanced solution methods. In the first approach, a set partitioning formulation is solved by a branch-and-price framework. We present efficient algorithms for the pricing problem and in ord...
Bookmarks Related papers MentionsView impact
Decision Support Systems, 2021
In this article, we describe the decision support system that was developed for the assignment of... more In this article, we describe the decision support system that was developed for the assignment of courses to teaching modalities and rooms for the Fall semester of 2020 at the University of Connecticut (UConn). With the adoption of safety/mitigation standards imposed by the COVID-19 pandemic, the seating capacities of rooms were reduced by more than 70%, thus making virtually every existing room assignment for Fall 2020 infeasible. The demand for in-person instruction required the reassignment of a large number of courses to rooms, where not all requests for physical space could be accommodated. In order to maximize opportunities for in-person instruction, UConn introduced a teaching modality in which class meetings are attended on campus by only 50% of the enrolled students. As decision makers were given partial flexibility to assign teaching modalities to classes, the complexity of the assignment problem increased considerably, especially because the real-world instances involved hundreds of rooms and thousands of classes and required a quick solution turnaround in practice. In this article, we introduce this flexible assignment problem and describe the two mixed-integer programming formulations that were used to solve the real-world instances of the problem;in particular, one of the formulations leverages structural properties presented in this work in order to represent the problem in a more compact way. We explain how we tailored our algorithms to solve the real-world problem, describe the dynamics of the interactive decision support system created in this initiative, and present insights derived from our study.
Bookmarks Related papers MentionsView impact
Proceedings of the XVI Brazilian Symposium on Human Factors in Computing Systems, 2017
Bookmarks Related papers MentionsView impact
Proceedings of the 13th International Web for All Conference, 2016
Digital education has potential to provide different possibilities for personalization and conseq... more Digital education has potential to provide different possibilities for personalization and consequently reach a larger and more diverse number of people. Personalization is a key component of solutions addressing important and long-standing pedagogical challenges in education, such as dealing with heterogeneity of learning styles. In particular scenarios where accessibility support is required, personalization depends on the creation of different representations for individual pieces of content. In this light, the main goal of this article is to describe how we addressed the challenges involved in the construction of a platform that satisfies this requirement. We thus present a system that supports the creation, adaptation, and delivery of personalized courses for people with multiple types of disabilities. More specifically, we introduce the technology, describe its main capabilities, and discuss the results of early evaluations by two instructors of an institution that provides vocational training for people with intellectual disabilities. Our initial results show that the tool was favorably assessed by the instructors and can potentially be adopted in this community.
Bookmarks Related papers MentionsView impact
IBM Journal of Research and Development, 2015
Bookmarks Related papers MentionsView impact
Communications in Computer and Information Science, 2015
Bookmarks Related papers MentionsView impact
Bookmarks Related papers MentionsView impact
Open Communication Interfaces have been applied as forms to foment social networking and citizen ... more Open Communication Interfaces have been applied as forms to foment social networking and citizen engagement in com- munity settings. In this work, we propose a new model of shared boards, tailored for low-income communities. We foresee this technology as a new type of social media, since it establishes an open digital interface for intra-community message exchange between members and has the potential to foster a new mode of citizen participation. In this article, we introduce the technology and report the results of a real- world experiment that was internally conducted on a large IT company.
Bookmarks Related papers MentionsView impact
Proceedings of the 11th Web for All Conference, 2014
ABSTRACT The access to information displayed in public spaces is a challenge faced by visually im... more ABSTRACT The access to information displayed in public spaces is a challenge faced by visually impaired people for which image processing techniques have the potential to deliver satisfactory solutions. However, object recognition algorithms must initially locate possible candidates in the images, which is a hard task in complex scenes. In this article, we introduce an image processing technique that relies on the incorporation of markers to panels and boards with fixed layouts displaying dynamic content. The markers allow: a) locating the objects to be recognized; b) correcting perspective in the input images; c) limiting the training set size for supervised learning; and d) guiding the visually impaired by indicating how they should position their devices for adequate pictures. The proposed technique can be used for automatic recognition of texts and images and is suitable for deployment on mobile devices, providing more independence to the citizens. Results of preliminary tests on vending machines show that this method is robust enough to be used in practice.
Bookmarks Related papers MentionsView impact
Proceedings of the 11th Web for All Conference, 2014
ABSTRACT In this work we present an image processing-based assistant for helping visually impaire... more ABSTRACT In this work we present an image processing-based assistant for helping visually impaired citizens with the task of recognizing dynamic content within fixed layouts of displays in public spaces. Our solution relies on the placement of markers, in order to facilitate the location and recognition of target objects and, at the same time, provide hints to users about how to better position their mobile device's cameras to capture the whole information contained in the display.
Bookmarks Related papers MentionsView impact
Uploads
Papers by Carlos Cardonha